Join Bridge Winners
All comments by Craig Zastera
1 2 3 4 ... 48 49 50 51 52 53 54 55 56 ... 94 95 96 97
You are ignoring the author of this comment. Click to temporarily show the comment.
A two day pairs event (with day 1 qualifying for day 2) would be a dream come true.

In the olden days, these were common at regionals, but I think they have been essentially extinct for a long time.
June 6
You are ignoring the author of this comment. Click to temporarily show the comment.
Richard,
You are preaching to the choir–I said that I like the idea of using competitive 2NT as a LR+ of partner's major, as advocated in Robson's “Partnership Bidding..” book (one of my favorites).

BTW, if we were not a PH, I think Robson's book advocates that 2NT here is a 4+ card LR+ on this auction–he would play 3 as a “fit non-jump” I think, and 2 as a 3 card high card raise.
I don't think that book addresses if this would change when advancer is a PH, so I'm supposing not.

Steve,
To me, a mixed raise *is* a “light high card LR”, i.e. 4+ card trump support with about the same loser count as a LR but fewer than 10 HCPs (again, some 9 HCP hands may be worth a LR).
June 6
You are ignoring the author of this comment. Click to temporarily show the comment.
I certainly would not pass that hand out!

Close between a 4th seat 2 and 1–I think I like 2, but this is a max as it has “hard” values.

I would surely bid game if partner had a Drury response (to 1), so that argues for 1.
Game would have play opposite as little as A and a black king (and 2-3 s), so 2 does risk missing a game, I think.
June 6
Craig Zastera edited this comment June 6
You are ignoring the author of this comment. Click to temporarily show the comment.
Richard,
I think this is just about precise agreements on what is a LR vs. a mixed raise–certainly neither view is wrong. What matters is that you and partner agree.
For me, a LR means 10 HCPs. Sure, I suppose three Kings, four card support, and a stiff might be enough.

So for me, this hand is a near perfect (if also near maximum) “mixed raise.”

For most in USA (I think), 2NT would not be a raise
(I like the idea of using competitive 2NT to show good 4 card raises of partner's major, but I've found it all but impossible to get partners to agree to play this–most feel a natural 2NT is needed).

On OP's auction, there are two “cue-bids” (I use the term with trepidation) available–2 and 3.

I believe a common agreement (again, in US anyway) in such situations is to use the *lower* cue-bid to show a 3 card LR and the higher cue-bid to show a 4+ card LR (even when both cue-bids force to the same level of our suit).
So playing that way, if one wanted to show a strong raise here, 3 would be the way to do it (saying nothing about s).
June 6
You are ignoring the author of this comment. Click to temporarily show the comment.
Since Drury is one of the best conventions there is, I do not see why you would abandon it just because 4th hand stuck in a 1 overcall. Our side still needs to show LR values while staying safely low when partner has opened light (made perhaps even more likely by the overcall).

Having to respond 2 (over (1) overcall) to show a decent hand with support may get us too high, and at the very least wastes bidding space (relative to 2 Drury) that we might be able to exploit in investigating game prospects at a low level.
June 6
You are ignoring the author of this comment. Click to temporarily show the comment.
Steve,
If you are right, I do not find the text to support that position.

What I do find about natural cue-bids w.r.t need to alert is (from the ACBL alert chart):
<alert required for>
“If played as natural,
a direct cue-bid of a natural opening bid”

But the cue-bid we are discussing ((1)-P-(1)-2) is not a direct cue-bid of a natural opening bid–it is a direct cue-bid of a natural response, which seems to me to be different.

Further, in the ACBL Alert Pamphlet, it explicitly says the following:
“The following are not considered cuebids:
◦ 1♦-P-1♥-2♥: If natural or 2-suited takeout, no Alert. ”

So that seems to answer this issue (if in a slightly contradictory way since this statement contradicts the earlier definition of a cue-bid).
June 5
Craig Zastera edited this comment June 6
You are ignoring the author of this comment. Click to temporarily show the comment.
Quoting directly from ACBL “Alert Procedures”:
“Cuebid:
A bid in a suit which an opponent has either
bid naturally or in which he has shown four
or more cards.”

Seems pretty clear (to me anyway) that a “cue-bid” could be a jump (or double jump)–the only requirment seems to be that an opponent has either bid the suit (naturally) or even made a bid that implies 4+ cards in the suit.
June 5
You are ignoring the author of this comment. Click to temporarily show the comment.
Well, I'm not sure I'm supposed to think less of RHO just because he is young, but if I suppose that by “junior” you meant to suggest (besides youth):
* extreme aggressiveness
* relative unconcern with vulnerability

then I guess this would push me towards acting. Thus, double becomes more appealing relative to pass (and I already said I thought this a close choice anyway).

I don't think I'm pushed towards 3–I dislike high-level overcalls on poor suits–I want my partner to feel he can raise with adequate HCP values without to much worry about modest trump “support”, e.g. Hx.

I think bidding 3 here with such a poor suit is too likely to result in rewarding the “Junior” for his perhaps over-aggressive pre-empt.

Besides, one of the ways to win against stereotypical “junior” would be for partner to leave in my TO double.
June 5
You are ignoring the author of this comment. Click to temporarily show the comment.
Close choice.
I think 3 would be terrible, but the choice between DOUBLE vs. PASS seems close to me.

The combination of the extreme disparity of your support among the three unbid suits, the fact that partner is a PH, and the fact that RHO opened (3) VUL vs. not and in 2nd chair (both of which, to me, suggest a sound pre-empt) sway me towards PASS.
But I certainly would not be at all critical if my partner DOUBLED with this hand, even after it worked out poorly.
June 5
You are ignoring the author of this comment. Click to temporarily show the comment.
I'm bidding 3 (mixed raise) whatever (2) meant.

But I'm wondering if, since many (most?) would play this (2) as Drury, whether it might not make sense to inquire (despite the failure to alert) of LHO what their agreements are. Or, perhaps you think you should not inquire if you are likely to make the same call regardless of the answer?
June 5
You are ignoring the author of this comment. Click to temporarily show the comment.
If partnership plays that a jump cue-bid is a mixed raise, then this hand is just about a textbook example of such–less than 10 HCPs (say, 7-9), 4 card support, some shape.
June 5
You are ignoring the author of this comment. Click to temporarily show the comment.
Leaving out all 2=2=4=5 hands is probably a slight error.
But I doubt that this will have huge impact on results because such hands will be but a small fraction of all the 12-14 HCP balanced hands opener might hold.

Secondly, some (many) 2=2=4=5 (and 2=2=5=4) hands *should* be left out because opener will often either open 1 and rebid 2 or open 1 and rebid 2 (depending on which shape and honor structure in the minors).

Only occasionally will hands of these shapes have strong enough major suit doubletons (I would never rebid 1NT with 2=2=(54) shape without at least Kx in the unbid major) for opener to plan to rebid 1NT (I would need at least Qx in both majors, preferably Kx or better) initially, and a few more hands where he might have thought he would rebid 2 but change his mind and rebid 1NT instead when his doubleton in the *unbid major* is strong while the one in responder's major is weak.
Since exact criteria for rebidding 1NT with 2=2=(54) will vary from player to player, I did not want to specify some arbitrary criteria with which all would not agree. Even so, perhaps I should have included some hands of these shapes on which the great majority would choose a 1NT rebid.
June 4
You are ignoring the author of this comment. Click to temporarily show the comment.
Certainly one would like to know if 2 is natural and forcing and how far does the force extend.

But on this particular hand, I don't see the problem as I don't consider our hand strong enough to force to game, so why would I even be considering 2 here (without a suit)?

Just make a nice invitational jump to 3 which won't be misunderstood and is a reasonable description of the hand.
Sure, you're supposed to have 6+ (good) s for this, but AKQJ9 is close enough.
June 4
You are ignoring the author of this comment. Click to temporarily show the comment.
There are many auctions where the issue arises as to whether a player can pass directly over RHO's bid in suit “x” and then later double suit “x” at a higher level and have that double be interpreted as take-out.

There are two schools of thought:
(a) no, he can't. If he couldn't double 1X (or 2X)
directly (for TO), then he cannot logically
be doubling 2X (or 3X) later for TO.
Such a later double must therefore be for penalty.

(b) yes, he can.
He might have appropriate shape for a TO double
but not have enough strength according to
partnership standards for an immediate double.
However, if 2X or 3X comes back around to him in
pass out seat, he can make a balancing TO double.

Particularly at matchpoints, I think the claim that choice (b) is perhaps the more practical is reasonable.

This OP sequence is an example of this. I really don't
think 9 HCPs and very imperfect shape (only 3 cards in the unbid major) is sufficient to double (2) directly.
At least not in my partnerships–if yours have agreements that say this is enough, then I hope you can handle the ensuing auction to reach the right level and strain.

But when (3) comes back around to South, there is an argument that partner must have at least a fair hand since RHO is limited and LHO with a fit did not try for game.
So it seems tempting to say South should be able to double now for take-out indicating a hand with acceptable TO double shape but insufficient HCP strength for a direct double.
Those who choose (a) above (insisting that South's balancing double is for penalties) will be stolen from frequently, I think. Here, they will have to defend (3) which may turn out OK (if others bid games they don't make).

If playing (b), on the OP deal, the auction could reasonably go:
(2)…Pass…(3)…Pass
(Pass)..Double.(Pass)..3
all pass

Here North, even though he has a good hand, advances a mere 3 because he knows his partner was too weak for a direct double, in light of which North's hand doesn't appear game-worthy.

Of course, North might choose a different option, e.g. passing (3X) hoping for +200 vs. a partscore.
Or, with length in both black suits, he might instead try a 4 cue-bid, which likely would result in reaching 5.
June 4
Craig Zastera edited this comment June 4
You are ignoring the author of this comment. Click to temporarily show the comment.
I did a 1000 deal simulation of this hand opposite 12-14 HCP balanced hands with 2-3 s (and 3 s was disallowed if opener had a weak doubleton in a red suit too, as I believe such hands would raise to 2 rather than rebid 1NT), 4+ s, and no more than 3 s (assumption is that 4=4 minor hands open 1).

I did not allow opener to be 2=2=4=5 or to have a stiff .

The results were that either 4 or 4 (sometimes both) made on almost 60% of the deals.
That suggests that even a mere game try on this hand is conservative and responder should not worry about getting to the 3 level in a major as a result of inviting game when partner doesn't accept.
There were only 135 deals where 9 tricks could not be made in either major.
June 3
You are ignoring the author of this comment. Click to temporarily show the comment.
You seem quite confused.
First, you acknowledge DD sometimes favors declarer, sometimes defenders, but “on average” (which, if you think about it means “in the long run over *many* deals”) it evens out.

But then you go on to quote some specific hands (e.g. with 3 two-way finesses) where DD will do better than most real world declarers.

But those are specific hands–not “on average over the long run” which means over thousands of deals.

So appropriate simulations using thousands of deals and DD analysis can give statistically meaningful results even though individual deals in that simulation may return results that differ greatly from what a SD declarer would achieve.

It is known (re Pavlicek) for various contracts by approximately how much real world results differ from DD results *on average over the long run*. So one can use those correction factors to make simulation results even more accurate *on average over many deals.*
June 3
You are ignoring the author of this comment. Click to temporarily show the comment.
Part of the difficulty in answering this problem is that OP has not specified his methods in enough detail.

Few today play “simple” one-way NMF over partner's 1NT rebid.
Hence, we may not know exactly what replies are allowed for opener over such a NMF 2:
(a) can he jump to 3 or 3? 4 or 4?
(b) can he jump to 3NT?

Also, playing “one-way”, would responder's 2 rebid be an absolute sign-off (i.e. “pass or correct”), or is opener allowed/expected to raise or even jump to 3 with a good hand for his previous bidding?
(Playing 2-way, 2 rebid by responder is a sign-off, i.e. pass or correct to 2).

Also, can opener be 2=2=4=5 for this bidding?
Or 1=3=4=5 ?
June 3
You are ignoring the author of this comment. Click to temporarily show the comment.
I play 2 way NMF and disagree that 1-1-1N-2-2-2 “shows” only four s.

While the same auction prelude with 3 instead of 2 for responder's final bid is “strongly invitational” and *guarantees* at least 5-5 majors, the 2 relay followed by only 2 does not deny being 5-5: it is just a milder invitation.

The OP hand strikes me as not good enough for 2-way NMF 2-2 followed by a “strongly invitational” jump to 3 because on that sequence, opener is only supposed to stop below game with a really bad hand, perhaps:
xx-xxx-KQJ-KQJxx

With the OP's hand, game would be poor opposite many decent hands opener might hold, e.g.:
Kxx-Kxx-Jxx-AKxx

so OP's hand, IMO, justifies only a milder invitation using
“2-way” (i.e. 2 relay followed by 2) despite being 5=5. If it were only 5=4 with same honor structure, it wouldn't justify and invite at all.

As others have mentioned, playing only standard NMF makes this hand virtually impossible to bid accurately.

If you risk 2 NMF, you are committed to playing at least 2NT or 3M even when opener's hand is a poor fit.

If you only bid 2 (which I'm presuming is “pass or correct”, but maybe some play that opener is allowed to raise s or jump to 3s over this with a good hand), then you will miss good games when opener has the right
cards (example: KQJ-KQx-xxx-xxxx would make game nearly 100% and that's not even an opening bid).
June 3
Craig Zastera edited this comment June 3
You are ignoring the author of this comment. Click to temporarily show the comment.
Agree that there are some (many) bridge questions that cannot easily be solved automatically with computers currently.

About the most you can use the computer for on such problems is to generate appropriate deals–but then you have to look at them one by one “by hand” and try to estimate what you think would be likely to happen at the table. And obviously that can be a lot of work.

Of course *eventually*, as “AI” progresses, computers will be able to provide useful answers to all bridge problems, but the pace of evolution of bridge software does not appear to be very fast, unfortunately.
One would think that by now at least good *single dummy* solvers would be available that could automatically compare various lines of play in a given contract by generating many lay-outs for the defensive hands and keeping track of how often each possible line succeeds, deeming the one that succeeds most often as the “best” single dummy line.
June 2
You are ignoring the author of this comment. Click to temporarily show the comment.
I like Max's comment.

I think Michael's comment is way too extreme in disparaging double-dummy simulations in general.

Enormous amounts of very useful information can be obtained using double-dummy simulations (e.g. DealMaster Pro, which is all I've got) for appropriately selected problems.

Yes, a lot of care has to be exercised in setting up the conditions. A lot of thought has to be given in advance as to what one is trying to find out and whether the simulation experiment has been properly designed to answer those questions.

As Max says, for appropriate problems the “double-dummy” bias may not be too important for various reasons.

For one thing, both the offense and defense play “perfectly”, so the DD biases (may) tend to cancel out.

Secondly, for some types of simulation (e.g. comparing the relative strengths of hands), the DD bias may be irrelevent since it applies equally to all hands being compared.

Third, the DD bias actually switches between “real world declarer does better than DD results” (generally true for lower level contracts) and “real world declarer does worse then DD results” (generally true at the slam level).

The inflection point is around the 4 or 5 level, so large scale DD simulations (thousands of deals) will probably suffer very little from DD bias, statistically speaking (i.e. the results on individual deals may not reflect reality well, but over thousands of deals the the average results will probably be pretty close to real-world).

Fourth, there have been studies (Pavlicek for one) of how DD play/defense results compare to real-world results, so it is possible to try to compensate for DD bias (average bias over many deals) using this kind of info.
For example, in 3NT contracts, real world declarers tend to outperfrom DD by about 4%. This figure will of course be meaningless on any single deal, but will be a very good “correction factor” when looking at the DD results for thousands of deals played in 3NT.

The problem Pavlicek has undertaken to simulate here strikes me as a very difficult one to attempt to answer using simulations for some of the reasons already mentioned, e.g. difficulty in anticipating the real-world bidding by the opponents.

With only my limited DealMaster Pro, I would not even attempt to try to simulate my way to a meaningful conclusion on this problem, so I admire RP for his efforts.
It appears that he has fancier software than DealMaster Pro, because some of the constraints he puts on defensive hands cannot be specified using DMP as far as I can tell (DMP doesn't even allow specifying “Quick Trick” ranges for each hand–a really annoying limitation which makes it difficult to even describe constraints for opening bids).
June 2
1 2 3 4 ... 48 49 50 51 52 53 54 55 56 ... 94 95 96 97
.

Bottom Home Top