Dirk Kuijt

ey I think I really have only three options here: 1. Go poll at least a handful of real experts. 2. Give my own best opinion. 3. Silence. Choice one will obviously get the best opinion. However, since I have no way of doing that, choice one is equivalent to choice three. Choice two has the obvious flaw that I haven't won the Spingold (or anything in that league, or the next league down or the next one, depending on how you measure leagues), so why should anyone take any notice of my opinion, anyway? That is a fair point, and it is for that reason that I don't post opinions in the advanced and expert forum. That leads us to choice three. Choice three has the obvious flaw that if people posting opinions on the forums are restricted to people who have won the Spingold or equivalent, there aren't going to be many opinions posted, period. So, I tried my best. FWIW, and I understand that you have no way of knowing this, whenever I give an opinion that I know is at variance with general expert opinion (on bridge or any other subject), I label it as such. From the answers to this question, my opinion is at variance with general expert opinion; I promise you I didn't know that at the time.

Speaking for the non 2 heart bidders: 2 hearts would be wonderful, or at least better than 3 hearts, on this hand if the only point of evaluation is play in hearts. However, that's not what 2 hearts means, at least to me. A 2 heart bid, to me, shows some high card values; something that is valuable in 3NT or on defense. This is not 2 hearts, even if it qualifies by the rule of 500 2 98765432 32 32 So, I'm not bidding 2 hearts on this hand; either 3 hearts (which doesn't promise values for anything except play in hearts) or, if the opponents look threatening and your nerve fails, then pass. Of course, in third seat, anything goes, since partner can't have the hand to jump to 3NT or double the opponents in a partscore.

One difficulty of rating the importance of card play for "beginners" is that many people who have never played a hand of bridge have played other trick taking card games: hearts, spades, or pinochle, for instance. So, they are acquainted with the card play ideas (counting suits, finesses, establishing long suits, etc.), perhaps even experts at this. However, they have no idea what Stayman, Blackwood, or takeout doubles are, and the idea that one would bid high (preempt) with a poor hand, but bid low (approach forcing) with a good hand, seems very weird to them.

Bab9 Yes, though it takes a bit of searching. Here are the current results. Note that these don’t relate to the previous run; the program comes up with a new ‘system’ each time: Pass – 1♣: 5+ spades Pass – 1♦: catchall, I couldn’t find a pattern Pass – 1♥: 4+ hearts Others seemed to not occur 1♣-1♦: catchall, I couldn’t find a pattern 1♣-1♥: 6+ spades 1♣-1♠: 5+ hearts 1♣-1NT: normally, both majors 1♣-2♣: seems not to occur 1♣-2♦: normally 3=6=2=2 1♦-1♥: normally balanced 1♦-1♠: 3 spades and a long red suit 1♦-1NT: seems not to occur 1♦-2♣: balanced (only one example) Others seemed not to occur 1♥-1♠: balanced or long spades or long clubs 1♥-1NT: long spades or both minors 1♥-2♣: seems not to occur 1♥-2♦: long diamonds

Nick Yes, my current training takes no notice of level or HCP at all. Obviously unrealistic, but my previous attempts to include them didn't lead to any convergence. I don't mean convergence on anything useful; I simply didn't get convergence at all. So, I decided to try a simpler problem that perhaps I could get some answer. I would definitely like to bias the bidding to lower calls; I've pondered this in the past, but have been unable to find a way to do that. That investigation will also continue. Incidentally, the problem that I set for the computer is not all that easy, given that you are restricted to three bids.

I decided on tackling an easier problem. Rather than trying to get to the right contract, I would try to have the program just get to the right strain (suit or NT). The weights for HCP were all set to zero, and the weight changes were all based on whether the program got to the right suit (or at least an acceptable suit) or not. Specifically: 1. If the contract is in a suit, the suit must be 7 cards combined or more, with penalty weight changes increasing linearly the shorter the suit, and reward weight changes increasing with the length of the suit. 2. If the hands have 9 or more spades, don’t play in hearts unless there at least 8, don’t play in a minor unless there are at least 9, and don’t play NT. 3. If the hands have 9 or more hearts, don’t play in spades unless there at least 8, don’t play in a minor unless there are at least 9, and don’t play NT. 4. If the contract is NT, the shortest suit in the combined hands must have at least 4 cards. The program was permitted to make 3 calls at most, with dealer’s initial pass allowed, but counting as one of the three calls. That is, opening bid, response, opener’s rebid is the longest possible sequence. The initial weights were random, as were the hands, so the results listed below might well not be repeated if the program were rerun (as I plan to do). During the training phase, each hand was bid until the program got it right or it tried 100 times. Also, during the training phase, successive hands differed only by the exchange of one card with one other in the deck. Often, such a change will make no difference in the hands, and it will never make a large difference. During the examination phase, however, hands were (pseudo) randomly dealt fresh for each pair of hands, and the weights were not changed. The opening call may be Pass, or up to 2NT. The response may be up to 3NT, and the rebid may be up to 5D. There were 900,000 hands in the training phase. The results are obviously not optimal in any sense; for one thing the program never opens 2S, or, at least, never did in the 50,000 deals of the examination phase. I have tried to eyeball the results, and can summarize the opening bids of the ‘system’ as follows: Pass: long diamonds, may also have long clubs 1♣: 5 diamonds, 6 or more clubs 1♦: usually 4 spades and 6 clubs, sometimes 4 diamonds and 7 clubs 1♥: exactly 2-1-5-5 1♠: a catchall for normally balanced hands, may be 4-4-4-1 1NT: long hearts 2♣: long spades 2♦: long clubs, may have 4 diamonds 2♥: exactly 4-2-0-7 2♠: unused, as far as I can tell 2NT: 5 hearts and a 5 card minor I recall an earlier poster had noticed a tendency to bid over the promised suit, rather than bidding the promised suit (as natural systems do, by definition) or bid under the promised suit, as in transfer situations. That happens in this set of results as well, with the 1♦, 1♥, 1NT, 2♣, and 2♦ bids, and arguably other cases as well. Why, I have no idea. More to come.

FWIW, I would, nevertheless, bid 1S. Suppose partner has xx QJxx xxx xxxx This hand will not play all that well in Spades. After three rounds of diamonds, you can use your one dummy entry to try to finesse spades (more or less giving up on finding a pitch for the third club), or play AK and another spade. If spades are 3-3 you're golden; if not, you're leaden. The hand doesn't play all that well in hearts, either, after three rounds of diamonds. If you ruff low, and it lives, you don't have a fast entry to the dummy; if you ruff high, you kind of need 3-3 hearts to make game. The problem, as I see it, is that, once you have bid a new suit, showing length and strength, you have canceled the message of the takeout double (that is, that you have support for the unbid suits). You would, or at least I would, double and rebid spades with: AKJxxx xx Ax AJx or, even AKJxxx x Ax AJxx so partner can not count on any heart support. So, even if partner has ♥ QJxxx, he will be wary of bidding hearts again. Tough hand.

Next approach (yes, I'm still at it): First, I'm going to just try to get a system that gets to the right suit, rather than penalizing both for the wrong suit and the wrong level. Second, I'm going to try having the program bid a hand until it gets it right, then modifying the hand *slightly* (at most one card in each hand, perhaps just one card in the 26), and having the program bid again. The idea is to make the program discover that similar hands should be bid in similar ways. We'll see how this turns out. More to come.

It was pointed out that, when I said my experiment was a failure, that really wasn’t true. We learn from failures. Indeed, true, which is actually why I posted it. I rated the results as a failure in the sense that, as far as I could tell, it was bidding at random. Final contracts didn’t bear any relation that I could see to what I was trying to teach, nor to the way it bid similar hands. This was still true even after trials with as many as 1,000,000 hands. My previous approach suffered (I realized in retrospect) from too much linearity, while bridge bidding systems are not linear. As such, even if I set the weights on input, I could not define standard bids. Specifically, five key inputs were HCP, number of spades, hearts, diamonds, and clubs. Consider a standard 1NT (let’s say 15-17, balanced). As the number of HCP goes up, the weight for the 1NT opener line should increase, but, when you get to 18 HCP, the value has to drop to zero. This is pretty much inexpressible in my previous approach. New approach: I now have 9 inputs for ranges of HCP, and 7 ranges for the length of each suit. That makes 37 inputs, which are essentially binary. This at least allows me to express a Standard American 1NT opener. I have also bitten the bullet and now have a set of weights for each possible bidding sequence. Gulp! Yes, that does take a lot of memory. At the moment I’m only dealing with 2 rounds of bidding (4 bids). I limit the opener to calls up to 2NT, the first response to 3NT, and the next two bids to 5D but it is still a lot of memory. At least I can now express a standard Stayman sequence or Jacoby Transfer. Results are still random, much the same as the previous approach. Why, I don’t know.

An update, but not a successful report. I tried a Neural Net with these inputs: HCP # spades # hearts # diamonds # clubs and one input for all previous bids. It was a failure. I couldn't teach the program anything. I'm going to make a next try with: several inputs for HCP, essentially as ranges ditto for suit lengths one input for each possible previous sequence. We'll see how this goes. The number of possible sequences builds up very quickly, of course, so this may be unworkable from that respect.

I've read all 11 pages of this discussion. Fascinating, yet it is clear that there is nothing close to agreement. (I've said before, and want to go on record again here, that my answer to these questions is: no requirement to disclose and no limitation on systems, and that I'm fully aware that this answer is incompatible with the current laws.) However, I would like to repeat a question to Fred that was asked before, but not, as far as I can tell, answered: 1. What is a 'serious' event? 2. How significant a problem is this, anyway? I realize that definitions of "basic sequence" and "frequent misinformation" are inherently vague. Basic in some relay system might go several rounds, simply because the bidding goes more 'slowly'. On the other hand, a second round sequence, especially in competition, may be 'rare'. And frequency is very hard to measure, since we are talking about rare events. If your opponents screw up twice in the eighth round of a Swiss team event, you are likely to conclude that they don't know their system. Of course, you haven't played against them on the previous seven rounds; if they were correct over that period, are they on the wrong side of the law or not? Defining a 'serious' event, however, is a different story. You know, or at least, should be able to know, when you enter an event, what the rules are for that event, before you start. So, I'm going to ask Fred to define, precisely what is a 'serious' event. The other question is even more basic. Is this really a problem, anyway? Of course, in bridge, rare things can be interesting (when was the last time you played a guard squeeze?). But interesting doesn't mean that we need to change rules to handle a super rare case. People that habitually make system mistakes aren't likely, in my humble opinion, to qualify for the semifinals of the Spingold. Are there people at that level making a bunch of system errors? Or is this whole thing just a tempest in a teapot?

I'm working on a neural net attempt at this problem, so I'm bumping this topic. More to follow.

(slight thread hijack) Albert Morehead, many, many, years ago asked this question: which is the better holding? QJT9xx Kxxxxx The answer is: it depends. If partner has xx, then the first holding can always be played for two losers (and easily, by just leading the long suit), while the second always has two losers, and that requires 3-2 with the ace onside, and the ability to lead up to the king, and can easily lose four tricks with a 4-1 break; But, if partner has Axxx, then the first holding still has half a loser (and you need to be able to lead from the QJ), while the second has a loser only 13% of the time, and there are no entry problems in playing the suit. So, I expect the first would be better opposite a 10 count, but the second better opposite a 20 count.

Mark Dean and hrothgar I'm no expert on the laws, but it seems like the only way to really make this law work would be for pairs to just not be able to look at the back of the opponent's convention card. Almost anyone is going to be influenced in style knowing that they opponents were playing Fishbein and penalty doubles of overcalls. Indeed, if I knew that I was sitting down to play Hare Trigger and Back Stabber (who I know for being demon doublers), I'm going to pull in a bit, even if Trigger and Stabber are playing exactly the same conventions as everyone else. As for the law you two have mentioned: which one, specifically, is it? Don't misunderstand; I am not saying you are wrong about this. I started by saying that I wasn't an expert on the laws, and many know them better than I. I'm just trying to learn here.

There was also a vigorous discussion on these forums about this sequence: 1C-(1S)-X-P 2D BWS 2001 says that this can be a minimum. I think I'm right in saying that the consensus on this board in 2010 is that the negative double only shows hearts, and, consequently, 2D is a reverse showing substantial extra values. That is a different kind of TOX that don't promise support.

hrothgar I certainly alter my preempting style depending on what my LHO plays. Is there some specific rule that says that an opening one bid is different?

Quoting Ira Rubin here; Not only do I need a mirror, but I need one behind the opponents as well. Double dummy, and with double dummy agreements with my partner, I would double

2C if this is inverted (which I don't play in compitition). 3C otherwise.

Well, playing "little canape" (and not showing it to best effect), it would go: 1♥-1 2♦ 1♥ is a four card heart suit, which may have a longer second suit. 1♠ over 1 ♥ is a Kaplan interchange, showing, essentially, a forcing 1NT bid without 5 spades. 2♦ is the long suit. At this point, responder has a serious problem: Try to get back to opener's major, get to NT, or just try to stay low on misfits by passing. My tendency is to pass. Yes,I know that neither of us has bid spades in the natural sense, and so staying in a 5-1 can not be optimum. OTOH, we are still low, which has its own advantages.

OK, that is really ROFL.

You are probably going down after this start, but I would bid 4♥. At least you are taking the force in the short hand, unlike playing in 4♠. Unfortunately, you are minimum, and it sounds like partner is also. I admit that I don't understand why partner is supposed to hold only 4 hearts. It seems to me that xxx Axxxx Jx AQx (and several variations including 2-5-3-3) is a perfectly normal 2NT bid. If partner holds the HJ as well, or even instead of the ♦J, then you certainly want to play in hearts.

IIRC, the standard to open in fourth position is 15 Pearson points (HCP + spades). West isn't close.

Let me give you a simple situation. You are West, and South deals, and the opponents have bid the exiting and unusual: (1C)-P-(1H)-P-(1NT)-P-P-P. You have: KTxx xxx Kxx Qxx You lead a spade from KTxx, and the dummy comes down with: xxx Axxx xx Kxxx Declarer wins the Queen, leads a club to the King, and a club to his Jack and your Queen. Your partner's carding shows that opener has 4 clubs. Do you shift to diamonds? Against me, you probably should. I should have 3=3=3=4 for this auction, since I didn't raise hearts, didn't bid spades, would open 1D with 4-4 in the minors and a balanced minimum hand, so I'm likely to be 3=3=3=4. Against Frank Stewart, a diamond shift is much more dangerous, since he, as declarer, might be (23)=4=4, since he opens 1C with a minimum balanced hand. If declarer has AQ (tight) of spades, then a spade continuation will likely beat the hand. If declarer has AQJ of spades, then a spade continuation is, shall we say, "unwise". I'm not enough of an expert to fully analyze this, or almost any situation, but it is clear to me that if you know what declarer has (i.e. declarer's agreements are) you are better off than otherwise, and getting that information is difficult at best, and impossible at worst (the unauthorized information situations are decreased by screens, but not eliminated.) Even if some better analyst than I can show that a diamond shift (or spade continuation) is clearly right regardless of the opponent's agreements, my point is: 1. The defense will do better with full disclosure than without it. 2. Full disclosure is rare, if not impossible. 3. What people think of as full disclosure is far from that; rather it assumes a large body of "standard" agreements, which may not, in fact, be all that "standard". This is, perhaps, acceptable if your opponents do have the same standard agreements, but it violates both the spirit and letter of the law if they don't. (A major problem is, if the opponents don't have the "standard" agreements, then they are unlikely to complain. They are the outsiders, since they are non-standard. Some, of course, will still ask, or complain if the explanation is inadequate, but, in general, outsiders keep their mouths shut. One consequence of this is that one can not conclude that since there are few people who complain that our rules are just fine. No. Outsiders don't complain, in general. They just go away quietly, and don't show up in statistics.)

Obviously I'm talking my book here, but I haven't seen anyone really answer the questions I raised earlier. 1. How do you explain your "style" adequately? 2. How can one ask questions without passing unauthorized information? Various people have pointed out that it is hard, that it is a matter of familiarity, and that it is matter of experience. I agree. But what is the answer, within the current rules? I say there is no answer.

First, sorry not to have answered awm’s question sooner; I have been traveling. Let me try to explain why I think that full disclosure is impossible. 1. As awm hinted at, style matters, and getting this information is hard. Look at the discussion held earlier in this forum about what to open with 4-4 in the minors. Frank Stewart generally favors 1C while I believe I’m right in saying that most of the people on this forum favor 1D. Stewart’s choice is going to mean that 1D will tend to be a better suit that the other choice. Obviously, when RHO opens 1D I *can* ask this question, but I’m not likely to. There are countless examples of this; sequences that may be defined by a practiced partnership (or might not be) whose meaning in terms of strength or distribution is likely to differ from my definition. While this doesn’t affect bidding all that much, simply because one side so often just passes, it will affect the defense. 2. As pointed out in a Bridge World editorial (which I can’t lay my hands on, so this is from memory), there is a serious risk of passing unauthorized information to partner *simply by asking about a bid*. Hence, even if I can get all the information I want perfectly (which I don’t think I can), I can’t ask the question. For example, this sequence, starting with LHO: (1H)-Pass-(2C)-?. Sounds simple, right? But, if this is against an Acol pair, where 2C might be only 8 or 9 HCP, I want my double to be takeout; this might be our hand, especially if we have the spade suit. On the other hand, if this is against Auken-von Arnim, where 2C is a semi-artificial game forcing response, then our side is outgunned in high cards; we can outbid them only with extreme distribution. But, double as a lead director is attractive, since responder doesn’t promise clubs. But, can I ask? Not really. If I have clubs, and want to consider the lead directing double, what will happen if I ask about 2C, and get the answer “long clubs, 10+”. Now, I have to pass; double risks them redoubling 2C and making overtricks. But, partner now knows that I had a problem, and it isn’t going to be hard to figure out what that problem is. Under the current laws, I don’t see a way out of this legal trap. Note that the same situation arises in point 1. If I ask, “what do you open with 4-4 in the minors?” over their 1C bid, partner knows I’ve got something good in diamonds, and partner and I have to spend the dinner break explaining our defense to a committee.