Join Bridge Winners
All comments by Henry Bethe
You are ignoring the author of this comment. Click to temporarily show the comment.
Marty,
You are of course right that the scores should be based on expected value. But except for Australia Bridge, which scored at imps, no bidding feature has ever done so. Bidding a 75% slam has always received the score that assumed it made. I'm not defending that - or advocating it - but observing that's the way it is. But I find it objectionable that the assumptions about the lie of the cards are often far from what I would assume in a decent standard matchpoint game.
Sept. 11, 2014
You are ignoring the author of this comment. Click to temporarily show the comment.
If you would like me to do more of these, please like the article. Thanks.
Sept. 11, 2014
You are ignoring the author of this comment. Click to temporarily show the comment.
As I recall the term NABC came (as so often) at the request of the Canadian Reps on the BOD who wanted to distinguish them from the Canadian Nationals, which is an annual tournament held in Toronto.
Aug. 31, 2014
You are ignoring the author of this comment. Click to temporarily show the comment.
Since I think 4 is a splinter in support of hearts, I think partner has a minimum hand with no slam interest. But obviously I did not think that when I bid 4, so it should show a long heart suit with no particular minor suit fit.
Aug. 24, 2014
Henry Bethe edited this comment Aug. 24, 2014
You are ignoring the author of this comment. Click to temporarily show the comment.
I must be the only person who does not understand the previous auction. Partner opened 1 and I raised to 2, inverted. In my world that is limit or better. Partner bid 2N showing a balanced hand, either too weak to open 1N or too strong, in any case forcing to 3. I bid 3 showing shortness, and partner bid 3. I still have a six LTC hand - no reason to look for slam unless partner has extras, which he has not shown.
Why wouldn't I bid 3N at this point? If partner can't bid again slam probably has no play.
Aug. 22, 2014
You are ignoring the author of this comment. Click to temporarily show the comment.
At least in my circles partner should not be leading the 6 from 7642 or 76(42), so partner either has Q76 or 6 doubleton.
Aug. 22, 2014
You are ignoring the author of this comment. Click to temporarily show the comment.
With both opponents passing I think it is important not to exaggerate high card values to partner.
Aug. 21, 2014
You are ignoring the author of this comment. Click to temporarily show the comment.
A rousing “Hurrah” for the Botswanna Juniors who won their first match in round 14.
Aug. 17, 2014
You are ignoring the author of this comment. Click to temporarily show the comment.
Actually, Yuan, very few people use “Butler” scoring anymore. When we talk about Cross Imps we mean compute each comparison and add them up. So if 50 pairs bid slam and 50 pairs don't, the bidders will gain 11 x 50 imps, or 550 imps. To reduce this to a meaningful number, divide by the number of scores. So each bidder-of-slam would gain 550/100 or 5.5 imps. Unfortunately some people divide by the number of comparisons, so they would compute 550/99 or 5.56 imps.

With computers Cross Imps are no more difficult than Butler.
Aug. 11, 2014
You are ignoring the author of this comment. Click to temporarily show the comment.
On the boards I looked at there were several on which there were plus and minus scores N-S. The plusses were typically +100, +110 with a few variations, and the minuses similarly distributed. So if, in a match, there was a swing, it would probably be 5 or 6 imps. But often there would be no swing.

There was one board on which every N-S pair was either +1010 or +510. About 1/3 of the scores were 1010. How would this board show if these were the scores from 26 team matches? It is conceivable that there would be almost no swings: in most matches either both pairs bid slam or neither pair. It is conceivable that every pair that bid slam would gain 11 imps. If the only evidence we are presented is the board total imp scores, e.g. 22 imps were scored or 154 imps were scored, we would make different conclusions about the flatness of the board. This is essentially the evidence I have available from match scores.
Aug. 11, 2014
You are ignoring the author of this comment. Click to temporarily show the comment.
Bob,
I do not regard five to seven imps as “small” scores. On every one of the boards at which I looked where no games were bid and made there were plus scores in both directions. Quite a few in each.
Aug. 10, 2014
You are ignoring the author of this comment. Click to temporarily show the comment.
Interesting. Leo and I make the same bid with very different meanings. I bid 2N natural and invitational. Given that partner did not make a takeout double I suspect he has some diamonds and that 4 would be vulnerable to diamond ruffs. Nine tricks may be easier than ten.
Aug. 10, 2014
You are ignoring the author of this comment. Click to temporarily show the comment.
Adam,
I did a quick eyeball check of the LM pairs first and second sessions. In each session there were four of the twenty-six boards on which virtually every pair bid and made game with the only variance being the number of overtricks. (Interestingly there were no ‘flat’ boards with all small scores.) The boards were played 52 times, so this seems like a reasonably large sample to find flat boards.

4/26 is 15%. My statistics are not as good as they once were, but I believe this would make the probability of a ‘flat set’ of even seven boards, say one in which five of seven boards are flat, vanishingly small.
Aug. 10, 2014
You are ignoring the author of this comment. Click to temporarily show the comment.
I posed the other hand as a problem before the 3 bid on August 4. 3 was the plurality vote holding x AK7xxx Kx AQxx with double, 4 and 4 trailing in that order. 6 by opener is an excellent contract, requiring 2-1 clubs and hearts not worse than 5-1, or if clubs are 3-0, hearts 4-2 or better. By responder, with the 2 bid where it is, 6 is awful.

Now I think it is something of an indictment of modern styles of bidding when the only people who will play clubs from the proper side of the table are those, roughly 15%, who bid 4. But perhaps I am harsh.
Aug. 9, 2014
You are ignoring the author of this comment. Click to temporarily show the comment.
I would have bid 4 over 3. I don't think this hand is good enough to drive to slam opposite a game force. And bidding Blackwood says you know what to do after partner's response. I don't think you do.
Aug. 9, 2014
You are ignoring the author of this comment. Click to temporarily show the comment.
Let me start by saying that I have no doubt that duplicated boards add to the sociability and the perceived fairness of events. If I have not said that before, I emphasize it now. I like duplicated boards. I like them as a player and even more so as an analyst and reporter.

Now there seems to be an issue of what constitutes swingy sets of boards and what constitutes flat sets. My definition, personal I admit, is that a swingy set is one on which a lot of imps are actually scored. If I play a seven board Swiss match and the final score is 32 to 20, I think of that as a swingy set. If it is 15-3 it isn't. At least not at my table. The score of the set matters to my perception. If there were opportunities for swings that weren't taken that does not affect my perception.

Various people have suggested that there are inherently flat sets of boards. Sathya gives an example. Bob argues from first principles. I can tell you that in the short matches of the USBC round robins there has not been a set that I can find on which the average number of imps per board has been less than 3.5. There has never been a set on which all or even a majority of margins have been small (less than 1 imp per board). I suspect that if someone was willing to do the work they would find the same phenomena in the large Swisses in Australia, and that the distribution of VP results does not vary significantly from one match to another. If there were flat and swingy sets of boards this would not be the case.

My argument is not that duplicated boards are not on balance a positive. It is that they are not necessary for a fair event.
Aug. 9, 2014
You are ignoring the author of this comment. Click to temporarily show the comment.
Aaron, I would really be interested to know what the facts are on which you would base your rebuttal. I showed the basis of my conclusion.
Aug. 8, 2014
You are ignoring the author of this comment. Click to temporarily show the comment.
FWIW, in the US I am told this is a geographic thing. In the NE and the West Coast there is near unanimity on the 20 point scale. In more central parts of the country there seem to be many districts/units that prefer either the 30 point scale or win/loss.
Aug. 8, 2014
You are ignoring the author of this comment. Click to temporarily show the comment.
The anecdote from the 1987 NA Swiss is for a seven-board match. The teams involved were ranked 1-2-3-4-6-8 of about 40 going into the last match. So all of the players were pretty good - and generally playing well - yet one match on these seven duplicated boards had essentially no swings while the other two produced a lot of imps.
Aug. 8, 2014
You are ignoring the author of this comment. Click to temporarily show the comment.
Statistics don't bear out your notion. The average imps per board in Swisses with table-dealt boards are indistinguishable from those with computer dealt boards.
Aug. 8, 2014
.

Bottom Home Top