Join Bridge Winners
All comments by Craig Zastera
You are ignoring the author of this comment. Click to temporarily show the comment.
Michael,
I think the points you make are valid which is probably why playing NF new suit advances (or at least some of
them NF) is probably the more common approach now-adays.

But there are advantages in playing them forcing when
(a) advancer is not a PH
and
(b) opening bidder's partner passes partner's overcall

Besides the obvious benefits with a goodish hand that can't be fully described in one bid, there are also the following:
1. new suit jump shift advances can be played as
“fit-showing”.
If the simple new suit advance is NF, then the
new suit jump advance is needed as natural with a
better hand (e.g. “invitational” as in BWS)

Fit showing jumps can be big winners when they
occur, so having them available can sometimes be
very useful.

2. If playing new suit advances as forcing, then the
cue-bid advance can be played as 100% guaranteeing
a good hand WITH A FIT for overcaller's suit.

When new suit advances are NF, the cue-bid becomes
“overworked” as it is then advancer's only forcing
call. Hence, it cannot promise a fit for
overcaller's suit. It may instead be any of a
variety of strong hands that can't afford to make
a NF advance. This makes it much harder for
overcaller to re-bid after the cue-bid because he
literally has no idea what sort of hand (i.e. what
shapes) his partner holds.
Feb. 12
You are ignoring the author of this comment. Click to temporarily show the comment.
Steve,
But if you believe this is “so very close” (presumably between PASS and 3NT?), then doesn't that make the “3rd option” the most appealing?

The 3rd option would be 3. This neither commits us to game nor to 2NT while simultaneously bidding out our pattern so that partner might be able to make a good decision.

In my view, this 3 is 100% forcing (could be up to 18 with the actual hand being an absolute minimum for this call) on partner but not 100% forcing to game.

If partner continues with e.g. 3 or 3, we can (and with this hand I think *should*) pass.

If partner has s adequately stopped opposite our announced shortness, he can bid 3NT (even with a minimum for his invite).

If he judges his holding inadequate for 3NT but otherwise thinks he has a max hand, he can try other calls, e.g. 3 or 4 or ..
Feb. 11
You are ignoring the author of this comment. Click to temporarily show the comment.
Let's assume we have 4 available as “reverse Baze” to agree s and indicate (some) slam interest but with no shortness.

If partner rebids 4 (artificial “Last Train”), we bid 4 and likely stop there (if partner goes on on his own, see below).

If partner bids 4 over 4, we obviously pass.

If partner bids 4 “Kickback” over 4, then he must have a good slammish hand with a control in all the side suits.
So we tell him about our 1 keycard (and, if he asks, lack of the Q).

If he bids 6 after that, then we know our side is missing at most one of A, K, Q, A, A *and* that partner has either the A or K.

How good is 6N under such conditions?
A 1000 deal simulation had 6N making (only?) 58.3% of the time. Subtract a couple of percent for “real world” declarer play, and we have 6N coming in only in the 56% or so range.
Still, that is marginally good enough to bid.

Our weak holding might cause problems, though.
After 4-4-4, if he lacks control but has a good hand, he might try 5 and we will have to play 5.

Or, if he has the minor suit aces and a control, he might bid Kickback and find we're off 2 keys. Again, we have to play 5.

If we have to play 5, how much risk is there of going down?
I tried a 1000 deal simulation where opener had 16-17 HCPs, both minor aces, but no control.
5 was down on 13.8% of these.

I also tried 1000 deals with North having all side suits controlled but where we are missing 2+ key-cards (including Q as a key). Here, 5N was down 11.7% of the time.

Keeping in mind that the cases where we have to play 5 are relatively uncommon, it appears to me that it might just be justified to try 4, reaching 6 only if partner takes over with key-card and finds our side holding enough despite the occasional 5-1 that will result.
Feb. 11
You are ignoring the author of this comment. Click to temporarily show the comment.
I think what I wrote is directly relevent to OP auction.
If West's immediate 2 advance were forcing, then his immediate 3 jump advance would be a fit-jump.

So if he had 6 s and no fit, he would presumably bid 2 whether it were forcing or not (unless his hand was so weak that PASS seemed better than a forcing 2).

Sure, it is possible that with some such hands he would *rather* be playing that 2 was non-forcing, but the dilemma of how to play new-suit advances of overcalls is well known, as is the fact that there is no solution that works well all of the time (I like transfer advances with new suits *below* the cue-bid forcing, so on OP auction with a suit this would be no different for advancer than just playing “new suit advances forcing”).
Feb. 11
You are ignoring the author of this comment. Click to temporarily show the comment.
Mike,
I think you've got this exactly backwards.

Playing the 2 advance as forcing is actually a boon here because in that case a jump-shift advance to 3 is 100% unambiguosly a “fit jump.”

It is those who play that 2 advance as NF who (probably) cannot afford to define the 3 jump as a (very useful) fit jump because in that style 3 is probably needed as natural and invitational (or perhaps some play it as natural and forcing).

In fact, the ability to play jump-shift advances as “fit-jumps” is quite possibly the single most important argument in favor of playing “forcing” new suit advances of overcalls.

When advancer is a passed hand and (probably) when opener's partner does something other than pass, it is still likely best to play 3-level new suit advances as “fit-showing”.

BTW, as I mentioned in a recent post on another thread, quoting at length from Robson/Segal, when 3rd hand *raises* opener's 1 of a suit to 2 of the same suit with partner having overcalled (at 1 or 2 level, non-jump) in between, advancer's 3-level new suit (non-jump) should probably be fit-showing whether or not he is a passed hand.

And when partner's overcall has been at the 2-level (i.e. non-jump in a suit lower ranking than opener's), then advancer's 3-level new suit non-jump advance should also be fit-showing if responder has made a “negative double” as well as when he has raised opener (again, even if advancer is not a PH).
Feb. 11
You are ignoring the author of this comment. Click to temporarily show the comment.
6 is poor.

After they take their A at trick 1 (I think a lead will be obvious on any reasonable auction), to make 6 you need to bring in the suit without loss.

The percentage play in the suit is a 1st round finesse against East's hoped for Q.

This will succeed (only) when East has Q, Qx, or Qxx.
The combined probability for one of these favorable lies of the suit is 36.74%.

Of course, you could alternatively bang down AK. That would succeed if Q is stiff or doubleton in either hand.
But the combined probabilities for one of those is only 32.78%, so that line gives even a poorer chance.
Feb. 11
Craig Zastera edited this comment Feb. 11
You are ignoring the author of this comment. Click to temporarily show the comment.
If you want 3 here to be natural (6 s, etc.), then I think you would need to define 3 as a strong game invite.

That is probably not unreasonable since if advancer had s he could have advanced 2 instead of 1NT originally.

I do think it is important for advancer to have some way to show a fit with a hand too good for a mere raise to 3 but not quite enough for 4. This is because there is quite a bit of ambiguity w.r.t. strength both for overcaller's 1 then 2 rebid and for advancer's initial 1NT advance.

IMO having a strong invite available here is more important than having a natural 3, so without discussion I would not take 3 here as natural. But using 3 for the invite seems like a good idea if the partnership has actually discussed such things.
Feb. 11
You are ignoring the author of this comment. Click to temporarily show the comment.
Just as “2/1 game force” does not have to be played as 100% GF in all auctions (i.e. it is perfectly reasonable to have auctions that can stop at 4m), so here it is not necessary to play that 1NT-3/3 has to be 100% GF in all cases.

Opener's first responsibility is to bid 3NT if he has sufficient stopper(s) in responder's short suit to expect 3NT to be good opposite minimum responder (say 10 HCPs).

But once opener decides that 3NT is not his call, he doesn't have to default to 4m on all hands.

Since there is quite a bit of ambiguity about responder's strength, opener, when not bidding 3NT, should attempt to differentiate minimum hands (say 15 with some wasted in the major) from really nice hands for play in the minor suit.

With that thought in mind, we could define opener's 4 and 4 rebids as suggesting play in that suit (so 3NT is out) but with just a minimum hand and/or not particularly great for the minor.

After such a rebid, responder (who, because he is looking at his hand should have a pretty good idea whether his 3M response was dead minimum or if he has “extras” such that even slam might be in the picture) is allowed to PASS 4m with a minimum hand.

But when opener, still with a hand not suitable for 3NT, has a nice hand for play in a minor such that he doesn't want to allow the responder the option of quitting at 4m, he can choose some call that is neither 3NT nor 4m.

Each partnership can decide the exact definitions for all opener's rebids that are not 3NT nor 4m.
There is certainly room for creative definitions for all of 4, 4, 4NT, 5, and 5 as well as 3 after 1N-3 to increase the partnership's chances of finding the right strain and level.
Feb. 11
Craig Zastera edited this comment Feb. 11
You are ignoring the author of this comment. Click to temporarily show the comment.
I think East's 1 and West's 1NT were both fine (even if West had had a natural 2NT available).

East's 2 was clear-cut.

But now I think West erred with 3. His hand is too good for this. He would raise to 3 with almost any hand with 4 card support (and strong enough for his previous 1NT).

Thus, his choices are between 4 or 3. I actually slightly prefer the latter.
When East continues with 3NT, West now removes to 4.
Feb. 11
You are ignoring the author of this comment. Click to temporarily show the comment.
IMO it is not beyond comprehension. Just not the call I would have chosen (1 seems clear to me).
Feb. 11
You are ignoring the author of this comment. Click to temporarily show the comment.
2NT Lebensohl to be followed by 3NT to show four s with s stopped (i.e. choice of games: 3NT or 4).

This is a fairly standard treatment after partner doubles their weak 2 (e.g. see http://kwbridge.com/leb.htm).

Play the same way after (1)-DBL-(2).
Feb. 10
You are ignoring the author of this comment. Click to temporarily show the comment.
I can't imagine why you would want to bid 5 unless you are trying to take a speculative sacrifice vs. their imagined game.

But why should they have a game? And if they do, you can always think about 5 later.

Surely you can't be bidding 5 because you think it likely to make? You have only 8 tricks, so would need 3 from partner (and your void means that they had best be in the other suits).

I think even bidding 4 here is aggressive because partner will expect a better hand and may raise to 5 expecting to make it and be disappointed. And let's hope he doesn't bid 4.

You actually made my life easier on this problem by specifying “no non-Leaping Michaels”. I actually play that convention, so would not have a natural 4 available.
In that case, I think I would pass this hand and await developments.
Feb. 10
You are ignoring the author of this comment. Click to temporarily show the comment.
Dave,
I think it is a trade-off.

Rebidding 2 with 5=3=3=2 so that a 2 rebid always delivers four s gains when opener can rebid 2 (more specificity about his length).

But that choice loses when he rebids 2 because responder with some length has to worry about opener's having only a doubleton , whereas with the (more) standard style he can be sure opener has at least 3 s (so a contract should be playable whenever responder has 4+ and definitely a good fit when he has 5).

I think aficionados of “BART” tend to favor the more frequent 2 rebid (i.e. with 5=3=3=2 as well as 5=3=2=3) so that opportunities to use their “BART” toy will be more frequent.

Personally, I (not a BART player currently) tend to favor “spreading the doubt” more evenly between the 2 and 2 rebids (both always showing at least 3) rather than concentrating it all into the 2 rebid so as to retain more integrity in the 2 rebid.
Even back when I played BART, I did not care to rebid 2 on a doubleton just to increase my BART opportunities.

I certainly cannot offer any objective quantitative evidence for one style over the other–just my gut feel which may (and apparently does) differ from yours.
Feb. 10
You are ignoring the author of this comment. Click to temporarily show the comment.
Michael,
It is quite normal for 1 HCP to make about a 20% change in the “make percentage” for a contract that is near the 50% mark.
I have done many such simulations for 3NT contracts and found that 1 additional HCP generally causes right around a 20% increase in the 3NT make percentage.

For the OP simulation I referenced (1000 deals for each North HCP value), the make percentage for 6S as a function of North's HCPs (10-13) was:

North HCPs 6S make %age
10 24.5%
11 41.3%
12 60.6%
13 79.5%
Feb. 10
You are ignoring the author of this comment. Click to temporarily show the comment.
I think 4+ s are likely but not 100% guaranteed.

With a 5=3=3=2 14 count, opener would likely rebid 2 as that would be too strong to pass a semi-forcing 1NT response.
Feb. 10
You are ignoring the author of this comment. Click to temporarily show the comment.
Without some special agreement to the contrary, the assumption is that opener's primary responsibility is to bid 3NT with s stopped adequately, or 4m (possibly some other call with special agreements, e.g. 4 = 5 s, 4 = very good hand for s, 4!N = very good hand for s) if his holding suggests that 3NT may not be good opposite responder's stiff.

There is nothing about responder's 3 bid that suggest he has any slam interest at all. He might well have barely enough for game (perhaps even a stretch with e.g. a 9 count). He is simply trying to describe his shape (primarily, his stiff which might be a problem for 3NT) so as to find the best game.
Feb. 10
You are ignoring the author of this comment. Click to temporarily show the comment.
Marion,
Yes–I am suggesting carrying over the technique of showing shortness and a major suit fit from the auctions:
1NT-2-2M-3OM
(agrees M, slam interest with unspecified shortness).

So when the major fit is in s, the compressed splinter is 3:
1N-2-2-3
or
1N-2-2-3

It is true (it seems to me) that these sequences do not work quite as well when our fit is in s as when it is in s because (as you note), in the former case the splinter is one step *above* 3 of our suit, whereas in the latter it is one step *below* 3 of our suit, leaving 2 fewer “steps” when we're playing s vs. when we're playing s.

This is true in both cases, so using this treatment after the Jacoby transfers is no worse than after the Stayman-fit sequences.

What to do about the reduced bidding room in the case?
That's up to each partnership. Maybe you give up the void vs. stiff distinction.
Maybe, you use the 3 “blind splinter” as always a stiff and use jumps to 4/4 to show voids.
Feb. 9
You are ignoring the author of this comment. Click to temporarily show the comment.
I think a 1NT balance is a better description than double.
But it is close and double is certainly not terrible.
Feb. 9
You are ignoring the author of this comment. Click to temporarily show the comment.
To answer this question quantitatively, need more info about what partner's 4 splinter shows.

To investigate this, I assumed the following:
* exactly four s
* exactly one (not A, K, Q, or J)
* 3-5 s and 3-5 s

I did simulations with the above constraints for various numbers of HCPs, starting with 10 HCPs and checked how often 6S would make.

When I got to 12 HCPs for North (with all above constraints), I found 6S making on a bit over 60% of the deals.

When North had only 11 HCPs, 6S made on only about 44% of the deals.

Some adjustment needs to be made for real-world results in slam contracts being (on average) a couple of percentage points below double-dummy.

Still, with 6S coming in at 60-61% when North has 12 working HCPs, one would expect the real world make percentage to be well above 50%.

And, in real-life we won't just automatically bid 6 whenever North has 12+ HCPs–we still can check for key-cards, etc. to eliminate some deals with obvious deficiencies.

So what range of hands can North have for his 4?
Of course, he *might* have five s. Or he might have
a void. Both of these might make 6 better than above.
Or, he might have a stiff *honor* which may or may not be useful.

Still, if 12 “working” HCPs (i.e. outside of s), a stiff small and four s is well within expected range for North's 4 splinter, then some slam move is probably justified.

In that case, I think that 4 (last train) would be the appropriate choice so as not to go beyond 4 if partner doesn't wish to. Of course, it is important that partner will read 4 as “Last Train” and not necessarily promising a control.
Feb. 9
You are ignoring the author of this comment. Click to temporarily show the comment.
To be sure, my simulations show that double-dummy South can make 6 when East has all 6 s on over 42% of those deals.

When East has 5 s, 6S is makeable on almost 64% of those deals.

Not quite so good when West has the long s:
When he has 5 s, 6S makeable on ~23.4% of those deals.
When he has all 6 s, 6S is never makeable.
Feb. 9
.

Bottom Home Top