NYC Board-a-Match

I had a good time last week at the NYC regional and came in a close 4th in the board-a-match...close to 2nd, that is. The winners were about 6 boards clear of the field! I wrote up a cute hand where I could have reduced the margin a small amount -- check out http://www.bridgewinners.com/blogs/nyc-bam-validating-an-insult.html

Check out Bridge Winners

Hi Everyone,

I'm not playing much lately and have less time to blog. In the unlikely event that any readers here haven't seen it, I highly recommend the new site Bridge Winners created by Gavin Wolpert, Jason Feldman and others. It's attracting columns by a lot of great people. I've been commenting regularly...you might check out Kit Woolsey's latest exciting hand and my comments. As a theory nerd I can occasionally point out something interesting even to players who are much better than I am.

Bottom line: I will leave this site up, and still love bridge, but don't expect too much activity! If I do come up with something I might write an entry over at Bridge Winners, which has become bridge central. Many thanks to everyone who has regularly read and commented here!

Jonathan

Robot madness

Regular readers know that I'm fairly unbiased on this blog, in terms of including lots of hands I messed up, etc., right? I hope this gives me license to publicize the following. Last night I substantially broke my previous record in a (12-board) robot duplicate, scoring 77.1%. I resolved not to play for a while, since my next effort would surely disappoint, right? Well, but I couldn't resist a game today, and as predicted I couldn't live up to last night, scoring 74.8%, my second-best score ever. :-) Yay, 2.4 masterpoints. Now I really shouldn't play for a while.

Good luck to everyone in Philly...I'm not going, have to get back to work.

Bidding Problem

Here’s an interesting hand to bid from the Spingold. I had a promising collection in 1^st seat white/red, AKQxxx Kxxxx x x, which got better when partner responded with Jacoby 2nt. What is your plan? Assume standard Jacoby where 4x=5-card suit and at least a sound opener. I don’t think any of the popular modifications exactly solve this hand either.

Update on the mysterious origins of the fatal hand

I was curious enough to get in touch with the author, Gary Pomerantz, who was kind enough to send a prompt reply:

---------

Hi Jonathan, Thanks for your very nice email. On the night Jack Bennett was killed, the Hofmans were interviewed by Kansas City police. Both Mayme and Charles Hofman indicated that they could not remember the distribution of cards in the so-called "Fatal Hand." Myrtle was delirious on that night, and into the wee hours, and was given a sedative. To my knowlege, she was never asked about the distribution of cards. My Best, GP

---------

I agree with Mr. Pomerantz that this makes it almost certain that the deal was a fabrication. What I gather, though, is that he doesn't have any first or second-hand accounts of the actual concoction, but as far as he knows it first appeared in The Bridge World so he assumes they were the ones who made it up. He is very likely to be right. If there were any bridge-playing cops on the scene (not so far-fetched in 1929) they *might* have been able to reconstruct the deal, but one would expect there to be a record of such a thing happening, and The Bridge World might have mentioned that, so I really doubt such a reconstruction happened.

The fatal hand

I just read "The Devil's Tickets" by Gary Pomerantz, a very entertaining book published a couple of years ago (targeted to non-bridge-players) which details both the famous Bennett bridge murder and the rise of Ely Culbertson as the nation's bridge guru. I've seen the "fatal hand" which led to the murder many times in bridge publications. Interestingly, this book states that aside from the bidding, the fatal hand itself is a fabrication! The participants were social players who wouldn't have been able to reconstruct the exact cards, especially since the declarer was dead.

Unfortunately, though the book is mostly footnoted, the author gives no source for the claim of fabrication, which I nonetheless tend to believe. He says that the hand was constructed by Sidney Lenz and first appeared in The Bridge World as part of Culbertson's never-ending search for publicity. It has since been reproduced many times. If someone has the latest Bridge Encyclopedia, I am curious whether there is any mention of the supposed hand being a construction.

Too bad Hercule Poirot wasn't on the scene with the police; he would surely have reconstructed the hands accurately, even though the tricks were gathered rubber-bridge style. Yes, I also recommend "Cards on the Table" to those who have missed it.

Not so fast (updated)

Here is a hand I enjoyed playing from our first-round loss in the Spingold. I was in 3C with

963 6542 QJ K873

KT54 7 K5 AQJ962

on the uncontested auction 1C-1D!(hearts)-1S-2C-3C. I considered the 3C bid very close; apparently at the other table they thought a long time and passed. Anyway, it’s nice to see the opponents can make 4D and probably 4H. Could I make 3C? They led a heart to the A and the DA. I dropped the DK in case I needed the entry; since I did this in tempo, I guess they were afraid I was 4-2-1-6 and they fatally played a second heart. I ruffed with the 9, played a diamond to dummy for a heart ruff, and played the CQ to the K, very pleased when lefty had stiff T. I could claim now; I eliminated the last heart with a high ruff, crossed to the C8 pulling the last trump, and led a spade planning to cover RHO’s card for the endplay. The opposing spades were AJx/Qxx so this was a very nice result requiring a little help from the defense. Fun. But on later sober reflection, I realized I played this hand wrong! Do you see why? Answer tomorrow.

The result was a 1-imp pickup when our teammates didn’t get in the auction either and were -90.

---------------------------------

The actual right way to play: I should use my trump entry before the diamond entry. This works just as well when the T drops, but leaves me better placed when it doesn't to make against both A onside and Qx or Jx onside, because I find out the trump position earlier. Work it out.

A 5-or-7 hand, at the one-level

I was kibitzing a hand from the Spingold QFs where Fantoni, in a 3-card ending, wound up having to guess whether the opposing trumps were initially 97/KJ54 or 95/KJ74. This was necessary in order to get out for 800 instead of 1100 in 1S doubled! In particular, declarer’s trumps were initially T863/AQ2. LHO had earlier overruffed declarer’s 8 with the 9, righty had ruffed a plain suit with the 4, dummy had ruffed with the A to avoid an overruff, and RHO had just ruffed ahead of declarer with the J, leaving a remaining trump position of

Q2

x Ky

T63

where {x,y}={5,7}. Now RHO led a suit where declarer and LHO but not dummy were void. To take 2 tricks, declarer must ruff with the 6 if x=5, but with the T if x=7. Yikes! The commentators were all saying this was a pure guess. A waste of time to even think about it, right? What possible inference could there be about the 5 and 7? There is one clue, though…do you see it? Answer below.

The key is to ask yourself whether either defender had a choice of how to defend with either holding. For lefty, there is no inference whatsoever as he had no choice from either 97 or 95. RHO, though, could have ruffed with the 5 or 4 from KJ54, but his choice would be restricted to the 4 from KJ74. That’s right, restricted choice operates on the 5 and 4 spots! Declarer should therefore play righty for KJ74 and ruff with the 6. By the way, it’s a very good habit to play randomly from equivalent spots when you are not signaling…this doesn’t require that you foresee esoteric positions like this, it’s just good general technique that minimizes declarer’s information.

In real life, Fantoni ruffed with the T and was punished with -1100. At the other table, oddly, they got to 2H-X in a slightly stronger 4-3 fit and did two tricks better for -800. You can see the hand record here.

Thanks to vugraph operator Dan Wolkowitz for the joke in the title. He’s the best I’ve seen at operating accurately and making good comments at the same time.

The patient lived, but...

Some of you saw the NYT column where I was in 6C with

x Axxxxx Kxx K92

Ax Kx AQJx AT543

after LHO overcalled 1S and righty bid 4S, and Franco did very well to pull my double to 5C. Unfortunately, I didn't play this hand well. My problems started when lefty led a diamond which I wrongly assumed was shortness, so that he (if anyone) was likely to have trump length. I then debated whether I was willing to safety the trumps (CA, club to 9) and decided it wasn't worth the extra risks and I should just stick to basics and play for clubs 3-2 after which it's a claim. BUT having decided that, I should make sure to also make against stiff Q or J of trumps. (For instance, CK, SA, spade ruff*.) However, with the idea of lefty having the trumps being an idee fixe, and thinking I was just banging down AK trumps anyway, I started trumps with CA and another, lefty having stiff J as it happened. Now, since I couldn't ruff a spade, I was forced to rely on hearts 3-2, and fortunately they were, but unfortunately I'm the kind of person who is haunted by this kind of thing anyway. For the record, my play blows the slam about 4% of the time, but more than that it's just ugly. Naturally I'm hoping this confession will be good for the soul.

*Best start is actually CK, C2...rho might split from QJxx. If no honor appears, you have a close decision whether to go up A or hook the T. Probably the T, but depends how you read the opening lead.

To cleanse the mental palate, here's a bread-and-butter, but non-routine, hand from the R16 that I played right:

Q9853 QJxx x Kxx

AJT74 --- AKxx Axxx

I was in 6S at white/red, after P-P-1S-(2S)-4S-(5H)-6S-AP. They led HK. Try it if you like. More next time. Update: see lengthy analysis on this hand in the comments.

Deep Six

After a slight bidding mishap in the round-robin, we wound up in a 6nt that (after a somewhat marked double-hook against an overcaller in his suit) required 5 tricks from AKJ62/T87 with plentiful transportation. Thanks to those juicy middle spots, you can pick up all 4-1 breaks with the Q onside after cashing the A in case of stiff Q, then running the T. In fact lefty did have Q9xx and we made it. I record this hand only because I can’t recall another instance of making a slam because of a crucial 6-spot.

Of course you can pick up Q9xx double-dummy without the 6, but only at the risk of blowing to 9x offside. This is hand 50 at http://usbf.org/docs/vugraphs/USBC2010/2ndRR46-54.PDF. Given the luck involved, I’m happy to report that both teams involved were going to qualify by at least 10 VPs regardless of the outcome on this hand.

Unusual sacrifice

After an unexpectedly long week, I must get back to work, so for the moment I’ll just finish the story of that highly successful “save” against 3nt from two posts ago. Note: We actually played more than any other team, since all the other semifinalists had QF byes.

Anyway this hand was from the penultimate match in the first round-robin. (Board 65 at http://usbf.org/docs/vugraphs/USBC2010/RR64-72.PDF.) At this point we were above water but certainly not a lock to qualify. Looking back at the hand records, I see I was 2^nd seat not 3^rd. I held AJ7 J3 KQJ98 J93 white/white and heard the auction I mentioned:

P-1D-2C-2H;

2NT-P-3NT-P;

P-??

Now here are the opponents in 3nt, and I know they have less than half the hcp. It has to be a bit tempting to double, but let’s give them some credit. They almost certainly have at least the two red aces and six club tricks for down 1…this was my first estimate. As I thought longer, I decided lefty almost had to have 7 clubs for both of them to have reasonable bids. Clubs were likely to be 7-3-2-1 around the table, and partner to have a light distributional 2H bid. So I wasn’t doubling…should I save? It’s certainly odd to save over 3nt with a balanced shape and no established fit, but partner must be distributional, so we shouldn’t be down very many with my having no wastage opposite his hope-for stiff club, and we might just make something. I decided to take a shot. I didn’t know if we should be in hearts or diamonds so I bid 4C, the famous pick-a-save cuebid. Partner chose hearts with K9x KQT98x xxx x and was able to wrap up 590 despite a 5-0 trump break when they could only tap him once – as Franco noted, both 3nt and 4H require clubs 7-2 to make. (Also, his S9 was good enough to freeze that suit from attack.) That was a 21-imp improvement on -400 since our teammates sold to 3D, -110. We seemed to gain momentum after this hand and blitzed this match to virtually clinch survival of the first cut.

USBC seeding is fairly accurate

The teams remaining were originally ranked 1,2,3... and 20 of 23 entrants. 1,2,3 had byes to the QF and all won by at least 73, though #1 Nickell's match was close most of the way. This is a completely neutral report :-).

Cute problem from Round Robin

Not going to take much time on posts since we're still alive in the QF today, but here's an interesting hand to declare. You are in 4S:

Tx QJxxx QJxxx x

AKQJxx Kxx A Kxx

You are in 4S and your diabolical LHO, Justin Lall, leads a trump. At the other table, no trump lead and it makes. How do you play to try to avoid losing 12 imps?

USBC RR in progress

I’m resting between days of the USBC and won’t describe any hands in detail. Just let me tell you the auction on one of my favorite hands of all time. I was 3rd seat both white and it went:

P-P-1D-2C;

2H-2NT-P-3NT;

P-P-4C(!)-P:

4H-X-AP

Cold for +590…on a hand where the opponents had done very well to bid a cold 3nt on 19 hcp! They fell from grace and didn't save in 4nt, but the doubler did have 5 trumps(!). Try to reconstruct the hand :-). More after the event is over.

K98xxx opposite J7x -- lose at most 2

I hadn't seen this one before today. The answer, I believe, is to run the 9 (or low to 7.) This loses only to stiff T offside. Other plays lose to 2 cases, one 3-1 break and one 4-0 break (lots of choice as to which), while this one picks up both 4-0s.

On a lighter note

A few years ago, someone clipped an article from the Bridge Bulletin about the Cavendish pairs to show my grandmother. They knew she had a bridge-playing grandson named Weinstein. My grandmother was pretty with it, but past 90 and could be a little confused about things sometimes. Anyway, she showed me the article and said, "It's wonderful you won this tournament, but it's a terrible picture of you! And they got your name wrong, you're not Steve!"

Congratulations to *Steve* Weinstein and Bobby Levin for winning the Cavendish an ever-more-absurd number of times. And Steve shouldn't take offense about the "terrible picture"; in my grandmother's world, all her descendants should be models.

USBC RR Format: Avoiding Sitouts

Hi all,

I'm planning to send this message to the USBF powers-that-be tomorrow. Any comments welcome.

I noticed that the current tentative schedule calls for RR1 to be 72 boards (81 appear on the schedule, but each team has a sitout for one of the nine 9-board matches.) Given that the conditions of contest call for “approximately 60 boards per day, erring on the low side,” 72 boards for a 1.5-day round-robin seems low. 10-board matches, 80 total, would be closer to ideal. One constraint which must be entering into this is that the sitout round takes time out of the day. With mostly 6-baggers, I doubt teams are clamoring for a sitout in the first stage. I want to suggest an 8-round movement without sitouts, which I have tried to design to function as smoothly as possible. (I did combinatorics research at one time and enjoy thinking about bridge movements. I hear there is a treatment for this.) The idea is that at any time you have 3 teams in a 3-way and the rest in 2-ways. Here is the movement:

Round 1: 1-2-3 4-8 5-7 6-9

Round 2: 1-2-3 4-9 5-8 6-7

Round 3: 4-5-6 1-8 2-7 3-9

Round 4: 4-5-6 1-9 2-8 3-7

Round 5: 7-8-9 1-5 2-4 3-6

Round 6: 7-8-9 1-6 2-5 3-4

Round 7: 1-4-7 2-6 3-8 5-9

Round 8: 1-4-7 2-9 3-5 6-8

Comments:
1. Three-ways create a slight security issue, because boards are played out of order. But with only one in each division at a time, the 3-ways can all be put in closed rooms, with the NSs expected not to leave their room for the 2 rounds and the EWs monitored as they switch rooms. Substitutions between halves of a 3-way are problematic and should probably be prohibited.

2. I recommend no comparisons be allowed between rounds 7 and 8. Comparisons can be allowed after other odd rounds if the 3-way teams are sequestered. The only teams with 2 3-ways have their second one at the end, and they could complain if everyone but them knows where they stand after 7. It’s probably best regardless of the movement, anyway, to disallow comparisons just before the final round.

3. This movement makes it impossible for all teams to play identical boards, to a slightly greater extent than do sitouts. I have tried to mitigate this issue, but a 3-way inevitably requires 3 sets of boards, so one of these must differ from the 2 sets played during the 3-way by the rest of the field. This mean 4 of the 36 matches will involve “odd boards,” which seems minor. The movement uses 12 sets of boards, with 8 being used in 4 matches each and 4 in 1 match each. Any pair of teams will play 6/8 or 7/8 sets of boards in common, while in the sit-out movement it would be always 7/8.

I believe that under this movement, 10-board rounds can be played in a reasonable time-frame. Only the first day might be long, with 6 10-board sets, but eliminating breaks after odd rounds would deal with this and doesn’t seem too onerous; many tables finish early anyway. Teams could be told “You can compare after odd rounds, but there is no extra time allotted for this.” The second morning would be very quick, 20 boards with no break, allowing plenty of time to transition into RR2. RR2, with an even number of teams, and the KOs are boring for the movementologist, but who knows, the bridge might be interesting.

Random seeding gripe

In the ACBL seeding system, points for winning or placing in a major event decay arithmetically. So, you get 11-n points for winning the Spingold n years ago. Suppose "Bob" won in 2000 and 2001, and "Zia" won in 2003. As of 2004, Bob gets more points than Zia, 15 to 10, which seems fair. But when we get to 2010, suddenly Zia's win is worth more than Bob's two, 4 to 3. To put it differently, it looks crazy that a win 9 years ago is twice as good as a win 10 years ago, while wins initially decay only mildly, 10% a year.

I suggest an exponential decay factor, maybe .9. This would avoid these odd reversals. Given my (lack of) record in major events, my gripe has no self-interest component.

Foul trouble

Most game-players are sports fans, so I'm throwing in a little essay I wrote on foul trouble. It was targeted for an econ blog, and uses one or two technical terms, but they aren't central.

----------

In a professional basketball game, a player is disqualified (“fouls out”) if he is charged with 6 personal fouls. Observers of the NBA know that the direct effect of fouling out actually has less impact than the indirect effect of “foul trouble.” That is, if a player has a dangerous number of fouls, the coach will voluntarily bench him for part of the game, to lessen the chance of fouling out. Coaches seem to roughly use the rule of thumb that a player with n fouls should sit until n/6 of the game has passed. Allowing a player to play with 3 fouls in the first half is a particular taboo. On rare occasions when this taboo is broken, the announcers will invariably say something like, “They’re taking a big risk here; you really don’t want him to get his 4^th.”

Is the rule of thumb reasonable? No! First let’s consider a simple baseline model: Suppose I simply want to maximize the number of minutes my star player is in the game. When should I risk putting him back in the game after his nth foul? The phrasing is deceptive, because I shouldn’t bench him at all! Those of you who haven’t been brainwashed by the conventional wisdom on “foul trouble” probably find this obvious. The proof is simple: if he sits, the only thing that has changed when he gets back in is that there is less time left in the game, so his expected minutes have clearly gone down (in fact the new distribution on minutes is first-order stochastically dominated, being just a truncation.)

OK, while I believe the above argument is very relevant, it oversimplified the objective function, which in practice is not simply to maximize minutes. I’ll discuss caveats now, but please note, there is tremendous value in understanding the baseline case. It teaches that we should pay attention to foul trouble only insofar as our objective is not to maximize minutes. I am very comfortable asserting that coaches don’t understand this!

First caveat: players are more effective when rested. In fact, top stars normally play about 40 of 48 minutes. If it becomes likely that a player will be limited to 30-35 minutes by fouling out, we may be better off loading those minutes further towards the end of the game to maximize his efficiency. Notice, though, that this doesn’t lead to anything resembling the n/6 rule of thumb. It says we should put him back in, at the very latest, when he is fully rested, and this isn’t close to what is done in practice. In fact players often sit so long the rest may have a negative impact, putting them “out of the flow of the game.”

Second caveat: maybe not all minutes are created equal. It may be particularly important to have star players available at the end of the game. On a practical level, the final minute certainly has more possessions than a typical minute, but it also has more fouls, so maybe those effects cancel out. I think the primary issue is more psychological: there is a strong perception that you need to lean more on your superstars at the end of the game. I think this issue is drastically overrated, partly because it’s easy to remember losing in the last minute when a key player has fouled out, but a more silent poison when you lose because you were down going into that minute having rested him too long. By the way, my subjective sense is that the last possession is more similar to any other than conventional wisdom suggests: a wide-open John Paxson or Steve Kerr is a better bet than a double-teamed Michael Jordan any time in the game. On a couple of major occasions, Jordan agreed. This isn’t to underestimate the star’s importance in scoring and getting other players good shots, just to say that this is not necessarily more important in the final minutes. You do often hear that a team will rise to the occasion when a star is injured or suspended, so even conventional wisdom wavers here. Finally, note that the foul-trouble rule of thumb is applied also to players who aren’t the primary scorer, so that this argument wouldn’t seem to apply. I will give coaches a little credit: they do sometimes seem to realize that they shouldn’t worry about foul trouble for bench players who often don’t play at the end anyway.

One more psychological caveat: a player who just picked up a foul he thinks is unfair may be distracted and not have his head in the game immediately afterward. This may warrant a brief rest.

Final note: Conventional wisdom seems to regard foul management as a risk vs. safety decision. You will constantly hear something like, “a big decision here, whether to risk putting Duncan back in with 4 fouls.” This is completely the wrong lens for the problem, since the “risky”* strategy is, with the caveats mentioned, all upside! Coaches dramatically underrate the “risk” of falling behind, or losing a lead, by sitting a star for too long. To make it as stark as possible, observe that the coach is voluntarily imposing the penalty that he is trying to avoid, namely his player being taken out of the game! The most egregious cases are when a player sits even though his team is significantly behind. I almost feel as though the coach prefers the certainty of losing to the “risk” of the player fouling out. There may be a “control fallacy” here: it just feels worse for the coach to have a player disqualified than to voluntarily bench him, even if the result is the same. Also, there is a bit of an agency/perception problem: the coach is trying to maximize keeping his job as well as winning, which makes him lean towards orthodoxy.

There are well-documented cases in the last decade of sports moving towards a more quantitative approach, so maybe there is hope for basketball strategy to change. The foul-trouble orthodoxy is deeply ingrained, and it would be a satisfying blow for rationality to see it overturned.

*Final outcomes are binary, so the classical sense of risk aversion, involving a concave utility function in money, doesn’t apply at all. But there is also a sense of what I call “tactical risk”: a decision may affect the variance of some variable on which your probability of final success depends in a convex (or concave) way. I might write an essay sometime on the different meanings of “risk.” Anyway, here you would presumably should be risk-averse in your star’s minutes if ahead, risk-loving if behind. But this is rendered utterly moot by first-order stochastic dominance!

The dog that didn't ask

A friend of mine got “accidentally jobbed” by “accidental UI” in a very simple auction last week. How do you think a director or committee would treat this?

1S-P-3C(nat, inv, alerted)-X-All pass

Now, clearly X here is takeout, right? Well…except the doubler never asked what 3C was. He had a 5-card club stack and assumed 3C, alerted, was Bergen (and doubled rather quickly, I gather). His partner knew the X was penalty and passed with a club void, because *doubler never asked*. Doubler’s partner never asked either, and probably also thought 3C was Bergen and passed in complete innocence.

It’s clearly wrong for this to stand, because, however far this was from the players’ intentions, it gives them the ability to make takeout or penalty doubles in the same auction. If you agree, would you trust a director or committee to get this right? (I wouldn’t.)

BTW, it is impossible to avoid conveying UI on this hand, unless you follow the “always ask regardless of your hand” principle, which very few do. If the club stack asks, is told 3C is natural, and passes, the message is clear unless he is an always-asker. That particular UI would have been fairly innocuous on this deal, but the problem remains.

I guess if I were arguing this hand with a committee, one of the strongest points is that doubler ignored the skip bid.