From richard.hills at immi.gov.au Fri Jul 4 06:20:23 2014 From: richard.hills at immi.gov.au (Richard James HILLS) Date: Fri, 4 Jul 2014 04:20:23 +0000 Subject: [BLML] IMP tables and IMPing multi-pair teams [SEC=UNOFFICIAL] Message-ID: <240635A98F59F24AAD8510EB05121DC331AE8901@SDCWPIPEX02.IMMI.LOCAL> UNOFFICIAL Bill Hood: [big snip] >What disturbs me is that there is no consistency or sense to >the default bidding strategy and, in the absence of a known >or practically knowable bidding strategy, most players >somehow trust that the scoring is organised in a way that >rewards their best efforts even as they separately use their >subjective and largely independent judgements. Richard Hills: Yes and No. At standard head-to-head contests between two teams it is well known that the break-even point for bidding a vulnerable game is a 37.5% chance of success (if one makes the simplifying assumption that the game will never be doubled). At standard multi-pair IMPed events (Butler pairs) it is less well known that the break-even point for bidding a vulnerable game is a 33.3% chance of success (if one makes the simplifying assumption that the game will never be doubled). Bill Hood: >I was finally prompted to raise these issues when the EBU >proposed the use of IMP tables for multi-pair teams by >multiplying end-range values by somewhat arbitrary >mathematical factors as that effectively randomises the >bidding strategy in a way that is practically unknowable to >players. [small snip] Richard Hills: Yes and No. Old-fashioned Butler pairs scored against a datum is deprecated by blml aficionados. They prefer cross-IMPs, a.k.a. "multiplying end-range values by somewhat arbitrary mathematical factors". But in both cases of Butler pair scoring the winning bidding strategy is very similar to the winning bidding strategy at head- to-head teams. That is, bid to thin games, avoid thin slams, do not over-compete in a partscore battle (a plus score is a good score at IMPs for a partscore battle in the long run). John Maynard Keynes (1883 - 1946): "In the long run we are all dead." UNOFFICIAL -------------------------------------------------------------------- Important Notice: If you have received this email by mistake, please advise the sender and delete the message and attachments immediately. This email, including attachments, may contain confidential, sensitive, legally privileged and/or copyright information. Any review, retransmission, dissemination or other use of this information by persons or entities other than the intended recipient is prohibited. DIBP respects your privacy and has obligations under the Privacy Act 1988. The official departmental privacy policy can be viewed on the department's website at www.immi.gov.au. See: http://www.immi.gov.au/functional/privacy.htm --------------------------------------------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.rtflb.org/pipermail/blml/attachments/20140704/a95557c6/attachment.html From hermandw at skynet.be Fri Jul 4 08:18:41 2014 From: hermandw at skynet.be (Herman De Wael) Date: Fri, 04 Jul 2014 08:18:41 +0200 Subject: [BLML] IMTables and IMPing multi-pair teams In-Reply-To: <7E660BC78E1F4B25BC88F314C5FADE17@Pathfinder> References: <4709DE22F2E74C58957F9789CCD3EFCD@magnifique> <7E660BC78E1F4B25BC88F314C5FADE17@Pathfinder> Message-ID: <53B64741.8020702@skynet.be> While this discussion is interesting, the real problem is not the current IMP-table, but the process of IMPing itself. You really need to imagine what you would need to do to tweek the IMP-table: all documents around the world need to be re-made, and for a while different versions will co-exist, leading to problems. All programs need to be re-written, and a switch date needs to be determined. And contrary to the new VP-scale, the two versions will not be readily distinguishable, because you are apparentlu staying with a discrete 24-IMP scale. I doubt that the changes will be worth-while to convince the powers that be that a change should be made. Rather, I think we should do away with IMPing altogether. Currently, there are a few irregularities. An overtrick in the majors is worth one IMP, and two overtricks are worth two IMPs. An overtrick in the minors is also worth one IMP, and two overtricks ... also one. And an overtrick is sometimes worth nothing if the other table happens to be in a radically different contract (or going down in the same one). If you bid game and they don't, you earn 10 or 12 IMPs. If you bid slam and they don't, you also earn 10 or 12. But if you bid slam and they don't even bid game, the gain is only 14 or 17. And I could go on. The answer is a linear IMPtable, or no IMPtable at all, so using straight point scoring. Of course this means that the scoring table needs to change. I have developed a start of such an adapted scoring table, and for the low numbers (part scores, games), it works. Any and all input may be helpful. Herman. Bill Hood schreef: > Hi Roger, > > I am indebted to you for the links to the articles on the early history > of IMPs as I had spent considerable effort in trying to find such material. > From grabiner at alumni.princeton.edu Fri Jul 4 16:34:55 2014 From: grabiner at alumni.princeton.edu (David Grabiner) Date: Fri, 04 Jul 2014 10:34:55 -0400 Subject: [BLML] IMTables and IMPing multi-pair teams In-Reply-To: <53B64741.8020702@skynet.be> References: <4709DE22F2E74C58957F9789CCD3EFCD@magnifique> <7E660BC78E1F4B25BC88F314C5FADE17@Pathfinder> <53B64741.8020702@skynet.be> Message-ID: "Herman De Wael" writes: > Rather, I think we should do away with IMPing altogether. Currently, > there are a few irregularities. An overtrick in the majors is worth one > IMP, and two overtricks are worth two IMPs. An overtrick in the minors > is also worth one IMP, and two overtricks ... also one. And an overtrick > is sometimes worth nothing if the other table happens to be in a > radically different contract (or going down in the same one). This could be handled by smoothing the table (I believe this is known as Bastille scoring); the difference between +110 and +130 on a part-score deal could be 0.6 IMPs. With nonlinearity, the value of an overtrick will vary, but it will always be positive; the difference between +110 and +130 when the other table is making a vulnerable game can be a consistent 0.2, rather than being 1 IMP if they are +620 but zero if they are +600 or +630. > If you bid game and they don't, you earn 10 or 12 IMPs. 8 or 10; the 10 or 12 applies if both tables bid game and they go down. > If you bid slam > and they don't, you also earn 10 or 12. 11 or 13. > But if you bid slam and they > don't even bid game, the gain is only 14 or 17. > And I could go on. > The answer is a linear IMPtable, or no IMPtable at all, so using > straight point scoring. Of course this means that the scoring table > needs to change. This issue came up when the IMP table was first created, and the reason for nonlinearity is still valid. With total points (or a linear conversion to IMPs), a single board is too likely to determine a match. If you go down in a vulnerable grand slam which was made at the other table, 20 IMPs is a big loss, but you can make it up with +620 versus +170, +110 versus -100, and +300 versus +100 on three subsequent deals. If you have lost 2310 total points, you might as well concede the match. From hermandw at skynet.be Fri Jul 4 16:53:06 2014 From: hermandw at skynet.be (Herman De Wael) Date: Fri, 04 Jul 2014 16:53:06 +0200 Subject: [BLML] IMTables and IMPing multi-pair teams In-Reply-To: References: <4709DE22F2E74C58957F9789CCD3EFCD@magnifique> <7E660BC78E1F4B25BC88F314C5FADE17@Pathfinder> <53B64741.8020702@skynet.be> Message-ID: <53B6BFD2.6070005@skynet.be> David Grabiner schreef: > "Herman De Wael" writes: > > > This issue came up when the IMP table was first created, and the reason for > nonlinearity is still valid. With total points (or a linear conversion to > IMPs), a single board is too likely to determine a match. If you go down in a > vulnerable grand slam which was made at the other table, 20 IMPs is a big loss, > but you can make it up with +620 versus +170, +110 versus -100, and +300 versus > +100 on three subsequent deals. If you have lost 2310 total points, you might > as well concede the match. > But this can also be remedied by lowering the 1500 bonus for the grand slam. Herman. From moysian at talktalk.net Sun Jul 6 21:34:28 2014 From: moysian at talktalk.net (Bill Hood) Date: Sun, 6 Jul 2014 20:34:28 +0100 Subject: [BLML] IMTables and IMPing multi-pair teams In-Reply-To: <53B64741.8020702@skynet.be> References: <4709DE22F2E74C58957F9789CCD3EFCD@magnifique><7E660BC78E1F4B25BC88F314C5FADE17@Pathfinder> <53B64741.8020702@skynet.be> Message-ID: <56CC7E49EB54411B87BDE163B4666A10@Pathfinder> For the attention of Herman de Wae et al: Many thanks for your observations Herman. I do not under-estimate the enormity of the changes that would be necessary to overhaul the existing IMP table and the associated IMPing process. However, nor do I under-estimate the enormity of the dis-functionality of the existing methodology with its inconsistencies and inadequacies. Are we really saying that we cannot have a scoring system that is fit for purpose for the game or that, even if we could devise one, it would be too troublesome to work towards its introduction (over time)? Frankly, I think the game deserves better and that it is diminished (perhaps forever) by accepting this fundamental undermining of its integrity. For instance, is there any real point in the finely judged arguments and mathematics for converting match IMPs to VPs if the IMP total is so haphazardly and carelessly calculated? We should not let the anticipation of future administrative issues prohibit a debate/review of a better scoring system. My starting point is to insist that it is a fundamental requirement of any scoring system that not only can it can be used to derive an understanding of how best to play the game to maximise our chances of coming out ahead but also that the derived strategy is sensible and consistent. >From there, it seems sensible to set design objectives for what we want the scoring system to do. By contrast, the existing IMP table and the IMP process is an example of having introduced a process of scoring and then looked at what the consequences of that process are in terms of the best bidding strategy and trying to somehow rationalise the many irregularities and inconsistencies that it throws up. I do not insist that my approach is "best" or the only path to a more acceptable solution but it offers the potential for a radical improvement and puts the issues into some sort of context. Moreover, I have adopted this approach to start moving gradually from the existing model and, for example, I am not wedded to the notion of a discrete 24-point IMP scale. Indeed, I think that this may well be intrinsically unmanageable and inconsistent with the objectives of the desired compression. Other approaches may indeed be better or more practical and I would be interested in your ideas for a revised scoring table. For my part, I have been reluctant to interfere with fundamental scoring units (trick values etc) as they are too engrained and have restricted adjustments to bonus values. You say that your revised scoring table works at the lower end of the scale (part-scores and games) - possibly in the same way that a reduced IMP scale "seems" to work? However, I don't think you can dismiss the idea of IMPing in some form and linear IMPing simply gets you back to the problem you were trying to solve in the first place. Best regards Bill -----Original Message----- From: blml-bounces at rtflb.org [mailto:blml-bounces at rtflb.org] On Behalf Of Herman De Wael Sent: 04 July 2014 07:19 To: Bridge Laws Mailing List Subject: Re: [BLML] IMTables and IMPing multi-pair teams While this discussion is interesting, the real problem is not the current IMP-table, but the process of IMPing itself. You really need to imagine what you would need to do to tweek the IMP-table: all documents around the world need to be re-made, and for a while different versions will co-exist, leading to problems. All programs need to be re-written, and a switch date needs to be determined. And contrary to the new VP-scale, the two versions will not be readily distinguishable, because you are apparentlu staying with a discrete 24-IMP scale. I doubt that the changes will be worth-while to convince the powers that be that a change should be made. Rather, I think we should do away with IMPing altogether. Currently, there are a few irregularities. An overtrick in the majors is worth one IMP, and two overtricks are worth two IMPs. An overtrick in the minors is also worth one IMP, and two overtricks ... also one. And an overtrick is sometimes worth nothing if the other table happens to be in a radically different contract (or going down in the same one). If you bid game and they don't, you earn 10 or 12 IMPs. If you bid slam and they don't, you also earn 10 or 12. But if you bid slam and they don't even bid game, the gain is only 14 or 17. And I could go on. The answer is a linear IMPtable, or no IMPtable at all, so using straight point scoring. Of course this means that the scoring table needs to change. I have developed a start of such an adapted scoring table, and for the low numbers (part scores, games), it works. Any and all input may be helpful. Herman. Bill Hood schreef: > Hi Roger, > > I am indebted to you for the links to the articles on the early history > of IMPs as I had spent considerable effort in trying to find such material. > _______________________________________________ Blml mailing list Blml at rtflb.org http://lists.rtflb.org/mailman/listinfo/blml From jimfox00 at cox.net Mon Jul 7 00:21:17 2014 From: jimfox00 at cox.net (Jim Fox) Date: Sun, 6 Jul 2014 18:21:17 -0400 Subject: [BLML] IMTables and IMPing multi-pair teams In-Reply-To: References: <4709DE22F2E74C58957F9789CCD3EFCD@magnifique><7E660BC78E1F4B25BC88F314C5FADE17@Pathfinder> <53B64741.8020702@skynet.be> Message-ID: Your high-toned rhetoric is overdone because there is not a significant interest in changing anything in the area of imp scoring. The idiosyncrasies of the imp scale actually make the game more interesting. The majority of players want there to be the amount of scoring randomness we have now (even if they don't consciously realize it). Mmbridge -----Original Message----- From: blml-bounces at rtflb.org [mailto:blml-bounces at rtflb.org] On Behalf Of Bill Hood Sent: 07/06/2014 3:34 PM To: 'Bridge Laws Mailing List' Subject: Re: [BLML] IMTables and IMPing multi-pair teams For the attention of Herman de Wae et al: Many thanks for your observations Herman. I do not under-estimate the enormity of the changes that would be necessary to overhaul the existing IMP table and the associated IMPing process. However, nor do I under-estimate the enormity of the dis-functionality of the existing methodology with its inconsistencies and inadequacies. Are we really saying that we cannot have a scoring system that is fit for purpose for the game or that, even if we could devise one, it would be too troublesome to work towards its introduction (over time)? Frankly, I think the game deserves better and that it is diminished (perhaps forever) by accepting this fundamental undermining of its integrity. For instance, is there any real point in the finely judged arguments and mathematics for converting match IMPs to VPs if the IMP total is so haphazardly and carelessly calculated? We should not let the anticipation of future administrative issues prohibit a debate/review of a better scoring system. My starting point is to insist that it is a fundamental requirement of any scoring system that not only can it can be used to derive an understanding of how best to play the game to maximise our chances of coming out ahead but also that the derived strategy is sensible and consistent. >From there, it seems sensible to set design objectives for what we want the scoring system to do. By contrast, the existing IMP table and the IMP process is an example of having introduced a process of scoring and then looked at what the consequences of that process are in terms of the best bidding strategy and trying to somehow rationalise the many irregularities and inconsistencies that it throws up. I do not insist that my approach is "best" or the only path to a more acceptable solution but it offers the potential for a radical improvement and puts the issues into some sort of context. Moreover, I have adopted this approach to start moving gradually from the existing model and, for example, I am not wedded to the notion of a discrete 24-point IMP scale. Indeed, I think that this may well be intrinsically unmanageable and inconsistent with the objectives of the desired compression. Other approaches may indeed be better or more practical and I would be interested in your ideas for a revised scoring table. For my part, I have been reluctant to interfere with fundamental scoring units (trick values etc) as they are too engrained and have restricted adjustments to bonus values. You say that your revised scoring table works at the lower end of the scale (part-scores and games) - possibly in the same way that a reduced IMP scale "seems" to work? However, I don't think you can dismiss the idea of IMPing in some form and linear IMPing simply gets you back to the problem you were trying to solve in the first place. Best regards Bill -----Original Message----- From: blml-bounces at rtflb.org [mailto:blml-bounces at rtflb.org] On Behalf Of Herman De Wael Sent: 04 July 2014 07:19 To: Bridge Laws Mailing List Subject: Re: [BLML] IMTables and IMPing multi-pair teams While this discussion is interesting, the real problem is not the current IMP-table, but the process of IMPing itself. You really need to imagine what you would need to do to tweek the IMP-table: all documents around the world need to be re-made, and for a while different versions will co-exist, leading to problems. All programs need to be re-written, and a switch date needs to be determined. And contrary to the new VP-scale, the two versions will not be readily distinguishable, because you are apparentlu staying with a discrete 24-IMP scale. I doubt that the changes will be worth-while to convince the powers that be that a change should be made. Rather, I think we should do away with IMPing altogether. Currently, there are a few irregularities. An overtrick in the majors is worth one IMP, and two overtricks are worth two IMPs. An overtrick in the minors is also worth one IMP, and two overtricks ... also one. And an overtrick is sometimes worth nothing if the other table happens to be in a radically different contract (or going down in the same one). If you bid game and they don't, you earn 10 or 12 IMPs. If you bid slam and they don't, you also earn 10 or 12. But if you bid slam and they don't even bid game, the gain is only 14 or 17. And I could go on. The answer is a linear IMPtable, or no IMPtable at all, so using straight point scoring. Of course this means that the scoring table needs to change. I have developed a start of such an adapted scoring table, and for the low numbers (part scores, games), it works. Any and all input may be helpful. Herman. Bill Hood schreef: > Hi Roger, > > I am indebted to you for the links to the articles on the early history > of IMPs as I had spent considerable effort in trying to find such material. > _______________________________________________ Blml mailing list Blml at rtflb.org http://lists.rtflb.org/mailman/listinfo/blml _______________________________________________ Blml mailing list Blml at rtflb.org http://lists.rtflb.org/mailman/listinfo/blml From richard.hills at immi.gov.au Mon Jul 7 01:23:10 2014 From: richard.hills at immi.gov.au (Richard James HILLS) Date: Sun, 6 Jul 2014 23:23:10 +0000 Subject: [BLML] QWERTY (was IMP tables) [SEC=UNOFFICIAL] Message-ID: <240635A98F59F24AAD8510EB05121DC331AEA739@SDCWPIPEX02.IMMI.LOCAL> UNOFFICIAL I do not wish to start a flame war, but in my opinion writing "Your high- toned rhetoric is overdone" about the enthusiastic contribution of a blml newbie is unfair. Especially when I cannot recall the critic making a single useful suggestion himself over the past decade. As for the reason why the Law 78B imps scale has been unchanged during the past half-century, it is NOT "The majority of players want there to be the amount of [imps] scoring randomness we have now (even if they don't consciously realize it)". Rather the reason for the unchanging Law 78B imps scale is that it is an adequate solution to the flaws of total-point scoring (as discussed by David Grabiner), and that the gains from a slightly more adequate hypothetical 2015 Law 78B would not overcome the lock-in costs. See: http://en.wikipedia.org/wiki/Network_effect#Lock-in UNOFFICIAL -------------------------------------------------------------------- Important Notice: If you have received this email by mistake, please advise the sender and delete the message and attachments immediately. This email, including attachments, may contain confidential, sensitive, legally privileged and/or copyright information. Any review, retransmission, dissemination or other use of this information by persons or entities other than the intended recipient is prohibited. DIBP respects your privacy and has obligations under the Privacy Act 1988. The official departmental privacy policy can be viewed on the department's website at www.immi.gov.au. See: http://www.immi.gov.au/functional/privacy.htm --------------------------------------------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.rtflb.org/pipermail/blml/attachments/20140706/4c5e878f/attachment.html From jimfox00 at cox.net Mon Jul 7 01:48:31 2014 From: jimfox00 at cox.net (Jim Fox) Date: Sun, 6 Jul 2014 19:48:31 -0400 Subject: [BLML] QWERTY (was IMP tables) [SEC=UNOFFICIAL] In-Reply-To: References: Message-ID: Most of my contributions were made 40 years ago (not in BLML). Your comments are nonsense since they are sensible only in the context of "expert" players. I speak of ALL players who in fact do gain from the randomness of the imp scale, random seeding, and the like. Try to think out of your box. Mmbridge From: blml-bounces at rtflb.org [mailto:blml-bounces at rtflb.org] On Behalf Of Richard James HILLS Sent: 07/06/2014 7:23 PM To: Laws Bridge Cc: hildalirsch at gmail.com Subject: Re: [BLML] QWERTY (was IMP tables) [SEC=UNOFFICIAL] UNOFFICIAL I do not wish to start a flame war, but in my opinion writing "Your high- toned rhetoric is overdone" about the enthusiastic contribution of a blml newbie is unfair. Especially when I cannot recall the critic making a single useful suggestion himself over the past decade. As for the reason why the Law 78B imps scale has been unchanged during the past half-century, it is NOT "The majority of players want there to be the amount of [imps] scoring randomness we have now (even if they don't consciously realize it)". Rather the reason for the unchanging Law 78B imps scale is that it is an adequate solution to the flaws of total-point scoring (as discussed by David Grabiner), and that the gains from a slightly more adequate hypothetical 2015 Law 78B would not overcome the lock-in costs. See: http://en.wikipedia.org/wiki/Network_effect#Lock-in UNOFFICIAL -------------------------------------------------------------------- Important Notice: If you have received this email by mistake, please advise the sender and delete the message and attachments immediately. This email, including attachments, may contain confidential, sensitive, legally privileged and/or copyright information. Any review, retransmission, dissemination or other use of this information by persons or entities other than the intended recipient is prohibited. DIBP respects your privacy and has obligations under the Privacy Act 1988. The official departmental privacy policy can be viewed on the department's website at www.immi.gov.au. See: http://www.immi.gov.au/functional/privacy.htm --------------------------------------------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.rtflb.org/pipermail/blml/attachments/20140706/90f6433f/attachment-0001.html From richard.hills at immi.gov.au Mon Jul 7 03:26:25 2014 From: richard.hills at immi.gov.au (Richard James HILLS) Date: Mon, 7 Jul 2014 01:26:25 +0000 Subject: [BLML] QWERTY (was IMP tables) [SEC=UNOFFICIAL] Message-ID: <240635A98F59F24AAD8510EB05121DC33954D42B@SDCWPIPEX03.IMMI.LOCAL> UNOFFICIAL Bill Hood: [big snip] >For my part, I have been reluctant to interfere with fundamental >scoring units (trick values etc) as they are too engrained and >have restricted adjustments to bonus values. [big snip] Richard Hills: In 1987 the World Bridge Federation overcame the QWERTY lock-in effect by making two changes to fundamental scoring; one minor change and one major change. The minor change was to increase the "insult" for a redoubled contract from 50 to 100. This meant that the score for a vulnerable redoubled 5C / 5D contract making with an over- trick increased from 1350 (less than 1370) to 1400 (more than 1370). Hence at matchpoint pairs the Striped Tailed Ape Double ploy would always fail should a redouble occur. The major change was to increase the penalty for non- vulnerable doubled undertricks from 200 to 300 for penalties greater than 500. This precluded further use of the Meckwell stratagem under the pre-1987 scoring of taking a non- vulnerable save in 7S over a vulnerable 7H and making an imps profit if achieving as few as two tricks. Best wishes, Hilda R. Lirsch hildalirsch at gmail.com UNOFFICIAL -------------------------------------------------------------------- Important Notice: If you have received this email by mistake, please advise the sender and delete the message and attachments immediately. This email, including attachments, may contain confidential, sensitive, legally privileged and/or copyright information. Any review, retransmission, dissemination or other use of this information by persons or entities other than the intended recipient is prohibited. DIBP respects your privacy and has obligations under the Privacy Act 1988. The official departmental privacy policy can be viewed on the department's website at www.immi.gov.au. See: http://www.immi.gov.au/functional/privacy.htm --------------------------------------------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.rtflb.org/pipermail/blml/attachments/20140707/03798e57/attachment.html From svenpran at online.no Mon Jul 7 08:05:16 2014 From: svenpran at online.no (Sven Pran) Date: Mon, 7 Jul 2014 08:05:16 +0200 Subject: [BLML] IMTables and IMPing multi-pair teams In-Reply-To: <56CC7E49EB54411B87BDE163B4666A10@Pathfinder> References: <4709DE22F2E74C58957F9789CCD3EFCD@magnifique><7E660BC78E1F4B25BC88F314C5FADE17@Pathfinder> <53B64741.8020702@skynet.be> <56CC7E49EB54411B87BDE163B4666A10@Pathfinder> Message-ID: <001901cf99a9$6cdefeb0$469cfc10$@online.no> In my profession (engineering) we have a saying: If it ain't broke then don't fix it! I can't help feeling that you have completely failed to substantiate your allegation that the IMP table is flawed. It does indeed fulfill its main purpose which is to increase the impact from part score contracts so that the victory of an event will not solely be determined by some few extreme game results during a match. My experience from thirty (plus) years of directing is that players are comfortable with the total points to IMP conversion as it is. (The last main "flaw" in scoring was rectified in 1987 when they increased the bonus for setting doubled or redoubled non-vulnerable contracts more than three down so that players could no longer benefit from sacrificing against grand slams with eleven down when green v/ red.) > -----Opprinnelig melding----- > Fra: blml-bounces at rtflb.org [mailto:blml-bounces at rtflb.org] P? vegne av Bill > Hood > Sendt: 6. juli 2014 21:34 > Til: 'Bridge Laws Mailing List' > Emne: Re: [BLML] IMTables and IMPing multi-pair teams > > For the attention of Herman de Wae et al: > > Many thanks for your observations Herman. > > I do not under-estimate the enormity of the changes that would be necessary > to overhaul the existing IMP table and the associated IMPing process. > > However, nor do I under-estimate the enormity of the dis-functionality of the > existing methodology with its inconsistencies and inadequacies. > > Are we really saying that we cannot have a scoring system that is fit for > purpose for the game or that, even if we could devise one, it would be too > troublesome to work towards its introduction (over time)? > > Frankly, I think the game deserves better and that it is diminished (perhaps > forever) by accepting this fundamental undermining of its integrity. For > instance, is there any real point in the finely judged arguments and > mathematics for converting match IMPs to VPs if the IMP total is so > haphazardly and carelessly calculated? > > We should not let the anticipation of future administrative issues prohibit a > debate/review of a better scoring system. > > My starting point is to insist that it is a fundamental requirement of any scoring > system that not only can it can be used to derive an understanding of how best > to play the game to maximise our chances of coming out ahead but also that > the derived strategy is sensible and consistent. > > >From there, it seems sensible to set design objectives for what we want > >the > scoring system to do. By contrast, the existing IMP table and the IMP process > is an example of having introduced a process of scoring and then looked at what > the consequences of that process are in terms of the best bidding strategy and > trying to somehow rationalise the many irregularities and inconsistencies that > it throws up. > > I do not insist that my approach is "best" or the only path to a more acceptable > solution but it offers the potential for a radical improvement and puts the > issues into some sort of context. Moreover, I have adopted this approach to > start moving gradually from the existing model and, for example, I am not > wedded to the notion of a discrete 24-point IMP scale. Indeed, I think that this > may well be intrinsically unmanageable and inconsistent with the objectives of > the desired compression. > > Other approaches may indeed be better or more practical and I would be > interested in your ideas for a revised scoring table. For my part, I have been > reluctant to interfere with fundamental scoring units (trick values > etc) as they are too engrained and have restricted adjustments to bonus > values. > > You say that your revised scoring table works at the lower end of the scale > (part-scores and games) - possibly in the same way that a reduced IMP scale > "seems" to work? > > However, I don't think you can dismiss the idea of IMPing in some form and > linear IMPing simply gets you back to the problem you were trying to solve in > the first place. > > > Best regards > > Bill > -----Original Message----- > From: blml-bounces at rtflb.org [mailto:blml-bounces at rtflb.org] On Behalf Of > Herman De Wael > Sent: 04 July 2014 07:19 > To: Bridge Laws Mailing List > Subject: Re: [BLML] IMTables and IMPing multi-pair teams > > While this discussion is interesting, the real problem is not the current IMP- > table, but the process of IMPing itself. > You really need to imagine what you would need to do to tweek the > IMP-table: all documents around the world need to be re-made, and for a while > different versions will co-exist, leading to problems. All programs need to be re- > written, and a switch date needs to be determined. > And contrary to the new VP-scale, the two versions will not be readily > distinguishable, because you are apparentlu staying with a discrete 24-IMP > scale. > I doubt that the changes will be worth-while to convince the powers that be > that a change should be made. > > Rather, I think we should do away with IMPing altogether. Currently, there are > a few irregularities. An overtrick in the majors is worth one IMP, and two > overtricks are worth two IMPs. An overtrick in the minors is also worth one > IMP, and two overtricks ... also one. And an overtrick is sometimes worth > nothing if the other table happens to be in a radically different contract (or > going down in the same one). > If you bid game and they don't, you earn 10 or 12 IMPs. If you bid slam and they > don't, you also earn 10 or 12. But if you bid slam and they don't even bid game, > the gain is only 14 or 17. > And I could go on. > > The answer is a linear IMPtable, or no IMPtable at all, so using straight point > scoring. Of course this means that the scoring table needs to change. > > I have developed a start of such an adapted scoring table, and for the low > numbers (part scores, games), it works. > > Any and all input may be helpful. > > Herman. > > Bill Hood schreef: > > Hi Roger, > > > > I am indebted to you for the links to the articles on the early > > history of IMPs as I had spent considerable effort in trying to find > > such > material. > > > > _______________________________________________ > Blml mailing list > Blml at rtflb.org > http://lists.rtflb.org/mailman/listinfo/blml > > > _______________________________________________ > Blml mailing list > Blml at rtflb.org > http://lists.rtflb.org/mailman/listinfo/blml From anwalt at bley-strafrecht.de Mon Jul 7 08:24:07 2014 From: anwalt at bley-strafrecht.de (Dr. Richard Bley) Date: Mon, 7 Jul 2014 08:24:07 +0200 Subject: [BLML] IMTables and IMPing multi-pair teams In-Reply-To: <001901cf99a9$6cdefeb0$469cfc10$@online.no> References: <4709DE22F2E74C58957F9789CCD3EFCD@magnifique> <7E660BC78E1F4B25BC88F314C5FADE17@Pathfinder> <53B64741.8020702@skynet.be> <56CC7E49EB54411B87BDE163B4666A10@Pathfinder> <001901cf99a9$6cdefeb0$469cfc10$@online.no> Message-ID: <98AEF93A-1180-4F32-B78C-EAE756469D80@bley-strafrecht.de> Well Said!!! Dr. Richard Bley Rechtsanwalt Bredeneyer Str. 2b 45133 Essen Tel.: 0201 / 45 17 496 Fax: 0201 / 86 28 40 24 anwalt at bley-strafrecht.de > Am 07.07.2014 um 08:05 schrieb "Sven Pran" : > > In my profession (engineering) we have a saying: If it ain't broke then > don't fix it! > > I can't help feeling that you have completely failed to substantiate your > allegation that the IMP table is flawed. It does indeed fulfill its main > purpose which is to increase the impact from part score contracts so that > the victory of an event will not solely be determined by some few extreme > game results during a match. > > My experience from thirty (plus) years of directing is that players are > comfortable with the total points to IMP conversion as it is. > > (The last main "flaw" in scoring was rectified in 1987 when they increased > the bonus for setting doubled or redoubled non-vulnerable contracts more > than three down so that players could no longer benefit from sacrificing > against grand slams with eleven down when green v/ red.) > >> -----Opprinnelig melding----- >> Fra: blml-bounces at rtflb.org [mailto:blml-bounces at rtflb.org] P? vegne av > Bill >> Hood >> Sendt: 6. juli 2014 21:34 >> Til: 'Bridge Laws Mailing List' >> Emne: Re: [BLML] IMTables and IMPing multi-pair teams >> >> For the attention of Herman de Wae et al: >> >> Many thanks for your observations Herman. >> >> I do not under-estimate the enormity of the changes that would be > necessary >> to overhaul the existing IMP table and the associated IMPing process. >> >> However, nor do I under-estimate the enormity of the dis-functionality of > the >> existing methodology with its inconsistencies and inadequacies. >> >> Are we really saying that we cannot have a scoring system that is fit for >> purpose for the game or that, even if we could devise one, it would be too >> troublesome to work towards its introduction (over time)? >> >> Frankly, I think the game deserves better and that it is diminished > (perhaps >> forever) by accepting this fundamental undermining of its integrity. For >> instance, is there any real point in the finely judged arguments and >> mathematics for converting match IMPs to VPs if the IMP total is so >> haphazardly and carelessly calculated? >> >> We should not let the anticipation of future administrative issues > prohibit a >> debate/review of a better scoring system. >> >> My starting point is to insist that it is a fundamental requirement of any > scoring >> system that not only can it can be used to derive an understanding of how > best >> to play the game to maximise our chances of coming out ahead but also that >> the derived strategy is sensible and consistent. >> >>> From there, it seems sensible to set design objectives for what we want >>> the >> scoring system to do. By contrast, the existing IMP table and the IMP > process >> is an example of having introduced a process of scoring and then looked at > what >> the consequences of that process are in terms of the best bidding strategy > and >> trying to somehow rationalise the many irregularities and inconsistencies > that >> it throws up. >> >> I do not insist that my approach is "best" or the only path to a more > acceptable >> solution but it offers the potential for a radical improvement and puts > the >> issues into some sort of context. Moreover, I have adopted this approach > to >> start moving gradually from the existing model and, for example, I am not >> wedded to the notion of a discrete 24-point IMP scale. Indeed, I think > that this >> may well be intrinsically unmanageable and inconsistent with the > objectives of >> the desired compression. >> >> Other approaches may indeed be better or more practical and I would be >> interested in your ideas for a revised scoring table. For my part, I have > been >> reluctant to interfere with fundamental scoring units (trick values >> etc) as they are too engrained and have restricted adjustments to bonus >> values. >> >> You say that your revised scoring table works at the lower end of the > scale >> (part-scores and games) - possibly in the same way that a reduced IMP > scale >> "seems" to work? >> >> However, I don't think you can dismiss the idea of IMPing in some form and >> linear IMPing simply gets you back to the problem you were trying to solve > in >> the first place. >> >> >> Best regards >> >> Bill >> -----Original Message----- >> From: blml-bounces at rtflb.org [mailto:blml-bounces at rtflb.org] On Behalf Of >> Herman De Wael >> Sent: 04 July 2014 07:19 >> To: Bridge Laws Mailing List >> Subject: Re: [BLML] IMTables and IMPing multi-pair teams >> >> While this discussion is interesting, the real problem is not the current > IMP- >> table, but the process of IMPing itself. >> You really need to imagine what you would need to do to tweek the >> IMP-table: all documents around the world need to be re-made, and for a > while >> different versions will co-exist, leading to problems. All programs need > to be re- >> written, and a switch date needs to be determined. >> And contrary to the new VP-scale, the two versions will not be readily >> distinguishable, because you are apparentlu staying with a discrete 24-IMP >> scale. >> I doubt that the changes will be worth-while to convince the powers that > be >> that a change should be made. >> >> Rather, I think we should do away with IMPing altogether. Currently, there > are >> a few irregularities. An overtrick in the majors is worth one IMP, and two >> overtricks are worth two IMPs. An overtrick in the minors is also worth > one >> IMP, and two overtricks ... also one. And an overtrick is sometimes worth >> nothing if the other table happens to be in a radically different contract > (or >> going down in the same one). >> If you bid game and they don't, you earn 10 or 12 IMPs. If you bid slam > and they >> don't, you also earn 10 or 12. But if you bid slam and they don't even bid > game, >> the gain is only 14 or 17. >> And I could go on. >> >> The answer is a linear IMPtable, or no IMPtable at all, so using straight > point >> scoring. Of course this means that the scoring table needs to change. >> >> I have developed a start of such an adapted scoring table, and for the low >> numbers (part scores, games), it works. >> >> Any and all input may be helpful. >> >> Herman. >> >> Bill Hood schreef: >>> Hi Roger, >>> >>> I am indebted to you for the links to the articles on the early >>> history of IMPs as I had spent considerable effort in trying to find >>> such >> material. >> >> _______________________________________________ >> Blml mailing list >> Blml at rtflb.org >> http://lists.rtflb.org/mailman/listinfo/blml >> >> >> _______________________________________________ >> Blml mailing list >> Blml at rtflb.org >> http://lists.rtflb.org/mailman/listinfo/blml > > > _______________________________________________ > Blml mailing list > Blml at rtflb.org > http://lists.rtflb.org/mailman/listinfo/blml From hermandw at skynet.be Mon Jul 7 08:36:14 2014 From: hermandw at skynet.be (Herman De Wael) Date: Mon, 07 Jul 2014 08:36:14 +0200 Subject: [BLML] IMTables and IMPing multi-pair teams In-Reply-To: <56CC7E49EB54411B87BDE163B4666A10@Pathfinder> References: <4709DE22F2E74C58957F9789CCD3EFCD@magnifique><7E660BC78E1F4B25BC88F314C5FADE17@Pathfinder> <53B64741.8020702@skynet.be> <56CC7E49EB54411B87BDE163B4666A10@Pathfinder> Message-ID: <53BA3FDE.8020804@skynet.be> Hello Bill, I thank you for the attention you have given my post, but you seem to misunderstand my point. I do understand your concerns, and I do agree with them. My main problem is that, if you try to change the current IMP-table for a "better" one, you are going to come across huge difficulties, some of which will be insurmountable. Just think of the reaction of those players that have memorized the current table! And even if you succeed, you may find that not all the problems concerning the IMP-table are solved. For instance, you seem to suggest a 24-IMP-scale. That will probably still leave 0 and 10 at 0 IMPs, something which I find wrong. That is why I suggest we go even further than what you are saying, and I would do without IMPing at all. I feel that the players would approve of one big change (the scoring table) if it means they can do away with IMPing altogether. OK? Herman. Bill Hood schreef: > For the attention of Herman de Wae et al: > > Many thanks for your observations Herman. > > I do not under-estimate the enormity of the changes that would be necessary > to overhaul the existing IMP table and the associated IMPing process. > > However, nor do I under-estimate the enormity of the dis-functionality of > the existing methodology with its inconsistencies and inadequacies. > > Are we really saying that we cannot have a scoring system that is fit for > purpose for the game or that, even if we could devise one, it would be too > troublesome to work towards its introduction (over time)? > > Frankly, I think the game deserves better and that it is diminished (perhaps > forever) by accepting this fundamental undermining of its integrity. For > instance, is there any real point in the finely judged arguments and > mathematics for converting match IMPs to VPs if the IMP total is so > haphazardly and carelessly calculated? > > We should not let the anticipation of future administrative issues prohibit > a debate/review of a better scoring system. > > My starting point is to insist that it is a fundamental requirement of any > scoring system that not only can it can be used to derive an understanding > of how best to play the game to maximise our chances of coming out ahead but > also that the derived strategy is sensible and consistent. > >>From there, it seems sensible to set design objectives for what we want the > scoring system to do. By contrast, the existing IMP table and the IMP > process is an example of having introduced a process of scoring and then > looked at what the consequences of that process are in terms of the best > bidding strategy and trying to somehow rationalise the many irregularities > and inconsistencies that it throws up. > > I do not insist that my approach is "best" or the only path to a more > acceptable solution but it offers the potential for a radical improvement > and puts the issues into some sort of context. Moreover, I have adopted this > approach to start moving gradually from the existing model and, for example, > I am not wedded to the notion of a discrete 24-point IMP scale. Indeed, I > think that this may well be intrinsically unmanageable and inconsistent with > the objectives of the desired compression. > > Other approaches may indeed be better or more practical and I would be > interested in your ideas for a revised scoring table. For my part, I have > been reluctant to interfere with fundamental scoring units (trick values > etc) as they are too engrained and have restricted adjustments to bonus > values. > > You say that your revised scoring table works at the lower end of the scale > (part-scores and games) - possibly in the same way that a reduced IMP scale > "seems" to work? > > However, I don't think you can dismiss the idea of IMPing in some form and > linear IMPing simply gets you back to the problem you were trying to solve > in the first place. > > > Best regards > > Bill > -----Original Message----- > From: blml-bounces at rtflb.org [mailto:blml-bounces at rtflb.org] On Behalf Of > Herman De Wael > Sent: 04 July 2014 07:19 > To: Bridge Laws Mailing List > Subject: Re: [BLML] IMTables and IMPing multi-pair teams > > While this discussion is interesting, the real problem is not the > current IMP-table, but the process of IMPing itself. > You really need to imagine what you would need to do to tweek the > IMP-table: all documents around the world need to be re-made, and for a > while different versions will co-exist, leading to problems. All > programs need to be re-written, and a switch date needs to be determined. > And contrary to the new VP-scale, the two versions will not be readily > distinguishable, because you are apparentlu staying with a discrete > 24-IMP scale. > I doubt that the changes will be worth-while to convince the powers that > be that a change should be made. > > Rather, I think we should do away with IMPing altogether. Currently, > there are a few irregularities. An overtrick in the majors is worth one > IMP, and two overtricks are worth two IMPs. An overtrick in the minors > is also worth one IMP, and two overtricks ... also one. And an overtrick > is sometimes worth nothing if the other table happens to be in a > radically different contract (or going down in the same one). > If you bid game and they don't, you earn 10 or 12 IMPs. If you bid slam > and they don't, you also earn 10 or 12. But if you bid slam and they > don't even bid game, the gain is only 14 or 17. > And I could go on. > > The answer is a linear IMPtable, or no IMPtable at all, so using > straight point scoring. Of course this means that the scoring table > needs to change. > > I have developed a start of such an adapted scoring table, and for the > low numbers (part scores, games), it works. > > Any and all input may be helpful. > > Herman. > > Bill Hood schreef: >> Hi Roger, >> >> I am indebted to you for the links to the articles on the early history >> of IMPs as I had spent considerable effort in trying to find such > material. >> > > _______________________________________________ > Blml mailing list > Blml at rtflb.org > http://lists.rtflb.org/mailman/listinfo/blml > > > _______________________________________________ > Blml mailing list > Blml at rtflb.org > http://lists.rtflb.org/mailman/listinfo/blml > > From richard.hills at immi.gov.au Mon Jul 7 08:42:27 2014 From: richard.hills at immi.gov.au (Richard James HILLS) Date: Mon, 7 Jul 2014 06:42:27 +0000 Subject: [BLML] QWERTY (was IMP tables) [SEC=UNOFFICIAL] Message-ID: <240635A98F59F24AAD8510EB05121DC3395517F0@SDCWPIPEX02.IMMI.LOCAL> UNOFFICIAL Sven Pran: >>..... >>(The last main ?flaw? in scoring was rectified in 1987 when they increased >>the bonus for setting doubled or redoubled non-vulnerable contracts more >>than three down so that players could no longer benefit from sacrificing >>against grand slams with eleven down when green v/ red.) Richard Bley: >Well Said!!! > >Dr. Richard Bley >Rechtsanwalt >Bredeneyer Str. 2b >45133 Essen >Tel.: 0201 / 45 17 496 >Fax: 0201 / 86 28 40 24 >anwalt at bley-strafrecht.de Richard Hills: But famous American expert Grant Baze disagrees with Richard Bley and Sven Pran. http://blakjak.org/law_crit.htm (For what it is worth I disagree with Grant Baze and agree with Richard Bley and Sven Pran.) Best wishes, Hilda R. Lirsch hildalirsch at gmail.com UNOFFICIAL -------------------------------------------------------------------- Important Notice: If you have received this email by mistake, please advise the sender and delete the message and attachments immediately. This email, including attachments, may contain confidential, sensitive, legally privileged and/or copyright information. Any review, retransmission, dissemination or other use of this information by persons or entities other than the intended recipient is prohibited. DIBP respects your privacy and has obligations under the Privacy Act 1988. The official departmental privacy policy can be viewed on the department's website at www.immi.gov.au. See: http://www.immi.gov.au/functional/privacy.htm --------------------------------------------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.rtflb.org/pipermail/blml/attachments/20140707/0834d23c/attachment.html From agot at ulb.ac.be Mon Jul 7 11:24:39 2014 From: agot at ulb.ac.be (Alain Gottcheiner) Date: Mon, 07 Jul 2014 11:24:39 +0200 Subject: [BLML] QWERTY (was IMP tables) [SEC=UNOFFICIAL] In-Reply-To: <240635A98F59F24AAD8510EB05121DC3395517F0@SDCWPIPEX02.IMMI.LOCAL> References: <240635A98F59F24AAD8510EB05121DC3395517F0@SDCWPIPEX02.IMMI.LOCAL> Message-ID: <53BA6757.4040108@ulb.ac.be> Le 7/07/2014 8:42, Richard James HILLS a ?crit : > UNOFFICIAL > Sven Pran: > >>..... > >>(The last main "flaw" in scoring was rectified in 1987 when they > increased > >>the bonus for setting doubled or redoubled non-vulnerable contracts more > >>than three down so that players could no longer benefit from sacrificing > >>against grand slams with eleven down when green v/ red.) > I wonder why many considered this a flaw. Or, more specifically, why it is a flaw to be able to play 1NT* -7 against a vulnerable small slam and win the board ? ISTR that a Polish player once achieved this in an international match. Was this one more step in the crusade against psyches ? IMOBO rules which allow for more strategic ploys are good for the game. Come to think of it, it is only an extension of the principles of sacrificing. Do you want to ban preemptive bids ? Best regards Alain -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.rtflb.org/pipermail/blml/attachments/20140707/d5f09851/attachment-0001.html From hermandw at skynet.be Mon Jul 7 11:50:39 2014 From: hermandw at skynet.be (Herman De Wael) Date: Mon, 07 Jul 2014 11:50:39 +0200 Subject: [BLML] QWERTY (was IMP tables) [SEC=UNOFFICIAL] In-Reply-To: <53BA6757.4040108@ulb.ac.be> References: <240635A98F59F24AAD8510EB05121DC3395517F0@SDCWPIPEX02.IMMI.LOCAL> <53BA6757.4040108@ulb.ac.be> Message-ID: <53BA6D6F.5010009@skynet.be> Well Alain, you cannot speak of a flaw. But you can speak of tactics that are undesirable for the game. If, against 7H, you can freely bid 7S, winning the board on just 2 tricks, then this becomes too easy a tactic. And some believe that this is undesirable for the game. I believe that it cas a good thing to put the penalties for going down (doubled) at just one trick different between vulnerable and non-vulnerable. That is my personal opinion. Herman. Alain Gottcheiner schreef: > Le 7/07/2014 8:42, Richard James HILLS a ?crit : >> UNOFFICIAL >> Sven Pran: >> >>..... >> >>(The last main ?flaw? in scoring was rectified in 1987 when they increased >> >>the bonus for setting doubled or redoubled non-vulnerable contracts more >> >>than three down so that players could no longer benefit from sacrificing >> >>against grand slams with eleven down when green v/ red.) >> > > I wonder why many considered this a flaw. Or, more specifically, why it > is a flaw to be able to play 1NT* -7 against a vulnerable small slam and > win the board ? > ISTR that a Polish player once achieved this in an international match. > > Was this one more step in the crusade against psyches ? > > IMOBO rules which allow for more strategic ploys are good for the game. > Come to think of it, it is only an extension of the principles of > sacrificing. Do you want to ban preemptive bids ? > > > Best regards > > > Alain > > > _______________________________________________ > Blml mailing list > Blml at rtflb.org > http://lists.rtflb.org/mailman/listinfo/blml > From ardelm at optusnet.com.au Mon Jul 7 12:00:41 2014 From: ardelm at optusnet.com.au (Tony Musgrove) Date: Mon, 7 Jul 2014 20:00:41 +1000 Subject: [BLML] IMTables and IMPing multi-pair teams In-Reply-To: <53BA3FDE.8020804@skynet.be> References: <4709DE22F2E74C58957F9789CCD3EFCD@magnifique><7E660BC78E1F4B25BC88F314C5FADE17@Pathfinder> <53B64741.8020702@skynet.be> <56CC7E49EB54411B87BDE163B4666A10@Pathfinder> <53BA3FDE.8020804@skynet.be> Message-ID: <002001cf99ca$4ffda460$eff8ed20$@optusnet.com.au> My 2 cents worth (for the tenth time). Suppose you are playing in a "butler swiss pairs", as they do in Australia quite often. Suppose there are 20 tables and the "butler average" is rounded up to the nearest 10 on each board. Suppose that 11 of the NS pairs record a "win". Is that fair? I always use Herman's Bastille method of linearizing the imp scales so that each small difference in scores leads to a different IMP score in cross imps. This method gets as close to zero as the average overall score in Howell type movements, as it is possible, and I have seen discrepancies amounting to more than 2 IMPs by using "Butler scoring". I think this is wrong. People like to have a published "datum" with which to compare their actual scores (it is said), but it is much easier (as on BBO) to find that your score is +2.3 IMPs on the board. They calculate their average cross imps score with integral IMPs as per the usual scale, but not to worry if compared with several scores. Cheers, Tony (Sydney) > -----Original Message----- > From: blml-bounces at rtflb.org [mailto:blml-bounces at rtflb.org] On > Behalf Of Herman De Wael > Sent: Monday, 7 July 2014 4:36 PM > To: Bridge Laws Mailing List > Subject: Re: [BLML] IMTables and IMPing multi-pair teams > > Hello Bill, > > I thank you for the attention you have given my post, but you seem to > misunderstand my point. > > I do understand your concerns, and I do agree with them. > > My main problem is that, if you try to change the current IMP-table for > a "better" one, you are going to come across huge difficulties, some of > which will be insurmountable. Just think of the reaction of those > players that have memorized the current table! > > And even if you succeed, you may find that not all the problems > concerning the IMP-table are solved. For instance, you seem to suggest a > 24-IMP-scale. That will probably still leave 0 and 10 at 0 IMPs, > something which I find wrong. > > That is why I suggest we go even further than what you are saying, and I > would do without IMPing at all. I feel that the players would approve of > one big change (the scoring table) if it means they can do away with > IMPing altogether. > > OK? > > Herman. > > Bill Hood schreef: > > For the attention of Herman de Wae et al: > > > > Many thanks for your observations Herman. > > > > I do not under-estimate the enormity of the changes that would be > necessary > > to overhaul the existing IMP table and the associated IMPing process. > > > > However, nor do I under-estimate the enormity of the dis-functionality > of > > the existing methodology with its inconsistencies and inadequacies. > > > > Are we really saying that we cannot have a scoring system that is fit for > > purpose for the game or that, even if we could devise one, it would be > too > > troublesome to work towards its introduction (over time)? > > > > Frankly, I think the game deserves better and that it is diminished > (perhaps > > forever) by accepting this fundamental undermining of its integrity. For > > instance, is there any real point in the finely judged arguments and > > mathematics for converting match IMPs to VPs if the IMP total is so > > haphazardly and carelessly calculated? > > > > We should not let the anticipation of future administrative issues > prohibit > > a debate/review of a better scoring system. > > > > My starting point is to insist that it is a fundamental requirement of any > > scoring system that not only can it can be used to derive an > understanding > > of how best to play the game to maximise our chances of coming out > ahead but > > also that the derived strategy is sensible and consistent. > > > >>From there, it seems sensible to set design objectives for what we > want the > > scoring system to do. By contrast, the existing IMP table and the IMP > > process is an example of having introduced a process of scoring and > then > > looked at what the consequences of that process are in terms of the > best > > bidding strategy and trying to somehow rationalise the many > irregularities > > and inconsistencies that it throws up. > > > > I do not insist that my approach is "best" or the only path to a more > > acceptable solution but it offers the potential for a radical > improvement > > and puts the issues into some sort of context. Moreover, I have > adopted this > > approach to start moving gradually from the existing model and, for > example, > > I am not wedded to the notion of a discrete 24-point IMP scale. Indeed, > I > > think that this may well be intrinsically unmanageable and inconsistent > with > > the objectives of the desired compression. > > > > Other approaches may indeed be better or more practical and I would > be > > interested in your ideas for a revised scoring table. For my part, I have > > been reluctant to interfere with fundamental scoring units (trick values > > etc) as they are too engrained and have restricted adjustments to > bonus > > values. > > > > You say that your revised scoring table works at the lower end of the > scale > > (part-scores and games) - possibly in the same way that a reduced IMP > scale > > "seems" to work? > > > > However, I don't think you can dismiss the idea of IMPing in some form > and > > linear IMPing simply gets you back to the problem you were trying to > solve > > in the first place. > > > > > > Best regards > > > > Bill > > -----Original Message----- > > From: blml-bounces at rtflb.org [mailto:blml-bounces at rtflb.org] On > Behalf Of > > Herman De Wael > > Sent: 04 July 2014 07:19 > > To: Bridge Laws Mailing List > > Subject: Re: [BLML] IMTables and IMPing multi-pair teams > > > > While this discussion is interesting, the real problem is not the > > current IMP-table, but the process of IMPing itself. > > You really need to imagine what you would need to do to tweek the > > IMP-table: all documents around the world need to be re-made, and > for a > > while different versions will co-exist, leading to problems. All > > programs need to be re-written, and a switch date needs to be > determined. > > And contrary to the new VP-scale, the two versions will not be readily > > distinguishable, because you are apparentlu staying with a discrete > > 24-IMP scale. > > I doubt that the changes will be worth-while to convince the powers > that > > be that a change should be made. > > > > Rather, I think we should do away with IMPing altogether. Currently, > > there are a few irregularities. An overtrick in the majors is worth one > > IMP, and two overtricks are worth two IMPs. An overtrick in the minors > > is also worth one IMP, and two overtricks ... also one. And an overtrick > > is sometimes worth nothing if the other table happens to be in a > > radically different contract (or going down in the same one). > > If you bid game and they don't, you earn 10 or 12 IMPs. If you bid slam > > and they don't, you also earn 10 or 12. But if you bid slam and they > > don't even bid game, the gain is only 14 or 17. > > And I could go on. > > > > The answer is a linear IMPtable, or no IMPtable at all, so using > > straight point scoring. Of course this means that the scoring table > > needs to change. > > > > I have developed a start of such an adapted scoring table, and for the > > low numbers (part scores, games), it works. > > > > Any and all input may be helpful. > > > > Herman. > > > > Bill Hood schreef: > >> Hi Roger, > >> > >> I am indebted to you for the links to the articles on the early history > >> of IMPs as I had spent considerable effort in trying to find such > > material. > >> > > > > _______________________________________________ > > Blml mailing list > > Blml at rtflb.org > > http://lists.rtflb.org/mailman/listinfo/blml > > > > > > _______________________________________________ > > Blml mailing list > > Blml at rtflb.org > > http://lists.rtflb.org/mailman/listinfo/blml > > > > > _______________________________________________ > Blml mailing list > Blml at rtflb.org > http://lists.rtflb.org/mailman/listinfo/blml From agot at ulb.ac.be Mon Jul 7 12:12:07 2014 From: agot at ulb.ac.be (Alain Gottcheiner) Date: Mon, 07 Jul 2014 12:12:07 +0200 Subject: [BLML] QWERTY (was IMP tables) [SEC=UNOFFICIAL] In-Reply-To: <53BA6D6F.5010009@skynet.be> References: <240635A98F59F24AAD8510EB05121DC3395517F0@SDCWPIPEX02.IMMI.LOCAL> <53BA6757.4040108@ulb.ac.be> <53BA6D6F.5010009@skynet.be> Message-ID: <53BA7277.4050205@ulb.ac.be> Le 7/07/2014 11:50, Herman De Wael a ?crit : > Well Alain, > > you cannot speak of a flaw. Some did. > > But you can speak of tactics that are undesirable for the game. > > If, against 7H, you can freely bid 7S, winning the board on just 2 > tricks, then this becomes too easy a tactic. Did you say "too easy" ? Hold your horses .... These tactics are in fact quite dangerous. Last time they sacrificed against our slam (that was in Deauville 2011, so not a very common occurrence), the score was 1100 for 5 down, no one vulnerable, in 7H against our 7D. The problem (not for us) was that there was an inescapable slow loser in Spades. Since the score at the other table was 500 (in 6HX) the 7H bid changed a score of +11 into -12. In a 4-board match. So, IMO, gaining some IMPs in 7SX against 7H making is a just reward for determining/gambling that he opponents were right to play in 7. Also, the old scale added to tactical considerations by making it occasionally right to stop in 6 and hope for no sacrifice (because they don't want to push us) rather than playing in 7 and seeing them sacrifice. Best regards Alain From agot at ulb.ac.be Mon Jul 7 12:25:10 2014 From: agot at ulb.ac.be (Alain Gottcheiner) Date: Mon, 07 Jul 2014 12:25:10 +0200 Subject: [BLML] QWERTY (was IMP tables) [SEC=UNOFFICIAL] In-Reply-To: <53BA6D6F.5010009@skynet.be> References: <240635A98F59F24AAD8510EB05121DC3395517F0@SDCWPIPEX02.IMMI.LOCAL> <53BA6757.4040108@ulb.ac.be> <53BA6D6F.5010009@skynet.be> Message-ID: <53BA7586.8000509@ulb.ac.be> Le 7/07/2014 11:50, Herman De Wael a ?crit : > Well Alain, > > you cannot speak of a flaw. > > But you can speak of tactics that are undesirable for the game. > > If, against 7H, you can freely bid 7S, winning the board on just 2 > tricks, then this becomes too easy a tactic. And some believe that this > is undesirable for the game. > I believe that it cas a good thing to put the penalties for going down > (doubled) at just one trick different between vulnerable and non-vulnerable. > > The new rules don't completely bar those extreme sacrifices. It is still possible to open a 9-11 NT, be doubled and play there, making just one trick, and register a good board against a vulnerable 6-in-major contract from the field. And of course it is still possible in undoubled contracts. Look at the following deal : KQTxxx Jx xxx xxx A xx Ax KQTx KQJxxxx Ax Qxx AJxxx Jxxx xxxxx x KTx North opened 2D (weak in H or S). East's methods called for a pass. South passed !!! Such 0-trick contracts are part of the game, making their use ineffective is cutting a part of the game. Best regards Alain PS : the score was an average. After 2D-p-3H it is difficult to find 6D abyway, and both sides were vulnerable. From cibor at poczta.fm Mon Jul 7 19:38:19 2014 From: cibor at poczta.fm (Konrad Ciborowski) Date: Mon, 7 Jul 2014 19:38:19 +0200 Subject: [BLML] QWERTY (was IMP tables) [SEC=UNOFFICIAL] In-Reply-To: <53BA6757.4040108@ulb.ac.be> References: <240635A98F59F24AAD8510EB05121DC3395517F0@SDCWPIPEX02.IMMI.LOCAL> <53BA6757.4040108@ulb.ac.be> Message-ID: From: Alain Gottcheiner Sent: Monday, July 07, 2014 11:24 AM To: Bridge Laws Mailing List Subject: Re: [BLML] QWERTY (was IMP tables) [SEC=UNOFFICIAL] Le 7/07/2014 8:42, Richard James HILLS a ?crit : UNOFFICIAL Sven Pran: >>..... >>(The last main ?flaw? in scoring was rectified in 1987 when they increased >>the bonus for setting doubled or redoubled non-vulnerable contracts more >>than three down so that players could no longer benefit from sacrificing >>against grand slams with eleven down when green v/ red.) I wonder why many considered this a flaw. Or, more specifically, why it is a flaw to be able to play 1NT* -7 against a vulnerable small slam and win the board ? ISTR that a Polish player once achieved this in an international match. [KC] It was Henryk Nied?wiecki back in 1957 in the European Championships (first time when Poland participated) against Austria. And the contract was 1Sx. The story began in Mitropa Cup a few months earlier when Schneider opened a psychic 1D in first seat on nothing (those were the days...). Nied?wiecki - Wieczorkiewicz somehow managed to find diamonds (it was like 1D - 1S - pass - 3D and it was a natural bid in those days - it would be pretty much impossible to play in diamonds in modern systems) but were unable to reach the slam bid in the other room. 5D they bid was a loss. Fast forward to the European Championships a few months later. Nied?wiecki who had taken Schneider's psyche personally was determined to get even. So come the match against Austria he opened 1S in first seat with a yarborough (those were the days...) and it went double, pass, pass to him. Instead of running he chose to pass! The result was spectacular. He went seven down which at the time was only -1300 and was a huge gain against 7S bid in the other room. The layout can be found in multiple places in Polish bridge literature. Not surprisignly this style of play didn't bring Poland many points, though, and Poland barely avoided the indignity of finishing last. Best regards, Konrad Ciborowski Krak?w, Poland From jimfox00 at cox.net Mon Jul 7 23:54:32 2014 From: jimfox00 at cox.net (Jim Fox) Date: Mon, 7 Jul 2014 17:54:32 -0400 Subject: [BLML] QWERTY (was IMP tables) [SEC=UNOFFICIAL] In-Reply-To: References: Message-ID: Here's my suggestion: "nothing". Seriously, my suggestion is let BLML be about "bridge laws", but not "bridge lawyering" nor esoterica. Mmbridge From: blml-bounces at rtflb.org [mailto:blml-bounces at rtflb.org] On Behalf Of Jim Fox Sent: 07/06/2014 7:49 PM To: 'Bridge Laws Mailing List' Cc: hildalirsch at gmail.com Subject: Re: [BLML] QWERTY (was IMP tables) [SEC=UNOFFICIAL] Most of my contributions were made 40 years ago (not in BLML). Your comments are nonsense since they are sensible only in the context of "expert" players. I speak of ALL players who in fact do gain from the randomness of the imp scale, random seeding, and the like. Try to think out of your box. Mmbridge From: blml-bounces at rtflb.org [mailto:blml-bounces at rtflb.org] On Behalf Of Richard James HILLS Sent: 07/06/2014 7:23 PM To: Laws Bridge Cc: hildalirsch at gmail.com Subject: Re: [BLML] QWERTY (was IMP tables) [SEC=UNOFFICIAL] UNOFFICIAL I do not wish to start a flame war, but in my opinion writing "Your high- toned rhetoric is overdone" about the enthusiastic contribution of a blml newbie is unfair. Especially when I cannot recall the critic making a single useful suggestion himself over the past decade. As for the reason why the Law 78B imps scale has been unchanged during the past half-century, it is NOT "The majority of players want there to be the amount of [imps] scoring randomness we have now (even if they don't consciously realize it)". Rather the reason for the unchanging Law 78B imps scale is that it is an adequate solution to the flaws of total-point scoring (as discussed by David Grabiner), and that the gains from a slightly more adequate hypothetical 2015 Law 78B would not overcome the lock-in costs. See: http://en.wikipedia.org/wiki/Network_effect#Lock-in UNOFFICIAL -------------------------------------------------------------------- Important Notice: If you have received this email by mistake, please advise the sender and delete the message and attachments immediately. This email, including attachments, may contain confidential, sensitive, legally privileged and/or copyright information. Any review, retransmission, dissemination or other use of this information by persons or entities other than the intended recipient is prohibited. DIBP respects your privacy and has obligations under the Privacy Act 1988. The official departmental privacy policy can be viewed on the department's website at www.immi.gov.au. See: http://www.immi.gov.au/functional/privacy.htm --------------------------------------------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.rtflb.org/pipermail/blml/attachments/20140707/fd4306a9/attachment-0001.html From richard.hills at immi.gov.au Tue Jul 8 05:29:42 2014 From: richard.hills at immi.gov.au (Richard James HILLS) Date: Tue, 8 Jul 2014 03:29:42 +0000 Subject: [BLML] QWERTY (was IMP tables) [SEC=UNOFFICIAL] Message-ID: <240635A98F59F24AAD8510EB05121DC339556986@SDCWPIPEX02.IMMI.LOCAL> UNOFFICIAL Douglas Adams, The Hitch-Hiker?s Guide to the Galaxy: ?Space,? it says, ?is big. Really big. You just won?t believe how vastly, hugely, mind-bogglingly big it is. I mean, you may think it?s a long way down the road to the chemist?s, but that?s just peanuts to space ?.. ? Bill Hood: [mind-bogglingly big snip] >More discretion is required at the higher end with grand slams >and, while you will come out ahead in the long run if you bid >grand slams at and above their threshold values, it is necessary >to temper aggression whenever there is reason to suspect that >your opponents may not bid beyond game. In such situations, >the differential gain between a small and a grand slam is not >that significant and doesn?t justify the risk. Richard Hills: Indeed. On three occasions during Edgar Kaplan?s IMPs career his partnership bid to an excellent grand slam, one off due to filthy breaks, only to discover that the pair at the other table had stopped in a Part-Score. ? ? Herman De Wael: >>Well Alain, >> >>you cannot speak of a flaw. >> >>But you can speak of tactics that are undesirable for the game. >> I>>If, against 7H, you can freely bid 7S, winning the board on just >>2 tricks, then this becomes too easy a tactic. And some believe >>that this is undesirable for the game. >> >>I believe that it was a good thing to put the penalties for going >>down (doubled) at just one trick different between vulnerable >>and non-vulnerable. >> >>That is my personal opinion. Richard Hills: Yes and No. I agree 100% with Herman with the trivial exception that I believe our views are not merely ?personal opinion? but rather objective reality. The 2007 Lawbook formalised the split of the old game into two distinct games, Contract (Rubber) Bridge and Duplicate Bridge. The purpose of Rubber Bridge is to have a high level of luck. The purpose of Duplicate Bridge is to have a lower level of luck, with skilfull bidding and play more frequently rewarded. A partnership which has accurately bid to a grand slam has demonstrated a high level of skill, so the scoring system must be designed to prevent that skill being cancelled by a random favourable vulnerability sacrifice. Best wishes, Hilda R. Lirsch hildalirsch at gmail.com UNOFFICIAL -------------------------------------------------------------------- Important Notice: If you have received this email by mistake, please advise the sender and delete the message and attachments immediately. This email, including attachments, may contain confidential, sensitive, legally privileged and/or copyright information. Any review, retransmission, dissemination or other use of this information by persons or entities other than the intended recipient is prohibited. DIBP respects your privacy and has obligations under the Privacy Act 1988. The official departmental privacy policy can be viewed on the department's website at www.immi.gov.au. See: http://www.immi.gov.au/functional/privacy.htm --------------------------------------------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.rtflb.org/pipermail/blml/attachments/20140708/5c42a0fa/attachment.html From moysian at talktalk.net Tue Jul 8 21:54:11 2014 From: moysian at talktalk.net (Bill Hood) Date: Tue, 8 Jul 2014 20:54:11 +0100 Subject: [BLML] QWERTY (was IMP tables) [SEC=UNOFFICIAL] In-Reply-To: <240635A98F59F24AAD8510EB05121DC331AEA739@SDCWPIPEX02.IMMI.LOCAL> References: <240635A98F59F24AAD8510EB05121DC331AEA739@SDCWPIPEX02.IMMI.LOCAL> Message-ID: <0AA261A403CA43ADB08F5C0A9EB74A97@Pathfinder> Thanks Richard for defending my newbie innocence but I suspect there was nothing to fear from the attack. Yes, I wanted to start a debate but not with anyone who advocates in favour of deliberately randomising the scoring and seeding just so that the game generates some "fun" results. First, it is not necessary as, while it is a game of skill and judgement, there will always be situations where the "right" thing is wrong and the "wrong" thing is right. Second, there is a dimension to the game which invites tactics over strategy and various ploys and gambits may prove effective according to the situation at the table or the state of the match. As they say, "stuff happens" (forgive the rhetoric) but we cannot legislate for all obscure, unlikely and unfortunate scores. However, we simply HAVE to build the scoring to reward skill and judgement and the scoring MUST make sense in the context of "expert" players, otherwise the game itself makes no sense. From time-to-time, we may have to make adjustments when clear injustices are noticed AND agreed to be undesirable (such as the increase in the "insult" and the cost of excessive non-vulnerable undertricks.) My argument is that the existing IMP table is flawed to an extent that is unacceptable for the demands/ expectations of the modern game. It is a relic from an earlier time when it was primarily intended as a means of curbing the worst excesses of aggregate scoring and concentration was mistakenly over-focussed on trying to develop equitable VP interval ranges to an extent that ignored the implications that that gave rise to in terms of a sensible bidding strategy. I would again clarify, that my approach has been deliberately geared to working away from the existing methodology in gradual steps but that I do not rule out ultimately abandoning it almost entirely if that proves to be best. Specifically, I would repeat - for the benefit of Herman de Wael - that I am not in any way convinced that a 24-point scale is viable and I have been looking at reduced IMP scales. It may be that we are more closely aligned than he thinks and I would be pleased to consider any suggestions he has to make with regard to his revised scoring table. For clarification, I would say that the EBU had suggested that additional IMP tables could be generated and used directly for scoring multi-pair teams (8s, 12s and 16s) by literally multiplying the end-points of the existing (teams-of-four) IMP table ranges by, respectively, v2 v3 and v4. This is not quite the aka cross IMPs (as Richard Hills possibly thought) and it effectively randomises the threshold values at which games etc have to be bid according to a combination of happenstance and the number of pairs in the team. Their suggestion was made on a take-it-or-leave-it basis but it was yet another vagary of the existing IMP scoring process that showed the need for reform. Bill Hood _____ From: blml-bounces at rtflb.org [mailto:blml-bounces at rtflb.org] On Behalf Of Richard James HILLS Sent: 07 July 2014 00:23 To: Laws Bridge Cc: hildalirsch at gmail.com Subject: Re: [BLML] QWERTY (was IMP tables) [SEC=UNOFFICIAL] UNOFFICIAL I do not wish to start a flame war, but in my opinion writing "Your high- toned rhetoric is overdone" about the enthusiastic contribution of a blml newbie is unfair. Especially when I cannot recall the critic making a single useful suggestion himself over the past decade. UNOFFICIAL -------------------------------------------------------------------- Important Notice: If you have received this email by mistake, please advise the sender and delete the message and attachments immediately. This email, including attachments, may contain confidential, sensitive, legally privileged and/or copyright information. Any review, retransmission, dissemination or other use of this information by persons or entities other than the intended recipient is prohibited. DIBP respects your privacy and has obligations under the Privacy Act 1988. The official departmental privacy policy can be viewed on the department's website at www.immi.gov.au. See: http://www.immi.gov.au/functional/privacy.htm --------------------------------------------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.rtflb.org/pipermail/blml/attachments/20140708/19e7b8f6/attachment.html From g3 at nige1.com Wed Jul 9 05:32:22 2014 From: g3 at nige1.com (Nigel Guthrie) Date: Wed, 9 Jul 2014 04:32:22 +0100 Subject: [BLML] QWERTY (was IMP tables) [SEC=UNOFFICIAL] In-Reply-To: <0AA261A403CA43ADB08F5C0A9EB74A97@Pathfinder> References: <240635A98F59F24AAD8510EB05121DC331AEA739@SDCWPIPEX02.IMMI.LOCAL> <0AA261A403CA43ADB08F5C0A9EB74A97@Pathfinder> Message-ID: <31B9FA71FB61431F921946C6C9937C7B@G3> [Bill Hood] Thanks Richard for defending my newbie innocence but I suspect there was nothing to fear from the attack. Yes, I wanted to start a debate but not with anyone who advocates in favour of deliberately randomising the scoring and seeding just so that the game generates some ?fun? results. First, it is not necessary as, while it is a game of skill and judgement, there will always be situations where the ?right? thing is wrong and the ?wrong? thing is right. Second, there is a dimension to the game which invites tactics over strategy and various ploys and gambits may prove effective according to the situation at the table or the state of the match. As they say, ?stuff happens? (forgive the rhetoric) but we cannot legislate for all obscure, unlikely and unfortunate scores. However, we simply HAVE to build the scoring to reward skill and judgement and the scoring MUST make sense in the context of ?expert? players, otherwise the game itself makes no sense. From time?to-time, we may have to make adjustments when clear injustices are noticed AND agreed to be undesirable (such as the increase in the ?insult? and the cost of excessive non-vulnerable undertricks.) My argument is that the existing IMP table is flawed to an extent that is unacceptable for the demands/ expectations of the modern game. It is a relic from an earlier time when it was primarily intended as a means of curbing the worst excesses of aggregate scoring and concentration was mistakenly over-focussed on trying to develop equitable VP interval ranges to an extent that ignored the implications that that gave rise to in terms of a sensible bidding strategy. I would again clarify, that my approach has been deliberately geared to working away from the existing methodology in gradual steps but that I do not rule out ultimately abandoning it almost entirely if that proves to be best. Specifically, I would repeat ? for the benefit of Herman de Wael - that I am not in any way convinced that a 24-point scale is viable and I have been looking at reduced IMP scales. It may be that we are more closely aligned than he thinks and I would be pleased to consider any suggestions he has to make with regard to his revised scoring table.For clarification, I would say that the EBU had suggested that additional IMP tables could be generated and used directly for scoring multi-pair teams (8s, 12s and 16s) by literally multiplying the end-points of the existing (teams-of-four) IMP table ranges by, respectively, ?2 ?3 and ?4. This is not quite the aka cross IMPs (as Richard Hills possibly thought) and it effectively randomises the threshold values at which games etc have to be bid according to a combination of happenstance and the number of pairs in the team. Their suggestion was made on a take-it-or-leave-it basis but it was yet another vagary of the existing IMP scoring process that showed the need for reform. [Nige1] I agree that the Bridge scoring should reward skill. Please tell us, Bill, what current IMP-table flaws concern you, and how a new IMP-table might remedy them, citing specific examples, if possible. From swillner at nhcc.net Wed Jul 9 18:23:09 2014 From: swillner at nhcc.net (Steve Willner) Date: Wed, 09 Jul 2014 12:23:09 -0400 Subject: [BLML] IMTables and IMPing multi-pair teams In-Reply-To: <002001cf99ca$4ffda460$eff8ed20$@optusnet.com.au> References: <4709DE22F2E74C58957F9789CCD3EFCD@magnifique><7E660BC78E1F4B25BC88F314C5FADE17@Pathfinder> <53B64741.8020702@skynet.be> <56CC7E49EB54411B87BDE163B4666A10@Pathfinder> <53BA3FDE.8020804@skynet.be> <002001cf99ca$4ffda460$eff8ed20$@optusnet.com.au> Message-ID: <53BD6C6D.7040905@nhcc.net> On 2014-07-07 6:00 AM, Tony Musgrove wrote: > Suppose you are playing in a "butler swiss pairs", > as they do in Australia quite often. It's probably futile to ask, but why is Butler popular? The flaws have been discussed at length and should be well known by now. > I always use Herman's Bastille method of linearizing > the imp scales Bastille doesn't linearize. It just interpolates across gaps in the IMP table. That removes the problem that an improved score on one board may give a worse placing in the event(!), but it doesn't eliminate the other problems of Butler (randomness depending on seat assignment, ignoring some valid results, and overemphasis on boards with bimodal scores). > so that each small difference in > scores leads to a different IMP score in cross imps. Cross-IMPs is not Butler. Cross-IMPs removes all the problems with Butler, and you don't need Bastille because each score (unlike the Butler datum) is a normal bridge score. > This method gets as close to zero as the average > overall score in Howell type movements, If there are no artificial scores, cross-IMPs forces the average EW and the average NS score on every board to be zero. That eliminates randomness based on seat assignment. (The other problems with Butler also go away.) > People like to have a published "datum" with which > to compare their actual scores (it is said), I don't know who these people are, but perhaps they exist. If players at a particular club prefer Butler over cross-IMPs, I have no objection to giving them what they want. Butler should never be used in a serious event, though. As Tony wrote, you can get an implied datum from the average IMPs you are plus or minus on each board. (The implied datum will differ for different pairs, but that's no problem.) From richard.hills at immi.gov.au Thu Jul 10 04:50:24 2014 From: richard.hills at immi.gov.au (Richard James HILLS) Date: Thu, 10 Jul 2014 02:50:24 +0000 Subject: [BLML] QWERTY (was IMP tables) [SEC=UNOFFICIAL] Message-ID: <240635A98F59F24AAD8510EB05121DC33955AC68@SDCWPIPEX02.IMMI.LOCAL> UNOFFICIAL Bill Hood: >..... >Yes, I wanted to start a debate but not with anyone who advocates >in favour of deliberately randomising the scoring and seeding just >so that the game generates some ?fun? results. First, it is not >necessary as, while it is a game of skill and judgement, there will >always be situations where the ?right? thing is wrong and the >?wrong? thing is right. George Orwell (in his essay Politics and the English language): I am going to translate a passage of good English into modern English of the worst sort. Here is a well-known verse from Ecclesiastes: I returned and saw under the sun, that the race is not to the swift, nor the battle to the strong, neither yet bread to the wise, nor yet riches to men of understanding, nor yet favour to men of skill; but time and chance happeneth to them all. Here it is in modern English: Objective considerations of contemporary phenomena compel the conclusion that success or failure in competitive activities exhibits no tendency to be commensurate with innate capacity, but that a considerable element of the unpredictable must invariably be taken into account. Bill Hood: >Second, there is a dimension to the game which invites tactics >over strategy and various ploys and gambits may prove >effective according to the situation at the table or the state of >the match. As they say, ?stuff happens? (forgive the rhetoric) >but we cannot legislate for all obscure, unlikely and unfortunate >scores. Richard Hills: Yes, ?time and chance happeneth to them all?. A website which offered Duplicate Poker went broke, because the average poker-faced player has a superstitious belief in their own luck. Bill Hood: >However, we simply HAVE to build the scoring to reward skill >and judgement and the scoring MUST make sense in the context >of ?expert? players, otherwise the game itself makes no sense. Richard Hills: Yes, ?commensurate with innate capacity?. The average Duplicate Bridge player likes the duplicate concept because they have an equally superstitious belief in their own skill. Bill Hood: >From time?to-time, we may have to make adjustments when >clear injustices are noticed AND agreed to be undesirable (such >as the increase in the ?insult? and the cost of excessive non- >vulnerable undertricks.) >..... Richard Hills: I was once the victim of a ?clear injustice? at the conclusion of the qualifying rounds of the South-West Pacific Teams. My team tied on Victory Points with another team for the final spot in the knock-out rounds.. The regulations mandated a four- board playoff. My team foolishly won the playoff. Because of the extra time taken to determine the result, I did not get a dinner break before the round-of-16 started, as I was required to attend a captain?s meeting. But this ?clear injustice? is most unlikely to recur. The ABF has adopted the new WBF Victory Point scale, which goes to two decimal places, thus vastly reducing the possibility of an exact tie on VPs between two teams. Best wishes, Hilda R. Lirsch hildalirsch at gmail.com UNOFFICIAL -------------------------------------------------------------------- Important Notice: If you have received this email by mistake, please advise the sender and delete the message and attachments immediately. This email, including attachments, may contain confidential, sensitive, legally privileged and/or copyright information. Any review, retransmission, dissemination or other use of this information by persons or entities other than the intended recipient is prohibited. DIBP respects your privacy and has obligations under the Privacy Act 1988. The official departmental privacy policy can be viewed on the department's website at www.immi.gov.au. See: http://www.immi.gov.au/functional/privacy.htm --------------------------------------------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.rtflb.org/pipermail/blml/attachments/20140710/d6ced134/attachment-0001.html From moysian at talktalk.net Thu Jul 10 15:02:46 2014 From: moysian at talktalk.net (Bill Hood) Date: Thu, 10 Jul 2014 14:02:46 +0100 Subject: [BLML] QWERTY (was IMP tables) [SEC=UNOFFICIAL] In-Reply-To: <240635A98F59F24AAD8510EB05121DC33955AC68@SDCWPIPEX02.IMMI.LOCAL> References: <240635A98F59F24AAD8510EB05121DC33955AC68@SDCWPIPEX02.IMMI.LOCAL> Message-ID: <48F2F5C6A6AE48B2A74C72CB659A39AF@Pathfinder> Wow - a review under the watchful eye of George Orwell! I guess it was worth dipping my toes into the water just for that. Bill _____ From: blml-bounces at rtflb.org [mailto:blml-bounces at rtflb.org] On Behalf Of Richard James HILLS Sent: 10 July 2014 03:50 To: Bridge Laws Mailing List Cc: hildalirsch at gmail.com Subject: Re: [BLML] QWERTY (was IMP tables) [SEC=UNOFFICIAL] UNOFFICIAL Bill Hood: >..... >Yes, I wanted to start a debate but not with anyone who advocates >in favour of deliberately randomising the scoring and seeding just >so that the game generates some "fun" results. First, it is not >necessary as, while it is a game of skill and judgement, there will >always be situations where the "right" thing is wrong and the >"wrong" thing is right. George Orwell (in his essay Politics and the English language): I am going to translate a passage of good English into modern English of the worst sort. Here is a well-known verse from Ecclesiastes: I returned and saw under the sun, that the race is not to the swift, nor the battle to the strong, neither yet bread to the wise, nor yet riches to men of understanding, nor yet favour to men of skill; but time and chance happeneth to them all. Here it is in modern English: Objective considerations of contemporary phenomena compel the conclusion that success or failure in competitive activities exhibits no tendency to be commensurate with innate capacity, but that a considerable element of the unpredictable must invariably be taken into account. Bill Hood: >Second, there is a dimension to the game which invites tactics >over strategy and various ploys and gambits may prove >effective according to the situation at the table or the state of >the match. As they say, "stuff happens" (forgive the rhetoric) >but we cannot legislate for all obscure, unlikely and unfortunate >scores. Richard Hills: Yes, "time and chance happeneth to them all". A website which offered Duplicate Poker went broke, because the average poker-faced player has a superstitious belief in their own luck. Bill Hood: >However, we simply HAVE to build the scoring to reward skill >and judgement and the scoring MUST make sense in the context >of "expert" players, otherwise the game itself makes no sense. Richard Hills: Yes, "commensurate with innate capacity". The average Duplicate Bridge player likes the duplicate concept because they have an equally superstitious belief in their own skill. Bill Hood: >From time-to-time, we may have to make adjustments when >clear injustices are noticed AND agreed to be undesirable (such >as the increase in the "insult" and the cost of excessive non- >vulnerable undertricks.) >..... Richard Hills: I was once the victim of a "clear injustice" at the conclusion of the qualifying rounds of the South-West Pacific Teams. My team tied on Victory Points with another team for the final spot in the knock-out rounds.. The regulations mandated a four- board playoff. My team foolishly won the playoff. Because of the extra time taken to determine the result, I did not get a dinner break before the round-of-16 started, as I was required to attend a captain's meeting. But this "clear injustice" is most unlikely to recur. The ABF has adopted the new WBF Victory Point scale, which goes to two decimal places, thus vastly reducing the possibility of an exact tie on VPs between two teams. Best wishes, Hilda R. Lirsch hildalirsch at gmail.com UNOFFICIAL -------------------------------------------------------------------- Important Notice: If you have received this email by mistake, please advise the sender and delete the message and attachments immediately. This email, including attachments, may contain confidential, sensitive, legally privileged and/or copyright information. Any review, retransmission, dissemination or other use of this information by persons or entities other than the intended recipient is prohibited. DIBP respects your privacy and has obligations under the Privacy Act 1988. The official departmental privacy policy can be viewed on the department's website at www.immi.gov.au. See: http://www.immi.gov.au/functional/privacy.htm --------------------------------------------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.rtflb.org/pipermail/blml/attachments/20140710/c1bd4e16/attachment-0001.html From moysian at talktalk.net Thu Jul 10 16:54:05 2014 From: moysian at talktalk.net (Bill Hood) Date: Thu, 10 Jul 2014 15:54:05 +0100 Subject: [BLML] QWERTY (was IMP tables) [SEC=UNOFFICIAL] In-Reply-To: <31B9FA71FB61431F921946C6C9937C7B@G3> References: <240635A98F59F24AAD8510EB05121DC331AEA739@SDCWPIPEX02.IMMI.LOCAL><0AA261A403CA43ADB08F5C0A9EB74A97@Pathfinder> <31B9FA71FB61431F921946C6C9937C7B@G3> Message-ID: [Nige1] I agree that the Bridge scoring should reward skill. Please tell us, Bill, what current IMP-table flaws concern you, and how a new IMP-table might remedy them, citing specific examples, if possible. Hi Nigel, I set out most of my concerns in an earlier post and they have been echoed by some of the responses and associated anecdotes. By way of a recap, I would re-iterate that my main concerns are: 1. It is a fundamental requirement of any scoring system that not only can it can be used to derive an understanding of how best to play the game to maximise our chances of coming out ahead but also that the derived strategy is sensible and consistent. I believe that the IMP table and the IMPing process fail both these tests. [If you ask different groups of bridge players what they think the correct bidding strategy is, you will get different answers.] 2. It is simply not good enough just to proceed on the basis that, within the margins of the interval ranges assigned, bigger swings will be rewarded by more IMPs. Nor do I believe that the IMP table compresses the aggregate scores to a level that adequately prevents the distortion of results from high-scoring contracts. [This is a difficult area and, frankly, I am not yet sure how much better it could be. Certainly, I have concerns that the objective may not be practically achievable with a 24-point scale or as an overlay on regulation scoring. I wanted to initiate a debate to test my concerns and establish the views of others.] Specifics On a positive note, the IMP table generates a differential between the threshold values for bidding V. and NV games. I think that this is both sensible and necessary and indeed it seems natural and correct that we should stretch for the extra points available when vulnerable. However i. Should the differential be wider (say 10%) ii. Should there not be a lower limit for such values (perhaps 40%) By the way, does anyone really think that the scoring should be arranged to reward bidding vulnerable games which are expected to fail 2/3rds of the time (as in multi-pair Butler IMP scoring)? iii. Somewhat perversely, there is no such differential for slams and this is inconsistent and, frankly, just silly. [If it makes sense to have a differential for V and NV games, why should it not make sense for slams? If we don't have to risk any more or do anything different for the additional (vulnerable) gains then what is the point of varying the rewards? I doubt that having to pay more attention to the play when vulnerable can be considered as a desirable or winning strategy.] Of course, it is the underlying scores which determine the parity and 50% threshold values of vulnerable and non-vulnerable small slams and the IMP table is unable to impose a differential - surely a critical weakness of the IMPing process. iv. Technically, there is a tiny differential which favours minor suit slams but this is just an untidy irrelevance of no practical value and it could and should have been eliminated by a "makes sense" adjustment of interval boundaries around the swing values. v. Why is there no score for a 10 point swing? Is it to eliminate the difference between contract strains? If so, why does it equalise NT and majors and minors and majors but not NT and minors and why are two major overtricks worth 2 IMPs while 2 minor overtricks are worth only 1? vi. The management of slams and, in particular, grand slams is unsatisfactory. a) At the top end of the scale, a slam swing (vs. a game) generates a 10/12 gain but the gain only increases to 14/17 if the other table doesn't even bid game. b) The reward for bidding a grand (even at the appropriate threshold value) is not worth the additional risk if you suspect that the opponents may not even bid the small slam. I will not elaborate on the "local" issues of the EBU suggestion that additional IMP tables for teams of 8, 12 and 16 can be generated by multiplying the end-points of the ranges of the teams-of-four IMP table by, respectively, v 2 v3 and v4. This effectively randomises the threshold values (and varies the vulnerable/ non-vulnerable differential) at which games should be bid according to happenstance and the number of pairs in the team. However, such excesses perhaps serve to underline the many issues associated with IMPing. As a worked example, if you adjust the bonus values slightly, you can readily generate a tidy IMP table that maintains a (close-to) 5% differential between vulnerable and non-vulnerable games and slams. This gives rise to a bidding strategy that is consistent, sensible and interesting in that it invites a measure of aggression when vulnerable. Other tables could be compiled against a consensus view of the design properties we wanted the IMP process to reflect. Bill _______________________________________________ Blml mailing list Blml at rtflb.org http://lists.rtflb.org/mailman/listinfo/blml -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.rtflb.org/pipermail/blml/attachments/20140710/5fcf25d2/attachment-0001.html From swillner at nhcc.net Thu Jul 10 17:19:52 2014 From: swillner at nhcc.net (Steve Willner) Date: Thu, 10 Jul 2014 11:19:52 -0400 Subject: [BLML] QWERTY (was IMP tables) In-Reply-To: References: <240635A98F59F24AAD8510EB05121DC331AEA739@SDCWPIPEX02.IMMI.LOCAL><0AA261A403CA43ADB08F5C0A9EB74A97@Pathfinder> <31B9FA71FB61431F921946C6C9937C7B@G3> Message-ID: <53BEAF18.1030803@nhcc.net> On 2014-07-10 10:54 AM, Bill Hood wrote: > /[If you ask different groups of bridge players what they think the > correct bidding strategy is, you will get different answers.]/ What do you mean by this? I'd have said that having different ideas of the best bidding strategy is a fundamental aspect of the game. > /By the way, does anyone really think that the scoring should be > arranged to reward bidding vulnerable games which are expected to fail > 2/3rds of the time (as in multi-pair Butler IMP scoring)?/ I don't follow your arithmetic, but in any case Butler is bad for so many reasons that I wouldn't use it as an argument for anything. [Independent of this thread, could people please avoid using HTML for BLML messages? Plain text is much easier to read.] From jimfox00 at cox.net Fri Jul 11 00:30:55 2014 From: jimfox00 at cox.net (Jim Fox) Date: Thu, 10 Jul 2014 18:30:55 -0400 Subject: [BLML] QWERTY (was IMP tables) [SEC=UNOFFICIAL] In-Reply-To: References: <240635A98F59F24AAD8510EB05121DC331AEA739@SDCWPIPEX02.IMMI.LOCAL><0AA261A403CA43ADB08F5C0A9EB74A97@Pathfinder> <31B9FA71FB61431F921946C6C9937C7B@G3> Message-ID: Comment below in snipped text. From: blml-bounces at rtflb.org [mailto:blml-bounces at rtflb.org] On Behalf Of Bill Hood Sent: 07/10/2014 10:54 AM To: 'Bridge Laws Mailing List' Subject: Re: [BLML] QWERTY (was IMP tables) [SEC=UNOFFICIAL] Bill: [If you ask different groups of bridge players what they think the correct bidding strategy is, you will get different answers.] MMbridge: And we think that is a BAD thing? Bill Mmbridge _______________________________________________ Blml mailing list Blml at rtflb.org http://lists.rtflb.org/mailman/listinfo/blml -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.rtflb.org/pipermail/blml/attachments/20140710/dcefb6a0/attachment.html From richard.hills at immi.gov.au Fri Jul 11 01:47:59 2014 From: richard.hills at immi.gov.au (Richard James HILLS) Date: Thu, 10 Jul 2014 23:47:59 +0000 Subject: [BLML] QWERTY (was IMP tables) [SEC=UNOFFICIAL] Message-ID: <240635A98F59F24AAD8510EB05121DC33955C763@SDCWPIPEX02.IMMI.LOCAL> UNOFFICIAL Bill Hood >>If you ask different groups of bridge players what they think the >>correct bidding strategy is, you will get different answers. Steve Willner: >What do you mean by this? I'd have said that having different ideas >of the best bidding strategy is a fundamental aspect of the game. Richard Hills: When world-class Aussie expert Paul Marston was editor of Australian Bridge magazine, he wrote an article titled, "Four card majors are better". In the next issue world-class Aussie expert Ron Klinger wrote an article titled, "Five card majors are better". Ron observed that one reason supporting the utility of the 5-card major method was that it was used by many partnerships throughout the world. And that opening 1H or 1S with five cards made it easy to reach a 5-3 fit. In the following issue world-class Aussie expert Richard Brightling wrote an article titled, "Six card majors are better". Richard observed that one reason supporting the utility of the 6-card major method was that it was used by zero partnerships throughout the world. And that opening 1H or 1S with six cards made it easy to reach a 6-2 fit. Best wishes, Hilda R. Lirsch hildalirsch at gmail.com UNOFFICIAL -------------------------------------------------------------------- Important Notice: If you have received this email by mistake, please advise the sender and delete the message and attachments immediately. This email, including attachments, may contain confidential, sensitive, legally privileged and/or copyright information. Any review, retransmission, dissemination or other use of this information by persons or entities other than the intended recipient is prohibited. DIBP respects your privacy and has obligations under the Privacy Act 1988. The official departmental privacy policy can be viewed on the department's website at www.immi.gov.au. See: http://www.immi.gov.au/functional/privacy.htm --------------------------------------------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.rtflb.org/pipermail/blml/attachments/20140710/f80f05c9/attachment-0001.html From richard.hills at immi.gov.au Fri Jul 11 02:28:04 2014 From: richard.hills at immi.gov.au (Richard James HILLS) Date: Fri, 11 Jul 2014 00:28:04 +0000 Subject: [BLML] QWERTY (was IMP tables) [SEC=UNOFFICIAL] Message-ID: <240635A98F59F24AAD8510EB05121DC33955C7E2@SDCWPIPEX02.IMMI.LOCAL> UNOFFICIAL Bill Hood: >It is a fundamental requirement of any scoring system that not >only can it can be used to derive an understanding of how best >to play the game to maximise our chances of coming out ahead >but also that the derived strategy is sensible and consistent. Richard Hills: If you are declarer in an obvious 4S contract at imps, your goal is to ensure the ten tricks required (that is, taking a safety play is a "sensible and consistent" strategy). If you are declarer in an obvious 4S contract at matchpoints, your goal is to take the optimum number of tricks (that is, eschewing a safety play is a "sensible and consistent" strategy). But the Conditions of Contest for an important English teams championship, the Pachabo Cup, state that 50% of each board is scored at imps, and 50% of each board is scored at matchpoints (point-a-board / board-a-match). Jeff Rubens has observed that declarer cannot formulate a "sensible and consistent" strategy under these Conditions of Contest. Best wishes, Hilda R. Lirsch hildalirsch at gmail.com UNOFFICIAL -------------------------------------------------------------------- Important Notice: If you have received this email by mistake, please advise the sender and delete the message and attachments immediately. This email, including attachments, may contain confidential, sensitive, legally privileged and/or copyright information. Any review, retransmission, dissemination or other use of this information by persons or entities other than the intended recipient is prohibited. DIBP respects your privacy and has obligations under the Privacy Act 1988. The official departmental privacy policy can be viewed on the department's website at www.immi.gov.au. See: http://www.immi.gov.au/functional/privacy.htm --------------------------------------------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.rtflb.org/pipermail/blml/attachments/20140711/ec42fb6f/attachment.html From hermandw at skynet.be Fri Jul 11 09:04:31 2014 From: hermandw at skynet.be (Herman De Wael) Date: Fri, 11 Jul 2014 09:04:31 +0200 Subject: [BLML] QWERTY (was IMP tables) [SEC=UNOFFICIAL] In-Reply-To: <240635A98F59F24AAD8510EB05121DC33955C7E2@SDCWPIPEX02.IMMI.LOCAL> References: <240635A98F59F24AAD8510EB05121DC33955C7E2@SDCWPIPEX02.IMMI.LOCAL> Message-ID: <53BF8C7F.2070609@skynet.be> Richard James HILLS schreef: > UNOFFICIAL > If you are declarer in an obvious 4S contract at imps, your goal > is to ensure the ten tricks required (that is, taking a safety play is > a ?sensible and consistent? strategy). > If you are declarer in an obvious 4S contract at matchpoints, your > goal is to take the optimum number of tricks (that is, eschewing a > safety play is a ?sensible and consistent? strategy). > But the Conditions of Contest for an important English teams > championship, the Pachabo Cup, state that 50% of each board is > scored at imps, and 50% of each board is scored at matchpoints > (point-a-board / board-a-match). Jeff Rubens has observed that > declarer /cannot/ formulate a ?sensible and consistent? strategy > under these Conditions of Contest. > Best wishes, > Hilda R. Lirsch > _hildalirsch at gmail.com_ > UNOFFICIAL > Jeff Rubens is wrong. (how i love to be able to say that, for once). There will, dependent on the actual method of calculation, be a percentage where the two strategies will be equal. If the chance of success of one of the alternative lines is higher than that percentage, that line ought to be chosen. Therefore, a player is able to formulate a sensible strategy. Moreover, that strategy will be the same for all players. But better mathematicians will be better able to calculate the border percentage. And better bridge players will be better able to calculate the probability of success of a particular line. Allow me to explain: if the choice is between a drop and a finesse, with the drop ensuring 10 tricks and the finesse yielding 9 or 11 (in a contract at the 4-level), then: At teams: the win for 11 tricks is +1 IMPs, and the loss for 9 is -10 IMPs. So you need around 95% chance of success to make the finesse worth while. A better mathematician will be able to refine the odds. They will calculate that the actual cut-off is 12/13 or 92.3%. At pairs, the win and loss are equal, so the cut-off is at 50%. A better bridge player will see that the chance of a successful finesse are above or below 50%, and they will act accordingly. At Patton, the cut-off will be somewhere between the two (depending on the actual method used). The better mathematician will refine this calculation, arriving at 72,3% (just a guess, actually). The better bridge player will calculate the odds of the alternative lines. In both cases, the better analyst will have the more successful outcome. But it is always a game of skill. Herman. From svenpran at online.no Fri Jul 11 11:18:44 2014 From: svenpran at online.no (Sven Pran) Date: Fri, 11 Jul 2014 11:18:44 +0200 Subject: [BLML] QWERTY (was IMP tables) [SEC=UNOFFICIAL] In-Reply-To: <53BF8C7F.2070609@skynet.be> References: <240635A98F59F24AAD8510EB05121DC33955C7E2@SDCWPIPEX02.IMMI.LOCAL> <53BF8C7F.2070609@skynet.be> Message-ID: <001901cf9ce9$1cf931c0$56eb9540$@online.no> > Herman De Wael > There will, dependent on the actual method of calculation, be a percentage > where the two strategies will be equal. If the chance of success of one of the > alternative lines is higher than that percentage, that line ought to be chosen. > Therefore, a player is able to formulate a sensible strategy. Moreover, that > strategy will be the same for all players. But better mathematicians will be > better able to calculate the border percentage. And better bridge players will > be better able to calculate the probability of success of a particular line. > > Allow me to explain: if the choice is between a drop and a finesse, with the drop > ensuring 10 tricks and the finesse yielding 9 or 11 (in a contract at the 4-level), > then: > At teams: the win for 11 tricks is +1 IMPs, and the loss for 9 is -10 IMPs. So you > need around 95% chance of success to make the finesse worth while. A better > mathematician will be able to refine the odds. They will calculate that the > actual cut-off is 12/13 or 92.3%. > At pairs, the win and loss are equal, so the cut-off is at 50%. A better bridge > player will see that the chance of a successful finesse are above or below 50%, > and they will act accordingly. [Sven Pran] Very well written, but there is a catch here (at pairs). If you estimate that most pairs will end in 3NT with 10 tricks (no less, no more) then your win and loss in a 4S contract are no longer equal. In fact the cutoff will approach 100% in favour of the finesse because there will be little difference in matchpoints between 420 and -50, you will need 450 to get points on the board. Such considerations are completely irrelevant at teams. From moysian at talktalk.net Fri Jul 11 13:17:47 2014 From: moysian at talktalk.net (Bill Hood) Date: Fri, 11 Jul 2014 12:17:47 +0100 Subject: [BLML] QWERTY (was IMP tables) [SEC=UNOFFICIAL] In-Reply-To: <240635A98F59F24AAD8510EB05121DC33955C7E2@SDCWPIPEX02.IMMI.LOCAL> References: <240635A98F59F24AAD8510EB05121DC33955C7E2@SDCWPIPEX02.IMMI.LOCAL> Message-ID: <4F2D3C9ED0334331A2A8DC66AB0AC99C@Pathfinder> Hi Richard, For the avoidance of doubt, I am at no point referring to systems of bidding or the rationale of specific card plays. Rather, I am concerned about which contracts are best rewarded by the scoring and under which conditions. Perhaps the key decisions are "what is the threshold probability above which it is correct (in terms of the rewards of the scoring system) to bid game or slam at the prevailing vulnerability". Certainly, they are the easiest to reflect on but you need also to consider when it is right to compete, take a penalty or sacrifice. How you choose to bid is an almost entirely different question though, perhaps, if you perceived that the scoring disproportionately rewarded slams, you might develop that aspect of your methods and be a bit more circumspect about your management of part scores. In a similar way, decisions about safety plays are important but secondary and your decisions must be pragmatic and influenced by the (relative) rewards of the scoring. I suspect that most (average plus) players don't really have a well developed bidding strategy and play along lines of pushing for bonuses regardless of vulnerability and doubling somewhat indiscriminately on a macho basis even when it is not correct to do so (in terms of the rewards of the scoring). I hope this clarifies my point. Bill P.S. Your point about the scoring in the Pachabo is interesting and the scoring is indeed idiosyncratic. Nobody quite understands how it works though it gives a perception of somehow being fair. For such a well developed game, it is more than curious that all our methods of scoring seem to be flawed in some way and it is almost as if the different methods have been devised to compensate for each other. _____ From: blml-bounces at rtflb.org [mailto:blml-bounces at rtflb.org] On Behalf Of Richard James HILLS Sent: 11 July 2014 01:28 To: Bridge Laws Mailing List Cc: hildalirsch at gmail.com Subject: Re: [BLML] QWERTY (was IMP tables) [SEC=UNOFFICIAL] UNOFFICIAL Bill Hood: >It is a fundamental requirement of any scoring system that not >only can it can be used to derive an understanding of how best >to play the game to maximise our chances of coming out ahead >but also that the derived strategy is sensible and consistent. Richard Hills: If you are declarer in an obvious 4S contract at imps, your goal is to ensure the ten tricks required (that is, taking a safety play is a "sensible and consistent" strategy). If you are declarer in an obvious 4S contract at matchpoints, your goal is to take the optimum number of tricks (that is, eschewing a safety play is a "sensible and consistent" strategy). But the Conditions of Contest for an important English teams championship, the Pachabo Cup, state that 50% of each board is scored at imps, and 50% of each board is scored at matchpoints (point-a-board / board-a-match). Jeff Rubens has observed that declarer cannot formulate a "sensible and consistent" strategy under these Conditions of Contest. Best wishes, Hilda R. Lirsch hildalirsch at gmail.com UNOFFICIAL -------------------------------------------------------------------- Important Notice: If you have received this email by mistake, please advise the sender and delete the message and attachments immediately. This email, including attachments, may contain confidential, sensitive, legally privileged and/or copyright information. Any review, retransmission, dissemination or other use of this information by persons or entities other than the intended recipient is prohibited. DIBP respects your privacy and has obligations under the Privacy Act 1988. The official departmental privacy policy can be viewed on the department's website at www.immi.gov.au. See: http://www.immi.gov.au/functional/privacy.htm --------------------------------------------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.rtflb.org/pipermail/blml/attachments/20140711/0dfe2a86/attachment.html From moysian at talktalk.net Sun Jul 13 12:40:49 2014 From: moysian at talktalk.net (Bill Hood) Date: Sun, 13 Jul 2014 11:40:49 +0100 Subject: [BLML] QWERTY (was IMP tables) [SEC=UNOFFICIAL] In-Reply-To: References: <240635A98F59F24AAD8510EB05121DC331AEA739@SDCWPIPEX02.IMMI.LOCAL><0AA261A403CA43ADB08F5C0A9EB74A97@Pathfinder> <31B9FA71FB61431F921946C6C9937C7B@G3> Message-ID: FAO MMbridge Whether or not you think it is a good idea that players do not know what the correct bidding strategy is depends on whether or not you think the game is intrinsically a serious test of skill and judgement. [Please also refer to my earlier post clarifying that my reference to bidding strategy (and tactics) is not about methods/ conventions of bidding - conventions and treatments such as the 4/5 card major openings may be employed in support of a given strategy but they are not in themselves a strategy.] You can obviously choose to do as you see fit and, since your stated objective is to get random outcomes, you can be take comfort from the fact that, if different pairs do different things, they will get different rewards (some fair/ fortunate and some less so). In that sense, everyone can (sometimes) win but they can't all be said to be pursuing winning strategies and, given near equivalent skill levels, the most consistent winners will be those whose strategy has been developed to take advantage of the way the scoring is done. LIFE - 'tis a puzzlement! Bill _____ From: blml-bounces at rtflb.org [mailto:blml-bounces at rtflb.org] On Behalf Of Jim Fox Sent: 10 July 2014 23:31 To: 'Bridge Laws Mailing List' Subject: Re: [BLML] QWERTY (was IMP tables) [SEC=UNOFFICIAL] Comment below in snipped text. From: blml-bounces at rtflb.org [mailto:blml-bounces at rtflb.org] On Behalf Of Bill Hood Sent: 07/10/2014 10:54 AM To: 'Bridge Laws Mailing List' Subject: Re: [BLML] QWERTY (was IMP tables) [SEC=UNOFFICIAL] Bill: [If you ask different groups of bridge players what they think the correct bidding strategy is, you will get different answers.] MMbridge: And we think that is a BAD thing? Bill Mmbridge _______________________________________________ Blml mailing list Blml at rtflb.org http://lists.rtflb.org/mailman/listinfo/blml -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.rtflb.org/pipermail/blml/attachments/20140713/f768ef57/attachment.html From jimfox00 at cox.net Sun Jul 13 18:36:48 2014 From: jimfox00 at cox.net (Jim Fox) Date: Sun, 13 Jul 2014 12:36:48 -0400 Subject: [BLML] QWERTY (was IMP tables) [SEC=UNOFFICIAL] In-Reply-To: References: <240635A98F59F24AAD8510EB05121DC331AEA739@SDCWPIPEX02.IMMI.LOCAL><0AA261A403CA43ADB08F5C0A9EB74A97@Pathfinder> <31B9FA71FB61431F921946C6C9937C7B@G3> Message-ID: You are missing the point. Bridge is a probabilistic game and that is a good thing. Most interesting games are. Any bidding system must adjust to take into account what the opponents are doing in their bidding. Thus there can never be a "correct bidding strategy" unless you somehow know exactly how your opponents will bid and how they will adjust to every situation and your actions. Let's suppose there were a few players, pairs or teams who knew "the correct playing strategy" and everybody else didn't. Those lucky few would then win every event. Do you think there is a "correct playing strategy" for golfers or is it a combination of how one player's strategy meshes with many random, unpredictable events on a golf course. How important is the strategy compared with execution of physical actions and ability to execute the strategy. My stated objective is not "to get random outcomes". My unstated objective is to have the game be as interesting to as many people as possible. I don't need bridge in my life to get a "serious test of skill and judgment". My work provides more than enough of that. For me, the purpose of playing bridge is to have an interesting way to exercise my brain that I can do with people I love or like. Nothing more. Winning top bridge tournaments (like golf tournaments) is mostly a "battle of numbers". Everybody is good. You execute strategy to give yourself the best chance to win. "Actually winning" is very much a random event. The fact of the matter is that the more individuals you allow to achieve that random event, the more popular the game will be. Scoring idiosyncrasies adds slightly to that number. Checkers and tic-tac-toe are deterministic games. Most people don't find them particularly interesting. Mmbridge From: blml-bounces at rtflb.org [mailto:blml-bounces at rtflb.org] On Behalf Of Bill Hood Sent: 07/13/2014 6:41 AM To: 'Bridge Laws Mailing List' Subject: Re: [BLML] QWERTY (was IMP tables) [SEC=UNOFFICIAL] FAO MMbridge Whether or not you think it is a good idea that players do not know what the correct bidding strategy is depends on whether or not you think the game is intrinsically a serious test of skill and judgement. [Please also refer to my earlier post clarifying that my reference to bidding strategy (and tactics) is not about methods/ conventions of bidding - conventions and treatments such as the 4/5 card major openings may be employed in support of a given strategy but they are not in themselves a strategy.] You can obviously choose to do as you see fit and, since your stated objective is to get random outcomes, you can be take comfort from the fact that, if different pairs do different things, they will get different rewards (some fair/ fortunate and some less so). In that sense, everyone can (sometimes) win but they can't all be said to be pursuing winning strategies and, given near equivalent skill levels, the most consistent winners will be those whose strategy has been developed to take advantage of the way the scoring is done. LIFE - 'tis a puzzlement! Bill _____ From: blml-bounces at rtflb.org [mailto:blml-bounces at rtflb.org] On Behalf Of Jim Fox Sent: 10 July 2014 23:31 To: 'Bridge Laws Mailing List' Subject: Re: [BLML] QWERTY (was IMP tables) [SEC=UNOFFICIAL] Comment below in snipped text. From: blml-bounces at rtflb.org [mailto:blml-bounces at rtflb.org] On Behalf Of Bill Hood Sent: 07/10/2014 10:54 AM To: 'Bridge Laws Mailing List' Subject: Re: [BLML] QWERTY (was IMP tables) [SEC=UNOFFICIAL] Bill: [If you ask different groups of bridge players what they think the correct bidding strategy is, you will get different answers.] MMbridge: And we think that is a BAD thing? Bill Mmbridge _______________________________________________ Blml mailing list Blml at rtflb.org http://lists.rtflb.org/mailman/listinfo/blml -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.rtflb.org/pipermail/blml/attachments/20140713/f5df00bf/attachment-0001.html From jfusselman at gmail.com Sun Jul 13 19:40:20 2014 From: jfusselman at gmail.com (Jerry Fusselman) Date: Sun, 13 Jul 2014 12:40:20 -0500 Subject: [BLML] QWERTY (was IMP tables) [SEC=UNOFFICIAL] In-Reply-To: References: <240635A98F59F24AAD8510EB05121DC331AEA739@SDCWPIPEX02.IMMI.LOCAL> <0AA261A403CA43ADB08F5C0A9EB74A97@Pathfinder> <31B9FA71FB61431F921946C6C9937C7B@G3> Message-ID: On Sun, Jul 13, 2014 at 11:36 AM, Jim Fox wrote: > You are missing the point. > Well, most of your points are questionable. > > Bridge is a probabilistic game and that is a good thing. Most interesting > games are. > Go and chess are very interesting games. > Any bidding system must adjust to take into account what the > opponents are doing in their bidding. Thus there can never be a ?correct > bidding strategy? unless you somehow know exactly how your opponents will > bid and how they will adjust to every situation and your actions. > I doubt that anyone thinks that there exists some perfect bidding strategy. We only try to do our best for most circumstances given limitations on memory. You can never completely overcome the limitations on memory. > > Let?s suppose there were a few players, pairs or teams who knew ?the correct > playing strategy? and everybody else didn?t. Those lucky few would then win > every event. > Nonsense. For some hands, ultra-strong jump-shifts are clearly wonderful. For some situations, fourth best is demonstrably superior to anything else. No set of methods dominates all other methods on 100% of the hands. Further, methods do not guarantee success if execution is lacking. > > My stated objective is not ?to get random outcomes?. My unstated objective > is to have the game be as interesting to as many people as possible. I > don?t need bridge in my life to get a ?serious test of skill and judgment?. Speak for yourself. A serious test of skill and judgement is exactly what I am after. > My work provides more than enough of that. For me, the purpose of playing > bridge is to have an interesting way to exercise my brain that I can do with > people I love or like. Nothing more. > Maybe rubber bridge is your game. > > Winning top bridge tournaments (like golf tournaments) is mostly a ?battle > of numbers?. Everybody is good. You execute strategy to give yourself the > best chance to win. ?Actually winning? is very much a random event. The > fact of the matter is that the more individuals you allow to achieve that > random event, the more popular the game will be. Also wrong. If the rules were changed to the extent that pros were in the bottom half half of the time, such as if results were 100% chance, I am sure that bridge would lose all of its popularity. Winning would be meaningless, essentially a lottery. Why bother holding cards in your hand if your efforts amount to nothing? > Scoring idiosyncrasies > adds slightly to that number. > "It's not a bug, it's a feature!" Jerry Fusselman From jimfox00 at cox.net Sun Jul 13 23:34:14 2014 From: jimfox00 at cox.net (Jim Fox) Date: Sun, 13 Jul 2014 17:34:14 -0400 Subject: [BLML] IMP and VP Tables and Bridge Scoring (was RE: QWERTY (was IMP tables) [SEC=UNOFFICIAL]) Message-ID: My comments embedded below. -----Original Message----- From: blml-bounces at rtflb.org [mailto:blml-bounces at rtflb.org] On Behalf Of Jerry Fusselman Sent: 07/13/2014 1:40 PM To: Bridge Laws Mailing List Subject: Re: [BLML] QWERTY (was IMP tables) [SEC=UNOFFICIAL] On Sun, Jul 13, 2014 at 11:36 AM, Jim Fox wrote: > You are missing the point. > Well, most of your points are questionable. > > Bridge is a probabilistic game and that is a good thing. Most interesting > games are. > Go and chess are very interesting games. Mmbridge: Although I believe that Go and chess are also probabilistic games, I won't argue the point other than to say that when you make a move in those games, there is some finite probability your opponent won't find the best move in response. One would probably try to estimate that probability when choosing which of several moves to make. More to the point, I guess I cannot tell whether or not you differ with "most". > Any bidding system must adjust to take into account what the > opponents are doing in their bidding. Thus there can never be a ?correct > bidding strategy? unless you somehow know exactly how your opponents will > bid and how they will adjust to every situation and your actions. > I doubt that anyone thinks that there exists some perfect bidding strategy. We only try to do our best for most circumstances given limitations on memory. You can never completely overcome the limitations on memory. Mmbridge: IAW, you agree. > > Let?s suppose there were a few players, pairs or teams who knew ?the correct > playing strategy? and everybody else didn?t. Those lucky few would then win > every event. > Nonsense. For some hands, ultra-strong jump-shifts are clearly wonderful. For some situations, fourth best is demonstrably superior to anything else. No set of methods dominates all other methods on 100% of the hands. Further, methods do not guarantee success if execution is lacking. Mmbridge: IAW, you agree. And I speak to the point of execution later (which you don?t excerpt). > > My stated objective is not ?to get random outcomes?. My unstated objective > is to have the game be as interesting to as many people as possible. I > don?t need bridge in my life to get a ?serious test of skill and judgment?. Speak for yourself. A serious test of skill and judgement is exactly what I am after. Mmbridge: Of course, I speak for myself. And I realize that others connect Duplicate bridge with their competitive and esteem issues. > My work provides more than enough of that. For me, the purpose of playing > bridge is to have an interesting way to exercise my brain that I can do with > people I love or like. Nothing more. > Maybe rubber bridge is your game. Mmbridge: I enjoy rubber bridge AND duplicate bridge. I don't have to love or like my opponents, (LOL) only my partner or teammates. I will not play in a partnership or team with people I don't like away from bridge. I'm pretty sure that many do. > > Winning top bridge tournaments (like golf tournaments) is mostly a ?battle > of numbers?. Everybody is good. You execute strategy to give yourself the > best chance to win. ?Actually winning? is very much a random event. The > fact of the matter is that the more individuals you allow to achieve that > random event, the more popular the game will be. Also wrong. If the rules were changed to the extent that pros were in the bottom half half of the time, such as if results were 100% chance, I am sure that bridge would lose all of its popularity. Winning would be meaningless, essentially a lottery. Why bother holding cards in your hand if your efforts amount to nothing? Mmbridge: I believe it's obvious I am not talking about causing pros to be pushed to the bottom half. What I am advocating is the situation we have right now. If you look at the results of most National events you will find pros all through the rankings from top to bottom alternating with non-pros. Sometimes non-pros even WIN!!! > Scoring idiosyncrasies > adds slightly to that number. > "It's not a bug, it's a feature!" Mmbridge: This "straw man" won't float. Jerry Fusselman _______________________________________________ Blml mailing list Blml at rtflb.org http://lists.rtflb.org/mailman/listinfo/blml From jfusselman at gmail.com Mon Jul 14 01:58:07 2014 From: jfusselman at gmail.com (Jerry Fusselman) Date: Sun, 13 Jul 2014 18:58:07 -0500 Subject: [BLML] IMP and VP Tables and Bridge Scoring (was RE: QWERTY (was IMP tables) [SEC=UNOFFICIAL]) In-Reply-To: References: Message-ID: Jim, your two statements that I most want to debate are these: 1. "Let?s suppose there were a few players, pairs or teams who knew `the correct playing strategy' and everybody else didn?t. Those lucky few would then win every event." 2. "Winning top bridge tournaments (like golf tournaments) is mostly a `battle of numbers'. [T]he more individuals you allow to achieve that random event, the more popular the game will be." In brief, the answer to 1 is no, they would not. The answer to 2 is that most players want less randomness in duplicate bridge, not more. On Sun, Jul 13, 2014 at 4:34 PM, Jim Fox wrote: > My comments embedded below. > > -----Original Message----- > From: blml-bounces at rtflb.org [mailto:blml-bounces at rtflb.org] On Behalf Of Jerry Fusselman > Sent: 07/13/2014 1:40 PM > To: Bridge Laws Mailing List > Subject: Re: [BLML] QWERTY (was IMP tables) [SEC=UNOFFICIAL] > > On Sun, Jul 13, 2014 at 11:36 AM, Jim Fox wrote: >> You are missing the point. >> > > Well, most of your points are questionable. > >> >> Bridge is a probabilistic game and that is a good thing. Most interesting >> games are. >> > > Go and chess are very interesting games. > > Mmbridge: Although I believe that Go and chess are also probabilistic games, I won't argue the point other than to say that when you make a move in those games, there is some finite probability your opponent won't find the best move in response. One would probably try to estimate that probability when choosing which of several moves to make. More to the point, I guess I cannot tell whether or not you differ with "most". This is revisionist history. In this thread, you wrote that "Checkers and tic-tac-toe are deterministic games." In those games too, your opponent's reply to your move is unknowable. You can't logically claim that go and chess are probabilistic immediately after claiming the checkers is not. Besides, "probabilistic" would lose all meaning if any game in which your opponent has choices is deemed probabilistic. > >> Any bidding system must adjust to take into account what the >> opponents are doing in their bidding. Thus there can never be a ?correct >> bidding strategy? unless you somehow know exactly how your opponents will >> bid and how they will adjust to every situation and your actions. >> > > I doubt that anyone thinks that there exists some perfect bidding > strategy. We only try to do our best for most circumstances given > limitations on memory. You can never completely overcome the > limitations on memory. > > Mmbridge: IAW, you agree. No, I rejected your straw man and reach a similar conclusion by completely different means: Limitations on memory, which you forgot to mention. > >> >> Let?s suppose there were a few players, pairs or teams who knew ?the correct >> playing strategy? and everybody else didn?t. Those lucky few would then win >> every event. >> > > Nonsense. For some hands, ultra-strong jump-shifts are clearly > wonderful. For some situations, fourth best is demonstrably superior > to anything else. No set of methods dominates all other methods on > 100% of the hands. > > Further, methods do not guarantee success if execution is lacking. > > Mmbridge: IAW, you agree. And I speak to the point of execution later (which you don?t excerpt). No, I don't agree that "the correct playing strategy" "would win every event." > >> >> My stated objective is not ?to get random outcomes?. My unstated objective >> is to have the game be as interesting to as many people as possible. I >> don?t need bridge in my life to get a ?serious test of skill and judgment?. > > Speak for yourself. A serious test of skill and judgement is exactly > what I am after. > > Mmbridge: Of course, I speak for myself. Also revisionist history. You wrote, "Winning top bridge tournaments (like golf tournaments) is mostly a `battle of numbers'. [T]he more individuals you allow to achieve that random event, the more popular the game will be." You're talking about all players, not just yourself. > And I realize that others connect Duplicate bridge with their competitive and esteem issues. > That's a delightful denigration of competition. >> >> Winning top bridge tournaments (like golf tournaments) is mostly a ?battle >> of numbers?. Everybody is good. You execute strategy to give yourself the >> best chance to win. ?Actually winning? is very much a random event. The >> fact of the matter is that the more individuals you allow to achieve that >> random event, the more popular the game will be. > > Also wrong. If the rules were changed to the extent that pros were in > the bottom half half of the time, such as if results were 100% chance, > I am sure that bridge would lose all of its popularity. Winning would > be meaningless, essentially a lottery. Why bother holding cards in > your hand if your efforts amount to nothing? > > Mmbridge: I believe it's obvious I am not talking about causing pros to be pushed to the bottom half. No, its not. You wrote "The fact of the matter is that the more individuals you allow to achieve that random event, the more popular the game will be." Make it fully random to the extent that pros are in the bottom half half of the time, and you have achieved exactly what you say will make the game the most popular. Exactly. But it would not make the game more popular. It would kill it. > >> Scoring idiosyncrasies >> adds slightly to that number. >> > > "It's not a bug, it's a feature!" > > Mmbridge: This "straw man" won't float. > Not a straw man---it's one of my main points. People don't really want any extra randomness in duplicate bridge. Jerry Fusselman From richard.hills at immi.gov.au Mon Jul 14 03:06:58 2014 From: richard.hills at immi.gov.au (Richard James HILLS) Date: Mon, 14 Jul 2014 01:06:58 +0000 Subject: [BLML] QWERTY (was IMP tables) [SEC=UNOFFICIAL] Message-ID: <240635A98F59F24AAD8510EB05121DC33955E6DA@SDCWPIPEX02.IMMI.LOCAL> UNOFFICIAL Bill Hood: >..... >depends on whether or not you think the game is intrinsically >a serious test of skill and judgement. >..... Richard Hills: Yes and No. One can test one's skill and judgement within a framework of rules that rewards skill and judgement BUT this does not prevent one from indulging in Pythonesque humour. An example is the Congratulatory Jack convention. Pard has found an excellent defensive play to avoid a bottom, so later in the defence you discard a surplus jack, which has the meaning, "Congratulations on your fine defence, pard." A similar convention (which is included on the Ali-Hills system card) is Cooper Echoes. When Hashmat is dummy he may be worried that our current contract will be a big disaster, especially if it is doubled. By playing high-low in one of dummy's suits I reassure Hashmat, "No worries, she'll be right." But is a Cooper Echo a claim? Grattan Endicott, 1st January 2004: +=+ The information conveyed by the echo is that, in 3NT, declarer will make nine or more tricks, is it not? If, instead, declarer says to dummy "I will make at least nine tricks" is this not a claim? Both communications appear to specify a number of tricks (nine) but to allow the number may be exceeded. The law does not appear to require that a claim should not allow of the possibility that more than the number claimed may accrue from the play about to be proposed in the statement of clarification. ~ Grattan ~ +=+ UNOFFICIAL -------------------------------------------------------------------- Important Notice: If you have received this email by mistake, please advise the sender and delete the message and attachments immediately. This email, including attachments, may contain confidential, sensitive, legally privileged and/or copyright information. Any review, retransmission, dissemination or other use of this information by persons or entities other than the intended recipient is prohibited. DIBP respects your privacy and has obligations under the Privacy Act 1988. The official departmental privacy policy can be viewed on the department's website at www.immi.gov.au. See: http://www.immi.gov.au/functional/privacy.htm --------------------------------------------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.rtflb.org/pipermail/blml/attachments/20140714/eda310f4/attachment-0001.html From jimfox00 at cox.net Mon Jul 14 05:53:27 2014 From: jimfox00 at cox.net (Jim Fox) Date: Sun, 13 Jul 2014 23:53:27 -0400 Subject: [BLML] IMP and VP Tables and Bridge Scoring (was RE: QWERTY (was IMP tables) [SEC=UNOFFICIAL]) In-Reply-To: References: Message-ID: For now, I will just state that checkers and tic-tac-toe are deterministic in the sense that a person who knows the correct strategy can never lose no matter what his opponent does. More later. Mmbridge -----Original Message----- From: blml-bounces at rtflb.org [mailto:blml-bounces at rtflb.org] On Behalf Of Jerry Fusselman Sent: 07/13/2014 7:58 PM To: Bridge Laws Mailing List Subject: Re: [BLML] IMP and VP Tables and Bridge Scoring (was RE: QWERTY (was IMP tables) [SEC=UNOFFICIAL]) >> Bridge is a probabilistic game and that is a good thing. Most interesting >> games are. >> > > Go and chess are very interesting games. > > Mmbridge: Although I believe that Go and chess are also probabilistic games, I won't argue the point other than to say that when you make a move in those games, there is some finite probability your opponent won't find the best move in response. One would probably try to estimate that probability when choosing which of several moves to make. More to the point, I guess I cannot tell whether or not you differ with "most". This is revisionist history. In this thread, you wrote that "Checkers and tic-tac-toe are deterministic games." In those games too, your opponent's reply to your move is unknowable. You can't logically claim that go and chess are probabilistic immediately after claiming the checkers is not. Besides, "probabilistic" would lose all meaning if any game in which your opponent has choices is deemed probabilistic. From jfusselman at gmail.com Mon Jul 14 06:30:27 2014 From: jfusselman at gmail.com (Jerry Fusselman) Date: Sun, 13 Jul 2014 23:30:27 -0500 Subject: [BLML] IMP and VP Tables and Bridge Scoring (was RE: QWERTY (was IMP tables) [SEC=UNOFFICIAL]) In-Reply-To: References: Message-ID: On Sun, Jul 13, 2014 at 10:53 PM, Jim Fox wrote: > For now, I will just state that checkers and tic-tac-toe are deterministic in the sense that a person who knows the correct strategy can never lose no matter what his opponent does. More later. ----------------------- Does any single human on earth know this correct strategy for checkers? No. Therefore, I'm not sure of the relevance of this one observation to the subject at hand, but I look forward to hearing your later thoughts. Thanks for continuing the discussion. I've already learned some things. (For example, I had no idea that checkers had been "solved" by a battery of computers.) Jerry Fusselman From jimfox00 at cox.net Wed Jul 16 23:55:24 2014 From: jimfox00 at cox.net (Jim Fox) Date: Wed, 16 Jul 2014 17:55:24 -0400 Subject: [BLML] IMP and VP Tables and Bridge Scoring (was RE: QWERTY (was IMP tables) [SEC=UNOFFICIAL]) In-Reply-To: References: Message-ID: -----Original Message----- From: blml-bounces at rtflb.org [mailto:blml-bounces at rtflb.org] On Behalf Of Jerry Fusselman Sent: 07/14/2014 12:30 AM To: Bridge Laws Mailing List Subject: Re: [BLML] IMP and VP Tables and Bridge Scoring (was RE: QWERTY (was IMP tables) [SEC=UNOFFICIAL]) On Sun, Jul 13, 2014 at 10:53 PM, Jim Fox wrote: > For now, I will just state that checkers and tic-tac-toe are deterministic in the sense that a person who knows the correct strategy can never lose no matter what his opponent does. More later. ----------------------- Does any single human on earth know this correct strategy for checkers? No. Therefore, I'm not sure of the relevance of this one observation to the subject at hand, but I look forward to hearing your later thoughts. Thanks for continuing the discussion. I've already learned some things. (For example, I had no idea that checkers had been "solved" by a battery of computers.) Mmbridge: As has qubic (a kind of 3-dimensional tic-tac-toe) Jerry Fusselman ----------------------------------------------------------- Mmbridge From richard.hills at immi.gov.au Fri Jul 18 04:23:20 2014 From: richard.hills at immi.gov.au (Richard James HILLS) Date: Fri, 18 Jul 2014 02:23:20 +0000 Subject: [BLML] QWERTY (was IMP tables) [SEC=UNOFFICIAL] Message-ID: <240635A98F59F24AAD8510EB05121DC339564DA1@SDCWPIPEX02.IMMI.LOCAL> UNOFFICIAL Richard Hills, July 2014: The original QWERTY typewriter contained ambiguities; for example the same key was used for lower-case "l" and the numeral "1". Similarly the 1997 Laws had ambiguities. Jesper Dybdal, October 2002: The first sentence of [the 1997] L11A is: (1) "The right to penalise an irregularity may be forfeited if either member of the non-offending side takes any action before summoning the Director." The second sentence is: (2) "The Director so rules when the non-offending side may have gained through subsequent action taken by an opponent in ignorance of the penalty." It seems to me that this can be read in two different ways: (A) (1) warns us that the players may forfeit a right, and (2) tells us exactly under which circumstances that is the case. I.e., the right is forfeited only in the circumstances specified by (2). and (B) (1) tells us that in some circumstances, players may forfeit a right, and (2) gives us one example of such circumstances. I.e., there may also be other circumstances, determined by the TD's judgement, where this right is lost. My original reading of it was (A), without having given it much thought; then, in the "When you don't call the TD" thread, Sven made it sound very convincing that (B) was right; but now I've read it a few more times and suspect that (A) may be right after all (I miss words like "for instance" in (2)). So I would appreciate it if someone would tell me what such a construct with "may" followed by "so rules when" actually means - in the English language and/or in bridge law. Richard Hills, July 2014: When I was perusing the drafts of the new Laws in 2006 - 2007 I remembered Jesper's request, so I mentioned it to the 2007 Laws coordinator Grattan. Thus the 2007 Law 11A is: "The right to rectification of an irregularity may be forfeited if either member of the non-offending side takes any action before summoning the Director. The Director does so rule, **for example**, when the non-offending side may have gained through subsequent action taken by an opponent in ignorance of the relevant provisions of the law." UNOFFICIAL -------------------------------------------------------------------- Important Notice: If you have received this email by mistake, please advise the sender and delete the message and attachments immediately. This email, including attachments, may contain confidential, sensitive, legally privileged and/or copyright information. Any review, retransmission, dissemination or other use of this information by persons or entities other than the intended recipient is prohibited. DIBP respects your privacy and has obligations under the Privacy Act 1988. The official departmental privacy policy can be viewed on the department's website at www.immi.gov.au. See: http://www.immi.gov.au/functional/privacy.htm --------------------------------------------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.rtflb.org/pipermail/blml/attachments/20140718/19be5797/attachment.html From jimfox00 at cox.net Tue Jul 29 06:47:10 2014 From: jimfox00 at cox.net (Jim Fox) Date: Tue, 29 Jul 2014 00:47:10 -0400 Subject: [BLML] Imp Tables (was QWERTY (was (Imp Tables (was QWERTY (was IMP tables)))) Message-ID: Test QWERTY Mmbridge -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.rtflb.org/pipermail/blml/attachments/20140729/633a292e/attachment.html