Join Bridge Winners
All comments by Kit Woolsey
You are ignoring the author of this comment. Click to temporarily show the comment.
Our negative and responsive doubles at the 2-level are normal. We can bid 2 of a major NF (negative free bids) to play there. If we want to play 3 of a suit lower than the enemy suit, we play 2NT is Lebensohl.

Yes, we do miss having a thrump double. Note that the relay double doesn't apply over a 3 call, since there is no suit for us to play at the 3-level. It is a tradeoff. A normal negative or responsive double at that level is vague to begin with, since it isn't clear what the doubler is after. Does he have the other suits, or is he trying to get to 3NT, and where should you put your priorities when responding to the double?
Nov. 10, 2012
You are ignoring the author of this comment. Click to temporarily show the comment.
Obviously I can't go into the kind of analysis at the table that I can when I have time to pour over every detail. At the table I just do the best I can, just as you and everybody else does.
Nov. 4, 2012
You are ignoring the author of this comment. Click to temporarily show the comment.
Got it – thanks
Nov. 4, 2012
You are ignoring the author of this comment. Click to temporarily show the comment.
Thanks for spotting the error. I have corrected it.
Nov. 3, 2012
You are ignoring the author of this comment. Click to temporarily show the comment.
I do not agree with the concept that preempts are less effective when the opponents have an implied (but not announced) fit. When a player has support for his partner he will be more inclined to show that support and less inclined to go after the preempter. Consequently, the preempt is far less likely to go for a number, while retaining the usual advantages of space consumption, description, lead directing, etc. I think this is the best time to preempt.
Nov. 2, 2012
You are ignoring the author of this comment. Click to temporarily show the comment.
Of course it is best. Not only does the artificial sequence give the opponents the option to double, it gives them the option to not double which may help the opening leader if he has a close choice. The last thing you want when you are in a tight 3NT is having them get off to the best lead.

In addition to the clear theoretical advantage, there is the practical advantage that a raise of 1NT to 2NT sounds natural, so there won't be a heat of the battle accident. The 2 followed by 2NT sequence is a bell-ringer, so you won't have a heat of a battle accident – partner will be forced to think.

I have played it as you suggest for years, and have never understood why anybody would play it otherwise.
Oct. 29, 2012
You are ignoring the author of this comment. Click to temporarily show the comment.
Of course world class players often fail to find the best line. Nobody is perfect, or even close to perfect. And the conclusions I reach in my analysis may be wrong also. Bridge is a difficult game.

When choosing which deals to present and how to present them, I try to pick the hand which has the more interesting decisions. It might be my hand, and it might be partner's hand. It is quite common that in my analysis I will conclude that the action taken at the table, either by myself or my partner, was not the best action. It doesn't really matter whether I am showing my hand or my partner's hand.
Oct. 29, 2012
You are ignoring the author of this comment. Click to temporarily show the comment.
Thanks for the catch. N-S vul is correct. I have corrected it.
Oct. 27, 2012
You are ignoring the author of this comment. Click to temporarily show the comment.
Steve,

Of course there are potential ambiguities in any method. But they can be more serious with odd-even discards. If you held A1096532 and wanted to encourage (i.e. not give a suit-preference signal for one of the other suits) you would play the 5 or the 6, choosing the card which, if mis-interpreted, would be least damaging. That will work well most of the time. But if you want to give a suit-preference signal for one of the other suits from your 7-card suit, you will never be playing a card which might be wrongly interpreted as a suit-preference signal for the other suit as might happen with odd-even discards. That is why odd-even discards are suspect. In the situation where they should be showing the greatest advantage (when partner knows you have a large choice of spots), they often come out worse than standard signalling.

On the actual hand, standard players wouldn't be playing 3-way signals since extra length isn't known. While there may be ambiguity anyway, at least they can give the signal they want to give.
Oct. 22, 2012
You are ignoring the author of this comment. Click to temporarily show the comment.
Steve,

Isn't the 10 considered an even spot card? If so, the missing even spot cards from West's point of view are 10862, and partner played the 6. Why do you say this is “obviously” for spades.

Suppose instead East's spades had been A1098765, giving declarer K32, but this time East wanted to suit-preference for diamonds. Would East not discard the 6 of spades, his lowest even spot card? Would you now say “this is the second lowest spot card so it is obviously for diamonds”. It would look exactly the same to West as my original example.

On the actual hand, suppose East had held A94 of hearts and wanted to encourage in hearts, but couldn't afford to discard anything but a heart. Now he would have no choice but to discard the 9 of hearts, and it would look exactly the same to West as on the actual deal.

I do agree that pairs who play odd-even signals often get lazy and just see the odd or even card. If they worked harder, they could determine the problem. But why have defensive agreements which make things more difficult when the gains from the agreements are questionable in the first place?

Oct. 22, 2012
You are ignoring the author of this comment. Click to temporarily show the comment.
I believe that most expert pairs would play that East's first spade card is count. That is the standard default signal for side suits declarer leads. On this hand it is not known whether declarer has 4 or 5 spades, so count would appear initially to be more important than suit-preference.

I agree that East's second spade play should be suit-preference, and that might have helped solve the problem. However, this was available only because East didn't have a potentially significant spade spot. If East had held some potentially significant spade spot, then his second spade play (or even his first spade play) might have been forced.

In Utopia the defense should have gotten this right. West did have enough clues. But a top pair (who have been playing odd-even discards for years so it wasn't a question of unfamiliarity) did get it wrong. We are all human, and if a carding agreement is such that it might lead us astray then that agreement is suspect unless the gains offset the losses. I do not believe that the gains from odd-even discards (if any) come close to offsetting the losses.

The only potential gain I can see from odd-even discards is the ability to give a suit-preference signal in addition to being able to encourage. However, for this to be effective:

1) You must have a choice of even spot cards. If you have only one spot card and it is a low card when you want to suit-preference high (or vice versa) you are either forced to discard an odd spot card (encouraging) or the even spot card and hope partner realizes that it isn't intended as suit-preference.

2) Partner must know you have sufficient length to have a choice of even spot cards. If partner doesn't know that, he won't know whether your even spot card is intended as suit-preference or the only even spot card you have.

If partner does know you have sufficient length to have a wide choice of spot cards, any sensible 3-way structure handles things at least as well. For example, lowest is suit-preference low, highest is suit-preference high, and middle is encouraging (which is probably the most common way to play) works fine. You always have a lowest card. You always have a highest card. You always have something in the middle. Thus, you can always give the signal you intend to give relative to your spot cards.

Here is an example I saw a few years ago:

Partner has shown a 7-card heart suit, and you are void in hearts. You lead a club against 3NT. Dummy has QJ4 of hearts. Declarer wins the club lead and leads a club back at you which you must win. Partner follows to the first club, and, playing odd-even discards, discards the 6 of hearts. Any of 3 defenses (punting with a third club, leading a spade, or leading a diamond) might be right from your point of view. Can you interpret partner's signal? Are you sure?

Well, partner wanted a spade shift, and that was the only winning defense. Partner's hearts were A976532, with declarer holding K108. Could partner have done differently? If you know partner has a 7-card suit and he can't give an unambiguous signal, then maybe something is wrong with your methods. Playing normal methods, there would not have been a problem. Partner would discard the 2 if he wanted a diamond shift, the 9 if he wanted a spade shift, and the 5 or 6 if he didn't want either, and his signal would certainly be readable.
Oct. 21, 2012
You are ignoring the author of this comment. Click to temporarily show the comment.
I'm not going to go into the full details here. The basics of the structure are:

1NT (or 1 over 1) by responder shows 0-5 points, < 4-card support

2 shows 6-8 points, < 3-card support

2 shows 6-8 points, 3-card support

2 (over 1) shows 6-8 points, 5 hearts, 3 spades

Higher bids show various hands with 4+ card support or a long suit in a coded structure

After 2 (the most difficult to handle obviously), 2 by opener is usually some minimal major-minor 2-suiter. Responder can show a doubleton in opener's major, 5 of the other major, or bid naturally without either of these. A 2M rebid by opener is minimal with a long suit. Other calls are generally stronger and game-forcing.
Oct. 20, 2012
You are ignoring the author of this comment. Click to temporarily show the comment.
It is simply a matter of the scoring table. Playing IMPs, slam bidding and getting to the right game are where the big bucks are on constructive auctions. For that reason, you want to establish a game force cheaply and with the best description possible so you can haul out your powerful relay or asking bid structure or whatever you use to maximize your accuracy on the slam and game hands. Sacrificing 1 as an natural positive loses a lot of this accuracy.

When responder has a negative, you will often not have a game and will seldom have a slam. If it is a part-score hand you are happy stopping in any playable contract. There might be a better part-score, but the equity loss of getting to an inferior (but playable) partial is relatively small. If it is a game vs. partial decision you might get it slightly wrong, but as long as you aren't way off the mark the equity loss is relatively small – sometimes bad games make, and sometimes good games go down. Thus, the importance of accuracy when responder has a negative isn't as great as when responder has a positive.

It must be emphasized that I am talking IMPs only. For matchpoints, it is an entirely different story. At matchpoints with each hand counting the same, accuracy in part-score bidding and the game vs. part-score decisions are as important in scoring as slam decisions, and since they occur more frequently it is better to gear your methods to these. So at matchpoints your suggestion of breaking down the ranges of the negative response makes a lot of sense. But at IMPs where the money is on the slams, I think you are giving up too much.
Oct. 20, 2012
You are ignoring the author of this comment. Click to temporarily show the comment.
Any structure will work well on some hands and work badly on others. If you find your structure works well for you, then more power to you. The important things for any structure are that both partners understand the ramifications of the methods, and that there aren't frequent hand types which are unbiddable.
Oct. 18, 2012
You are ignoring the author of this comment. Click to temporarily show the comment.
When the 1 opener has a minimum and responder has less than a positive, you usually won't have a game and you are scrambling to find any playable strain you can find. The first focus is a 7-card fit at the 2-level. We use the artificial 2 call to maximize our chances to find such a fit. If there is no major-suit fit, there will usually be a decent minor-suit fit we can play.

With our structure, if responder is 0-5 and we wind up in notrump responder will be declarer. In theory that is wrong-siding, but sometimes a lead up to a jack or queen is better than a lead up to an ace or king. Obviously there are other approaches which will allow the strong hand to become declarer, but these approaches have sacrifices of their own.
Oct. 14, 2012
You are ignoring the author of this comment. Click to temporarily show the comment.
For the same reason that when playing Standard you don't raise to 2 with 3 spades and a yarb when partner opens 1. Partner needs to know if you are broke or not, so he can make an intelligent decision with his 19-count.

Playing Standard, you can pass partner's opening 1 call with a yarb. Since we play that 1-1;1 is forcing (might be a standard 2 opener) we cannot do that, so we use 1NT to show a very weak hand without 4-card support.
Oct. 14, 2012
You are ignoring the author of this comment. Click to temporarily show the comment.
These are not constraints on the bidding. They are explanations of what partner will expect from various calls, and what calls partner is expected to make with various hands.

Of course the issue is a matter of judgment, not system. But judgment decisions can be meaningful only with understanding of the methods in use.

For example, suppose the expectation were that responder would raise on any hand with 3 spades, even on a zero-count. Surely you would agree that bidding 1 becomes a lot less attractive than if you know partner will never raise without 4-card support. Maybe you would choose the call anyway, but the possibility of getting to a lousy 4-3 fit has to be incorporated into your judgment.
Oct. 14, 2012
You are ignoring the author of this comment. Click to temporarily show the comment.
Probably a heart. It looks safe since North is relatively short in hearts. North bid spades so it is unlikely you will be setting up your spade suit, and if North's spades are strong that might be declarer's source of tricks.
Oct. 13, 2012
You are ignoring the author of this comment. Click to temporarily show the comment.
Perhaps he would. However, you never know whether an opponent might or might not open a preempt. Some players might think the South hand too strong to open a weak 2.

The point is that East had a game plan. Perhaps it wasn't the best game plan, and perhaps the hand he was playing for was unlikely. That could be argued. But there was a real plan.
Oct. 12, 2012
You are ignoring the author of this comment. Click to temporarily show the comment.
The need for murky signals is much greater when employing a 3-way signal, such as odd-even discards or, when playing what Fred and I play, suit-preference at trick 1 vs. suit contracts. The reason is that you are sometimes not dealt the spot which corresponds to the signal you want to give, so you do best by giving the least damaging signal. This is often a signal for the least logical preference. Most of the time partner can work out what you really want by noting which spot cards are missing and deducing your problem.

When giving a 2-way signal, this problem doesn't exist. It is true that you may not be dealt a readable spot card, but you can always give the signal you want to give relative to your spot cards since you are always dealt a highest card and always dealt a lowest card. Thus, there is less need for a murky signal when employing a 2-way signal, although they can be used to advantage as Mike points out.

If there is a technical reason why murky signals are more or less desirable when playing upside-down vs. standard, I do not see it. I think the reason those playing upside-down signals tend not to give mixed messages is that the signals are usually clearer. When discouraging one usually has nothing in the suit, so one can afford to pitch a high spot. When encouraging, every spot may count, so one might not be able to afford a high spot if playing Standard signals. This is the main reason many pairs prefer upside-down signals. Thus, since the signals tend to be clearer those playing upside-down signals don't want to make them murky.
Oct. 2, 2012
.

Bottom Home Top