Jboling

I think playing Odwyrtka (found on Dan's WJ2005 webpage) is the best alternative: 2NT = 4+ support, 2♦ = diamonds or balanced less than 4 card support. Jassem suggest natural followups, but after 2NT you can for sure play your favorite Jacoby/Stenberg structure, and after 2♦ I think you can play something similar to Bubrotka: 2♥ = any minimum 2♠ = 11+, 4+M 2NT = 11+, 6+M, singlesuiter 3m = 9-11, 5+m, 4M I think that it is good that responder usually bid 2♥ or 2♠, after which opener can show a balanced hand with 2NT, and other bids are descriptive with 4+ diamonds (lowest rebid can be 4441). Furthermore it can noted that 1♣-1M-3♦ is also free for something. I think 5+♦-4+M is a good choice, as this means that 1C-1M-2D denies 4M.

I think it would be beneficial if 1♣ would deny clubs, as you then use the 2♣ rebid for various purposes with no additional cost, for example: 1♣-1M-2♣ = 10+ with diamonds (as I think Helene already implied). 1♣-1M-2♦ = 18+ Bubrotka or similar artificial. 1♣-1♦ = 0-6 1♣-1♦-2♣ = artificial GF 1♣-1NT = 7-11 no major 1♣-1NT-2♣ = 15+ with 5+♦, responders rebid 2♦ showing 7-8 1♣-1NT-2♦ = 10-14 with 5+♦ A problem with this is that 1♣-1M-2♣-2♦ must be as non-forcing, so no low-level relaying here.

I think the reasons for using 15-17 have been mentioned already, but I think 14-16 is optimal together with T-W. When I play like that I do not open 1♣ with 11 (unless it is an upgrade to 12). This means that the most common balanced range (12-13) is more accurately bid. The problem that Adam mentioned with getting too high with 17 can be reduced by splitting up 17-19 into two parts. And then you might as well combine it with 20-21 balanced. What I mean is that 1♣-1R-1M is either 12-13 or 19-21, and 1♣-1R-1NT is 17-18. 19 feels like a GF against a 1-over-1 response, while 17-18 is not it, so I like this a lot. Starting low with 20-21 balanced is also good. Typically you do 1-over-1 with 5 points anyway, and missing games with 21+4 is not too bad (21 balanced against any hand with 4hcp makes 3NT only about 45% of the time), so you can play like this without big changes (except you need to play 1♣-1R-1M as almost forcing, and agree on how to show 19-21 balanced later on in the bidding). And naturally you also win by freeing the 2NT opening for something else than 20-21 balanced.

I thought 1♣-1♦-3♦ is more narrowly defined, something like 21-23. With 18-20 you have to choose between 1♥, 1♠ and 1NT, in this case probably 1♥. Of course change one of the hearts to a minor card, and you have to bid a very non-ideal 1NT.

Yes this was the number I used as the standard deviation of VP round-off. This assumes that all results are possible and equally likely, which neither are true, but it is an ok approximation. The standard deviation for rounding off imps is then also about the same, but when you express that in VPs it is not a constant, it depends on the actual imp-difference. You can most conveniently observe it using the new imp-to-vp table for 12 deals matches, when you have a standard deviation of 1 expressed in IMPs. This means that we have a maximal standard deviation of 0.36 VPs at imp-difference 0, and below 0.3VPs at imp-differences greater than 8. So on average we should have about the same standard deviation.

Good point, I did some calculations regarding this too, I estimated that the standard deviations for the errors due to round off to whole numbers are about equal (~0.3VP per match) for the IMP and the VP conversion.

I did also try Henrys formula, and did niether get the same as USBF nor the same as woefuwabit. Below mine and USBF (in that order): 0 0 0 1 1039 1039 2 1077 1077 3 1114 1114 4 1150 1150 5 1185 1185 6 1218 1218 7 1251 1251 8 1283 1283 9 1314 1314 10 1343 1344 11 1372 1373 12 1400 1401 13 1428 1428 14 1454 1454 15 1480 1480 16 1505 1505 17 1529 1529 18 1552 1552 19 1575 1575 20 1597 1597 21 1618 1618 22 1639 1639 23 1659 1659 24 1678 1678 25 1697 1697 26 1716 1716 27 1734 1734 28 1751 1751 29 1768 1768 30 1784 1784 31 1800 1800 32 1815 1815 33 1830 1830 34 1844 1844 35 1858 1858 36 1871 1872 37 1884 1885 38 1897 1898 39 1910 1910 40 1922 1922 41 1933 1933 42 1944 1944 43 1955 1955 44 1966 1966 45 1976 1976 46 1986 1986 47 1996 1996 48 2000 2000 I think the difference compared with woefubit comes from that the type cast (int) does not round, I think it truncates. By the way I did the calculations using floating point, and first forgot that you then have to add a factor that takes into account the finite precision of floating point numbers in the comparison. For example ((V[i+1] - V) > (V - V[i-1]+0.005))

I have actually tried to reverse engineer it, I do not think that there is any simple underlying function, they have probably just constructed the tables by hand. The imp-limit for 20-0 seems to be 15 multplied by the square root of the number of deals. Using this (for normalization of the imp-ranges) and a third order polynomial I could get down to an maximal error of about 0.02, but no significant improvement if I used higher order polynomials. Which was an disappointment, an error below 0.005 would have resulted in a simple formula for perfect generation of the tables, which also could have been used for generation of conversion tables for any other number of deals as well.

Interesting, thanks for the tip. Here is another variation, transfer-Gazzilli, found under Note [5]: http://www.ecatsbridge.com/documents/files/cc/OpenTeams/Germany/Fritsche-rohowsky.pdf

If dbl was negative, 2D is just showing a hand without interest of game against a minimal negative double, with reasonable support for diamonds (and denying support for hearts)

When I played Wilkosz according Matulas scheme given above, I changed 3♣ to threesuited invite with short spades, because I thought it was missing from the scheme. With a weak hand with long clubs, we passed and corrected to 3♣ if it was doubled. The responses to 3♣ were pass/3♦/3♥ = to play (typically with spades as the other suit) 4m = m+♠, good hand, reinvite 4♥ = to play 4NT = both minors 5m = to play (typically m+♠ with weak spades).

I did the full table as well, but only with 1000 deals like Cascade: hcp west : 10+ tricks : average tricks 0 279.0000 8.9210 1.0000 302.0000 8.9730 2.0000 321.0000 9.0280 3.0000 310.0000 8.9660 4.0000 296.0000 8.9840 5.0000 269.0000 8.8920 6.0000 274.0000 8.9280 7.0000 312.0000 8.9450 8.0000 286.0000 8.9590 9.0000 294.0000 8.9190 10.0000 307.0000 8.9670 11.0000 294.0000 8.9210 12.0000 313.0000 9.0030 13.0000 332.0000 9.0330 14.0000 319.0000 9.0060 15.0000 334.0000 9.0410 16.0000 343.0000 9.0740 17.0000 334.0000 9.0350 18.0000 326.0000 9.0600 Looks quite similar to Cascades table, except that I limited it to a 4-4 fit in spades, so my numbers are lower. Having more than 8 cards in trumps increases the number of tricks, when I redid the 9hcp with west situation with a 4-4+ fit the number of 10+ deals increased from 294 to 391, and the average number of tricks increased from 8.9 to 9.2. So the effect of additional trumps is clearly bigger than the distribution of points with the opponents (in a DD setting, it is probably the other way around single dummy). Keeping everything else but hcp constant would be the ideal when you study the effect of hcp. By the way I got the same strange maximum at 2 points with west. The reason for that you get more tricks with the majority of points with east is probably that he is often endplayed when leading. Why this trend does not last to the end seems strange. But still I think the connection between these investigations and real play is rather weak, as in a single dummy setting you really gain a lot single dummy, if you know based on the bidding where the majority of the points are.

I tested a double dummy study, but realized that it probably was pointless. DD is DD, so you already know the location of every single card from the start, so nothing is gained from that you know that LHO has 14 points from the bidding. The average number of tricks, under the conditions Ken suggested, with opponents points 4-14 and 9-9 were 9.0 and 8.8 respectively. And the likelihood for making at least ten tricks (with a 4-4 spade fit, with spades as trumps) were 31% and 28% respectively. Or maybe this was what Ken actually was asking for? Most likely you gain much more in a single dummy setting from knowing more about the point distribution based on bidding.

No Name is one of the strong pass systems found in the book "Introduction to Weak Opening Systems" by Slawinski and Ruminski. Suspensor is a later developement of No Name by Balicki and Zmudzinski. All this according to http://www.bridgeguys.com/pdf/forcingpass.pdf, written by Jan-Erik Larsson and Ben Cowling.

I play these tranfers slightly different: 1. Transfer to a already bid suit is forced, because it might be sign-off. Super-accepting is ok, you then bid something else but completion of the transfer. 2. Transfer to a new suit asks for a four card fit, and is only completed with one, other bids are natural and denies a fit in the transfer suit. This makes it almost impossible to stop at the three-level in a new suit, but makes it easier to agree the trump suit, in case responder is interested in slam.

A third alternative is to add some convention that can handle such "hands of death". You can play gazzilli, which in this case means 2♦ is artificial and forcing, or include such a hand into a transfer-Walsh structure, where you gain a rebid compared to natural, due to the transfer. After minor openings, the latter is my preference (by opening 1♣ with all 18-19 balanced hands, you gain a rebid after 1♦ opening as well).

I choose between showing shortness and a five card side suit based on which message I want to transfer to my partner: - Your honors in my five card suit excellent, honors in other side suits are of less value - Your honors in my shortness suit are of little value, honors in the other suits are much better. Thus a I always tend to show a Hxxxx suit, while with a HHxxx-suit I might show the shortness. Especially a five card suit headed by AK works against almost any holding. I recall a discussion some time ago, here it is http://forums.bridgebase.com/index.php?showtopic=29348, where showing a AKJxx side suit did not work well. Well it was a hard hand to bid anyway, but in my opinion 3♥ seems as a better start on that hand.

I have lived in New York (3 areas) and California, as well as now living in Nevada. Played plenty of bridge in all places. Still haven't come across anyone in person that thinks anyone but 5-5 is standard. Was there really a time the jump showed a void? In the swedish version of Jacoby 2NT, commonly referred to as Stenberg 2NT, openers jump to the 4-level shows a void. Stenberg is common in Scandinavian countries, and I have seen people stating that they play J2NT when they in fact play Stenberg. In Stenberg 3-level rebids shows side-suits. Sounds suboptimal to have all suit rebids showing shortness.

I did one more calculation related to the original thread, trying to answer the following question: If we have the average 22.5 after 20 deals (a 1% chance, equally big chance for 17.5-), and the total constraint on the 26 deal set is 22.7, how does the likelihoods change for getting a certain number of hcp? hcp orig. lh new lh new/orig 0 0.0001 0.0001 1.0899 1.0000 0.0005 0.0005 1.0897 2.0000 0.0022 0.0024 1.0895 3.0000 0.0062 0.0068 1.0893 4.0000 0.0180 0.0196 1.0889 5.0000 0.0431 0.0469 1.0884 6.0000 0.0919 0.1000 1.0877 7.0000 0.1821 0.1980 1.0869 8.0000 0.3404 0.3697 1.0858 9.0000 0.5900 0.6398 1.0844 10.0000 0.9540 1.0328 1.0826 11.0000 1.4648 1.5825 1.0804 12.0000 2.1220 2.2866 1.0776 13.0000 2.9397 3.1576 1.0741 14.0000 3.8845 4.1561 1.0699 15.0000 4.8966 5.2137 1.0647 16.0000 5.9067 6.2525 1.0585 17.0000 6.8372 7.1868 1.0511 18.0000 7.5739 7.8947 1.0424 19.0000 8.0400 8.2978 1.0321 20.0000 8.2218 8.3869 1.0201 21.0000 8.0568 8.1074 1.0063 22.0000 7.5668 7.4950 0.9905 23.0000 6.8095 6.6232 0.9726 24.0000 5.9072 5.6270 0.9526 25.0000 4.8920 4.5506 0.9302 26.0000 3.8795 3.5131 0.9056 27.0000 2.9510 2.5927 0.8786 28.0000 2.1292 1.8085 0.8494 29.0000 1.4635 1.1970 0.8179 30.0000 0.9556 0.7496 0.7845 31.0000 0.5865 0.4394 0.7491 32.0000 0.3375 0.2404 0.7121 33.0000 0.1846 0.1243 0.6737 34.0000 0.0942 0.0597 0.6342 35.0000 0.0435 0.0259 0.5940 36.0000 0.0176 0.0097 0.5533 37.0000 0.0067 0.0034 0.5125 38.0000 0.0020 0.0009 0.4721 39.0000 0.0005 0.0002 0.4322 40.0000 0.0000 0.0000 0.3934 The original likelihoods (denoted lh above) for low and high hcp are inaccurate, as they are obtained through simulation. In the 10 million deal set there were for example 7 deals with 0 points and 3 deals with 40 points. Fortunately these extreme values have little impact on the relative change in the likelihoods, which were obtained as follows: 1. Assume hcp between 0 and 40 for deal 21 2. Calculate the upper limit on the average for the remainng 5 deals (trivial) 3. Calculate the likelihood for that 5 random hands will be under that limit (obtained by numerical integration of the pdf, with the sigma = 2.13 for sets of 5 deals). 4. Repeat 1-3 for all hcp 5. This gives a multiplicative modifier on the likelihood for each hcp. The final likelihoods are obtained by normalization so that the total likelihood will be 100%. So the likelihood for getting 25 points is reduced by a factor of 0.93, and likelihood that the oppos have 25 is increased by a factor 1.06. And having soft constraints shouldn't change these much. It is a bit more than I expected, but still I'm not sure if it would affect my bidding. When I preempt I do it under the assumption that the opponents have game anyway (I use the 4-3-2-rule). And the likelihood for our side having 20+ points is still almost 52%. But anyway I agree with the others about that such constraints will probably never be accepted, for the reasons already mentioned. PS. For those who teaches probabalities and statistics: The frequency table for sum of north and south hcp, which hotshot published, is a nice illustration of the central limit theorem. The hcp frequency table for each player is not that normally distributed, but the hcp sum of two players is very close to normally distributed. And the hcp averages of sets of more than one deal become even more normally distributed. So my above given numbers, which were calculated assuming normal distribution, should be quite accurate. Edit: Forgot that the above likelihoods are a priori likelihoods, that is before looking at your hand. Typically you do that before you start bidding. The current hcp has of course a big impact on this, for instance if you have 9hcp instead of the expected 10 the unbiased expected value for the total hcp will be 19.33. And so on, a totally different situation for each current hcp-value. The above relative numbers could still be quite close to the truth. More accurate numbers could be calculated, but I'm not doing it.

Are you implying that a limit of 22.7 on 27 deals is equivalent to any limit between 21.86 and 23.36 on 26 deals? A don't buy that. If the limit 21.86 is enforced on 26 deals, I think most of the discarded deal sets would be within the limit 22.7 on 27 deals. And a par with an average of 17 hcp allowed by the 23.36 limit would have every reason to be disappointed and angry since you have promised to deliver quality checked deal sets... What I meant is that you could aswell genereate the extra deal first (it makes absolutely no difference if you do it last), and calculate the hcp for N-S, which will be random number between 0 and 40. And then you can use this number to modify the upper and lower hcp limits for the set of deals to used in the competition. In the discussed case the upper limits will then be between (22.7*26)/27=21.86 and (22.7*26+40)/27=23.34. This limit will be close to normal distributed, with the mean (22.7*26+20)/27=22.6 (for the upper limit)and 4.77/27=1.8 as standard deviation (thanks to hotShot for a well timed calculation). So that would give that in 52% of the cases the limit is 22.7 or lower, and in the remaing 48% it is higher. Still no time to study variance. Another interesting calculation would be how much does the likelihood for getting for example 25+ or 30+ hcp in the next deal decrease due to the limits.

I think this analysis is flawed. With the given constraints, each side knows they will have in total 520 +-70 hcp in a 26 deal set. If the NS average is 22.5 after 20 deals, they have had 450 hcp. Thus, they know with certainty that they will have 70 +-70 hcp, i.e. between 0 and 140 hcp, in the remaining 6 deals. They know, for example, that they cannot have the points for game on all remaining deals. That is knowing a lot compared to random deals. And of course, the situation could become more extreme closer to the last deal. Similarly, EW know they will have between 100 and 240 hcp in the remaining 6 deals. This means, for example, that balancing and competing for the part score will be pretty risk-free. EW know they will have some values and will not be unlucky in terms of hcp. Note that in all cases the table only tells about the next deal. After that deal, the average so far and the expected future average will be different. What khaggblo probably is thinking about is that if you for example get 25 on the next 5 deals, you will then know that you cannot have more than 15 points on the last deal. But this is something different, it means that you have the average 23 after 25 deals, which is much less likely. For the average 22.5 after 25 deals you know that you cannot have more than 27 on the last deal, and the average 22 after 25 deals means the upper limit is 40 on the last deal, that is it is completely unbiased. But as I mentioned previously, problems appear especially on the last deal, and this can be most easily remedied by using a random limit, so that the exact limit is unknown to everybody, which adds up to a big uncertainty for the average calculating players especially on the last deal. Which actually is the same as having an extra deal or two, which nobody plays. If we have one extra deal, with the total upper limit of 22.7, would mean that actual upper limit on the first 26 would then be between 21.86 and 23.36. Which is already a quite a big uncertainty, in the above mentioned case we knew that we had 15 or less on te last deal, which will change to that you can have 32 or less. With two extra deals the upper limit will be between 21.1 and 23.9. Even better results you get if you add a truly normally distributed offset to the limit, when you get no known limits. Except that you have to check that the upper limit does not go too close to 20, but a known lower limit on the upper limit does not help anybody, as far as I can see. Another issue is that my table is about mean values, it does not say anything about how the upper limit affects the variance, it might become more biased. Anybody want to make a try? I don't have the time at the moment.

If you cancel the whole set based on some statistic as the average number of hcp, you will introduce constraints on that statistic, and the set will become less random in that sense that you can predict something about the future deals based on the past deals. But if the constraints are loose enough, and even better the constraints would be unknown to everbody, I think it could become impossible for the players to predict the future. I did some sims two years ago, it was more difficult than usual when you need to do statistics on statistics of sets of deals. My conclusion was that about three sigmas (standard deviations), was about possible without the limits becoming relevant almost ever, even in a case when the limits were known to players. This means accepting 99.7% of the deals (in case you have one constraint, for instance hcp for N-S, which automatically limits E-S as well). I used sets of 26 deals as an example, with the three sigma limits 17.3-22.7: xxxx 10.0 16.0 20.0 23.0 24.0 25.0 -------------------------------------------- 21.5: 20.0 20.0 20.0 20.0 20.0 20.0 22.0: 20.0 20.0 20.0 20.0 20.0 20.0 22.5: 20.0 19.9 19.8 19.7 19.6 19.6 23.0: 20.0 19.8 19.4 18.2 17.0 13.4 23.5: 19.9 19.5 18.5 15.5 12.2 2.4 In the table I have the expected average hcp of the remaining deals, above we have the played deals so far, and to the left the average so far. So if your average will stay under 22.5 (and over 17.5) you will practically have no information at all. For instance at 20 deals we have total bias of a 6*0.2=1.2, that is slightly more than a jack totally on the remaining deals. The average will stay between 17.5 and 22.5 99.4% of the time for 26 deals, and it will be less common for less deals. So I think most of you overestimate dramatically the impact of having hcp-limits on actual playing decisions. I also did the same statistics for two sigmas, that is 18.2-21.8 limits, which probably is too tight limits: xxxx 10.0 16.0 20.0 23.0 24.0 25.0 -------------------------------------------- 21.5: 19.9 19.8 19.7 19.7 19.7 19.8 22.0: 19.8 19.5 19.1 18.1 17.1 14.4 22.5: 19.7 19.2 18.2 15.3 12.4 3.6 23.0: 19.6 18.7 16.9 11.9 6.8 -7.7 23.5: 19.5 18.1 15.5 8.3 1.2 -20.2 Here you could probably use information about the limits around 10% of the deal sets, but only at the end of the set. I recall that only our old Unix machine with Solaris operating system had flexible enough command tools to do the above statistics. I think I needed the command flex to grep the relevant statistics from Staverens dealer, which can do the hcp-statistics for 26 deals. The Unix machine is unfortunately no longe with us, so unfortunately I cant easily do more statistics of the above type. Helene mentioned that she would accept such limits only in holiday bridge. I think competitve bridge could benefit from such limits as well. As I mentioend earlier if you get less points your bridge skills will have less impact on your result. HAving hcp-constraints could be viewed as design of experiments, where you try to estimate the skills of participating players. And as some of might know, truly random experiments are not optimal when you make only a finite number of tests on your test subject. I'm not suggesting that adding hcp-constraints would make tournaments optimal tests, just slightly better than truly random deals, so you could get more accurate estimates of bridge skills using less deals Somebody mentioned finesses in this context, it is safer to use constraints on the pairs than on the individual players, as the latter could have impact on making finesses, while the former have no impact on this. During play of the hand you know the sum of the opponents hcp exactly when you see the dummy.

What? Where do you live? That's awful if it's true, I've never heard of such a thing happening. I think it had more to do with that people complain about computer generated hands that are too freakish (becouse they are used with balanced hand dealt ones), and earlier days when there actually were problems with computer generated deals. So my statement might not be accurate for todays situation. But still I'm not revealing where I'm from, I can tell that it's outside US, so Josh doesn't have to start mistrusting his TDs :). But do you really think that any TD with a instict for self-preservation would let through a set of deals with the 40-0 hcp-distribution on all deals? Of course not, the TD would (at least) be accused of manipulation of the deals, if he/she wouldn't manipulate the deals by redealing them! Of course the likelihood for such a set of deals is extremely small, I haven't even seen a single deal with all the points on one side.

Let me give you a counterexample, let us say all the deals in a tournament are distributed 40 points to N-S and 0 points to E-W. It is extremely unlikely, but it is possible. Would it be a fair and nice tournament, especially for the pairs that would sit E-W all the time? No, and the results would probably be close to 50% to everybody, and if a director would note this in advance he would redeal the hands without further thought. I asked directors about this, and yes they do look through the deals before the tournaments, and occasionally redeal the deals. So we actually have some kind of quality control of the deals already. And it could be automated, and it could be done so that nobody would notice the difference, except that the freak sets of deals would disappear, but nobody would miss those. For instance such that 99% of the deals have expected hcp for any pair is 20.0, and with the limits 0 and 40. And in the 1% of the deals where we have bias, one could do it such that in most cases this information would be uncertain to the players. Yet another way of rephrasing it, the set of deals used in a tournament could be viewed as a test of bridge skill. The test should aim at testing all the players skill, and the more high cards you have the more your skill will be put to a test. Thus if a player gets significantly less points in a tournament, the players skill will have a smaller impact on his/her result. Bridge players are here the customers that buy these bridge skill tests called tournamanents, and they would deserve tests that are as fair as possible. And yet another argument (I'll stop after this one), I am quite convinced that hand dealt deals are much more biased (towards balanced distributions) than deals that would go through a well designed hcp-limit test. And the bias from hand dealing concerns more or less every deal, but still we agree to play hand dealt deals.

I agree with Fluffy, I suggested something similar in a domestic discussion some time ago. My argument was that competitions would me more fair if all players would have about the same number of points to play around with. Especially in short tournaments the average number of points for a certain pair can vary a lot, and a pair with a low average on the hcp will obviously have less impact on its results. My interpretation of limiting the average is that is a sort of quality control of the set of the deal, one just discard the obviously unfair sets, with the aim of getting fairer results and more pleasant experiences at the bridge table. The main counterargument in the domestic discussion was that hard contriants on the average hcp could enable people to estimate partners hand strength in the later deals. I countersuggested that the constraints should be random and thus unknown, and possibly even soft (that is just reducing the likelihood for averages lying outside the constraints). And I also tried to show that keeping track of ones average would be very rarely of any use, especially if one intruduced some randomness in the constraints as well. PS. During the domestic discussion I estimated the standard deviation for hcp for pair in a 26 deal competition, it was about 0.91. So in 95% of the cases the average should stay between 18.2 and 21.8, in a 26 deal tournament. Two sigmas was my first suggestion as constraints, but I think that could be slightly too narrow constraints. Due to that people could then with a reasonable frequency use the constraints for some information about partners strength in the last deals. With some information I mean that maybe 1/20 of the sets one can on the last five deals know that our sides expected average hcp is about 19 and not 20. It was much more unlikely that one could judge on the last deal that the other side has 30+ points, but it was possible. So introducing constraints is somewhat problematic, but I think it could be done, using random and unknown limits. And it would for sure increase fairness, on average at least...