Join Bridge Winners
All comments by Craig Zastera
1 2 3 4 ... 52 53 54 55 56 57 58 59 60 ... 88 89 90 91
You are ignoring the author of this comment. Click to temporarily show the comment.
Daniel,
You must have a strange definition of “top half of his range.”
My definition of the “top half” of a (balanced) 15-17 HCP range is “the (on average) best performing 50% of all the hands in the range”.
Therefore, by definition, if 1NT opener accepts an invite with hands that are “in the top half of his range”, then 50% of all the hands in the 1NT opening range will accept the invite.
So the “top half” of a 15-18 HCP (i.e. 4 point) range is NOT just those hands with 17 or 18 HCPs, as such hands constitute well less than 50% of the 15-18 HCP range.
May 3
You are ignoring the author of this comment. Click to temporarily show the comment.
Easy to argue that this “rule of 17” cannot be correct because it counts 8s, 9s, and 10s equally.

That is, imagine two hands with the same shape and the same aces, kings, queens, jacks, and low spots (2..7).
But one hand has all four eights (and no nines or tens), while the other hand has all four tens (and no eights or nines). It seems clear that the one with all the tens and none of the eights will on average perform better than the one with all the eights and none of the tens (but it would be easy enough to do a simulation to demonstrate this.

I am intrigued, though, by the idea that when we have a balanced hand (no interest in a major) and partner opens 1NT that we ought to dispense with “balanced invites” entirely and, instead, either pass 1NT or raise to 3NT.
Eliminating the need for a “balanced invite” sequence frees such up for alternative uses (and it is easy to think of responding hand types that one might like to be able to show, e.g. specific 5-5s with 6-7 HCPs).

I once tried to do some simulations to investigate how much benefit (if any) accrues from responder's having a “balanced invite” (e.g. 1N-2N) available vs. just deciding himself whether to pass 1NT or bid 3NT. Wasn't able to demonstrate that such invites are “useless”, but I recall that at least in the case of VUL playing IMPs, one seemed to do very nearly as well with responder just making the 1N/3N decision himself.
May 3
You are ignoring the author of this comment. Click to temporarily show the comment.
The definition of “down the middle invites” is that partner will accept half the time.
For a 15-17 1NT, that means he accepts with all 17s and most 16s, since 15 HCPs constitute almost half (44%) of all 15-17 HCP hands when partner has 8-9.
Thus, there is something wrong with your claim that you will play 2NT 121 times and 3NT 79 times–should be 100/100.
May 3
You are ignoring the author of this comment. Click to temporarily show the comment.
What is this discussion about?
4 is obviously a vastly superior contract to 3NT.

4 is essentially 100% (it is 100% if we assume weak 2 opening denies a void).

But 3NT is around 58%.

So any auction that results in a final contract of 3NT is clearly inferior to one that ends in 4. Period.
May 3
You are ignoring the author of this comment. Click to temporarily show the comment.
I never suggested that a 3 opening was (necessarily) a psyche.
Presumably, the partnership has some definition for what a 3 opening shows (likely varying depending on type of scoring, seat position, vulnerability, perhaps quality of opponents,… etc.).
If you open 3 with a hand that conforms to partnership requirements for that bid, then it is a descriptive bid, not a psyche. The partnership may have very loose requirements or not.
But if you open 3 with a hand that does not conform to the partnership agreements for what that call is supposed to show under the current conditions, then that call would be a psyche.
May 2
You are ignoring the author of this comment. Click to temporarily show the comment.
I agree–that this hand may present (later) bidding difficulties should be immediately (i.e. before bidding the first time) apparent. Therefore, opener needs to pick a bidding strategy before making his first bid.
I personally would be content to open 1 and rebid 2 after partner's (forcing) 1NT response.
May 2
You are ignoring the author of this comment. Click to temporarily show the comment.
I disagree strongly with the last comment about invites.
The most effective invitations are those which are accepted about half the time.
If you deviate too much from that guideline, you are greatly diminishing the value of the invite.
If partner accepts most of the time, then you might as well just bid game yourself because realizing benefits from the invite will require a too rare parlay–partner must be very minimum and (therefore) not accept the invite and (b) the game contract must fail (never a sure thing because of uncertainty about fit, enemy card locations, etc.), and © the higher part-score contract resulting from the invite must make (sometimes, e.g. 2NT will be -1).
Conversely, if partner rarely accepts the invite, then you may as well just pass and play a level lower.
May 2
You are ignoring the author of this comment. Click to temporarily show the comment.
Leonard,
2 followe by 3 is, IMO, objectively “better” than jumping directly to 3NT because the former highlights our distribution (stiff heart) and suggests alternative contracts. If partner wants to bid 3NT after that, at least we have given him the picture.

I believe that 2 followed by 3 is *forcing* but not game forcing. If opener, appraised of the heart situation, decides that 3NT is not a great idea *and* he is minimum, he can bid 4 which is not forcing on responder. Responder knows *his* strength, and can decide whether to pass 4 or push on to game.
May 2
You are ignoring the author of this comment. Click to temporarily show the comment.
Brian,
True, but now you're getting into the area of psyching, i.e. making bids that do not represent the true value of your hand in the hopes that misleading opponents will outweigh misleading partner.

Here, I am only trying to answer the question of whether the given hand is worth a 15-17 1NT opener or not.
May 2
You are ignoring the author of this comment. Click to temporarily show the comment.
I don't like the 2NT rebid choice at all–hand is too weak.
I think any of the following are better choices:
(a) open 1NT (15-17)
(b) open 1, rebid 2 over 1NT
© open 1, rebid 2 over 1NT
might even consider:
(d) open 1, rebid 2 over 1NT
May 2
You are ignoring the author of this comment. Click to temporarily show the comment.
good point. Maybe 5 would be enough.
May 2
You are ignoring the author of this comment. Click to temporarily show the comment.
No–a forcing 1NT response to 1M is not a natural bid.
It does not promise a balanced hand.
Also, it covers an extremely wide HCP range–4 or 5 up to (at least) 12 (some play that a forcing 1NT may be bid with even more).

Therefore, opener's 2NT rebid does not simply say “bid 3NT with a max, pass with a min.”
Rather, opener's 2NT rebid describes a balanced 18-19 HCPs.
With 16-17 HCPs, opener typically rebids 2m (may well be a 3 card minor). If responder then preferences back to 2M, opener can continue with a delayed 2NT to show 16-17 balanced.

Opener's 2NT rebid after 1m-1NT is somewhat different, because a 1NT response to 1m does show a (more or less) balanced hand and also has a more narrowly defined point range (6-9, perhaps a very drab 10).
So in this case, one might claim that opener's 2NT is more like what you say–a request for responder to bid 3 with a max or pass with a min. Still, if opener is balanced he probably has 18 HCPs for this call as he would have opened 1NT with 15-17.
May 2
You are ignoring the author of this comment. Click to temporarily show the comment.
Bidding here, especially with methods not well suited to describing this hand, exemplifies a style of bridge that some play, but that I find unappealing.

It is essentially just trying to inject “randomness” into the result *AND* in a situation where the odds are against us.
The vulnerability is wrong. The opponents will be better placed to do the right thing than our side will because RHO has made a narrowly defined bid, so LHO will have a better chance of evaluating his side's offensive and defensive potential than our partner will.

I'm not against occasionally injecting some “randomness” into the auction, thereby forcing the opponents to guess at a high level. But I believe in picking situations where the odds of success are in my favor. This is most definitely not one of them.
May 1
You are ignoring the author of this comment. Click to temporarily show the comment.
Don't see why fact (?) that 2 followed by 3 is forcing should be much of a problem here–many just jumped to 3NT and 2 followed by 3 must be better than that.

Kx-xxx-Kx-AQxxxx opposite might make 6, and partner could easily have more than that.
May 1
You are ignoring the author of this comment. Click to temporarily show the comment.
I don't thinks this is “2NT+.” I think it is “2NT-” since 2NT shows 18-19 balanced (3NT shows a different hand type).

I would bid 2NT with this hand, though.
May 1
You are ignoring the author of this comment. Click to temporarily show the comment.
Ah–Bergen raises.
Those are not very popular in my area (Seattle), but may be more popular in other parts of the US.
My impression, though, is that Bergen raises are not especially popular among the top (US) players, but I may be wrong.
May 1
You are ignoring the author of this comment. Click to temporarily show the comment.
In US, most play 2NT as natural and invitational after
1-(1)-?, with a small (but possibly growing?) minority using 2NT as some kind of strong raise.

I think the majority use 3 and 3 after 1-1-??
as natural, pre-emptive (“weak jump shift in competition”), with an again small (but maybe growing?) minority using these calls as game invitational “fit-showing jumps”, that is, showing length/strength in the minor suit bid together with heart support and strength appropriate for a “light” game if the hands fit well in both hearts and responder's minor.
May 1
You are ignoring the author of this comment. Click to temporarily show the comment.
Interesting.
Seems somewhat surprising to me that around here (US), most people do not play 1M-3M the same way with vs. without interference.
I think most play it as pre-emptive when 2nd hand doubles or overcalls (1), but stronger–either a LR or, for some, a mixed raise–when 2nd hand passes.
May 1
You are ignoring the author of this comment. Click to temporarily show the comment.
Jyri,
Quite correct.
A “complete” evaluation would have to consider how this hand performs in various other scenarios.

For example, heart games when responder has four hearts and, say, 10 HCPs. And spade partials when responder has 5 spades and 4 HCPs, etc., etc.

Also, would have to consider how this hand compares to average 15 HCP balanced hands when responder is in the slam zone.

What I did was just a basic NT strength evaluation for when responder is balanced with minimum game going strength.

Another useful test would be to weaken responding hands slightly to make them just invitational (with the attendent difficulty of deciding just how to specify those–but let's say 8-9 HCPs, balanced, no 8+ card major fit).
How often would 2NT fail with this hand vs. an average 15?

Still, my guess is that the conclusion suggested by my one (important) simulation would carry through–this hand is very borderline between deserving to be downgraded to “just” a 14 HCP balanced hand vs. meriting a (15 HCP) 1NT opening.

I also tried the same simulation reported above but with a slightly weakened version of the OP hand:
QJ-K876-KQJ-K752

This time (5000 deals) the make percentage for 3NT dropped to 32% (1597 / 5000) !!
That this hand performed worse than an average 14 HCP balanced hand dramatically demonstrates the cruicial importance of a couple of decent spot cards when evaluating border-line hands!
May 1
You are ignoring the author of this comment. Click to temporarily show the comment.
Straight hand strength evaluation questions like this are particularly easy to answer using computer simulations.

Note that in answering this type of question, arguments about double dummy vs. “real world” play and defense (generally overblown anyway) are irrelevent since here we are comparing the specific hand's performance vs. “average” hands, both using the same (double dummy) protocol. So any differences between double dummy results vs. real world would apply equally to both.

Anyway, I have a large database of how average, random balanced hands of various strength play in 3NT opposite random balanced hands of various strengths. For these simulations, I specified “no 8+ card major suit fits.”

Anyway, here are a couple of relevent results:
1. Average 14 HCP balanced hand
opposite
Average 10 HCP balanced hand (no 8+ card major fit)
3NT makes: 39.04%

2. Average 15 HCP balanced hand
opposite
Average 10 HCP balanced hand (no 8+ card major fit)
3NT makes: 59.14%

These two results are particularly convenient w.r.t determining whether a particular balanced hand is “worth” a 15-17 1NT opening or not. You will note that the average of the above two is just a little less than 50% (49%).

Thus, to determine if the OP hand is or is not worth 15 HCPs (thus qualifying it for 1NT opening), we merely run
a simulation with it opposite random 10 HCP balanced hands and see how often 3NT makes. If > 50%, then this hand performs closer to an average 15 than to an average 14, hence would be worth 1NT. If < 50%, then hand is closer to an average 14, hence should be opened 1.

I did such a simulation (5000 deals).
3NT made on 2503 of them (50%).
The margin of error is probably 1 - 2%.
Thus, this hand is clearly right on the cusp between behaving like an average balanced 14 HCP vs. an average balanced 15 HCP.

My conclusion, based on this simulation, would be that it is a toss-up (w.r.t playing strength) between 1NT vs. 1 then rebid 1NT.

One might break the tie in various subjective ways–e.g. opening 1NT has more pre-emptive effect.
Conversely, opening 1 will sometimes allow us to discover a 4=4 heart fit when responder has 6-7 HCPs and four hearts whereas opening 1NT would not.
May 1
1 2 3 4 ... 52 53 54 55 56 57 58 59 60 ... 88 89 90 91
.

Bottom Home Top