Some doubts about fit jumps

[awm]

For several years now I've been playing Robson and Segal style fit-showing jumps. These have certainly gotten us good results on occasion, but I'm starting to doubt that this is the right way to play jump shifts. Comparing fit jumps to natural and weak, the frequency seems pretty similar. I ran the following sim:

 

(1) Opener has 11-15 points, 5+♥, more hearts than spades (i.e. A 1♥ opening in a pretty standard precision style).

(2) Opener's LHO has 10-16 points with 6+♣ (i.e .a 2♣ overcall; I know some hands with five clubs overcall too but it seemed hard to model since many of them would pass or double).

(3) Opener's partner has either: (3a) 5-10 points with 6+♦, at most 2♥, and two of the top three diamonds (3b) 4+♥ and 5+♦ with two of the top three diamonds.

 

The frequencies for (3a) 1.47% and (3b) 1.03%.

 

Now obviously your exact constraints for suit quality on a fit jump or weak jump will be more complex, and probably some hands without two of the top three diamonds will qualify in either case. However, both bids should have a decent diamond suit. Some of the maximum "constructive weak jumps" might actually be too good for the bid (i.e. 7♦ and ten points) but some of the "fit jumps" might also be better suited to make another call (club splinter or fit jump to the four level). Anyway, I don't claim these probabilities are exact or anything, but they do indicate there is not a big frequency difference in favor of fit jumps.

 

Of course, we also have to consider quality of results. And the fit jumps do generally get good results when they come up. However, the weak jump hands are nearly un-biddable when playing fit jumps... whereas the fit jumps can normally make some other sort of raise that is less descriptive about location of values but still gets the basic message of support and strength across.

 

I'm very much considering switching back to weak jumps. Anyone else thought much about this?

[glen]

Read the Bloom/Colchamiro article "The Second Fit" in the December (2011) The Bridge World (in short, SF is better than LOTT), and understand that it is the length in the second suit that does the work. Next, determine the frequency of the fit jump without the 2-of-the-top-3 requirement.

[awm]

Sure, fit jump is more frequent (by about 3:1 ratio) if you have absolutely no requirements on suit quality or overall strength (along with the previously specified pretty strict requirements on the weak jump). Do you really think you will get good results from your fit jumps in this case? My experience has been that fit jumps on lousy suits usually lead to bad results (yes, I have tried it).

 

I'm not going to buy a copy of the bridge world just to read this article.

[Siegmund]

My fit-jumps promise 2 of the top 3, but only 3 trumps. I imagine that too makes them much more frequent than weak jumps.

 

I can remember some number of good games and slams found only because I was playing fit-jumps. I have a hard time remembering many hands where I thought "gee, I wish I had a weak jump available here." There may have been some... but I think the potential for a fit-jump to swing a board to our side is a lot bigger than a WJS's is.

[straube]

I have doubts about it, but I also have doubts about weak jump shifts opposite a major. Maybe this isn't what you were looking for, but I've been thinking of a third way.

 

In our system, we only allow for 1H-2S and 1M-3C as weak jump shifts in uncontested auctions in part because we use other jump shifts to show raises and in part because we seldom have a suit that wants to preempt against a pattern that has 5+ non-fitting cards. After an overcall, I know we're more likely to have a weak jump shift hand, but we're also constrained for space to show fits.

 

I have reservations about the Robson-Segal structure (although we presently use it). As I understand it...

 

cue=3-cd limit raise+

2N-4-cd limit raise+

3M=4-cd mixed raise

FSJs

splinters

 

Double then has to serve as both negative and an invitational hand that would ordinarily make a natural 2N bid. I know 2N isn't a wonderful contract, but it seems like I have lots of hands that have a stopper and invitational strength. The Robson-Segal double feels overloaded. Partner must wonder whether I really have the other major or minors.

 

Also not sure about 3M as mixed. I think most people have altered their structure so that 3M is weak and a mixed raise bids 2 and then competes to 3, but I don't think that's how Robson-Segal initially envisioned it. I think they wanted that the single raise would seldom bid again. Maybe I remember that wrong. Not sure.

 

So one point is that many people who play FSJs are not using the whole of Robson-Segal's structure and then they are frustrated when they don't have sufficient bids to adequately describe their raises. OTOH, Robson-Segal's structure seems to devote almost too much to raises.

 

I've wondered if instead of FSJs (at the 3-level anyway) that one could play some sort of Bergen raises...depending on space. I know that some folks play Bergen after

 

1H (1S)

 

which has the same or more space available. I know a lot of experts seem to frown on this, but then they often get dealt those hands that don't neatly fit into a fit-showing jump and then they wind up bidding 2H, 2S, and 3H a lot.

 

After

 

1H (2D)

 

At the 3-level, there isn't any room for FSJs at the 3-level, but after...

 

1S (2D)

 

there's only one and very valuable 3-level jump shift...3H. I think Robson-Segal assign both 3H and 4H for FSJs. I'm thinking there is a better use.

 

For 1S (2D) I kind of like...

 

dbl-negative

2H-hearts

2S-raise, often 3 trump

2N-natural

3C-clubs

3D-limit+, 3 trump

3H-limit+, 4 trump

3S-mixed raise

 

But even if this is not optimum, my point is that 3H might better be used as a Bergen-style raise. One could play these sorts of bids at the 3-level and FSJs at the 4-level (or 1H (2D) 3S etc).

 

Another point in this direction is that if one has something like the agreement that you have to have KQxxx or AJ9xx in the suit to make a FSJ and you have to have four trump (or three very good trump) and you are barred from having outside kings and queens, but you're allowed outside aces...then in total, you are more likely to have something like Kxxx xx KQxxx xx or xxxx Axx AJ9xx x than xxxx xx KQxxx xx. So lots of the time that one has a halfways decent suit one will have distribution and an outside card such that bidding to the 4-level is not much of a stretch....and at least it prepares partner for a 5-level decision.

[gwnn]

I think if you get that the frequencies of FJS and WJS are similar, you're doing something wrong. WJS always seemed to me to be much more frequent (at least, when I was playing FJS!). Obviously, high frequency alone does not high utility make.

[akhare]

For 1S (2D) I kind of like...

 

dbl-negative

2H-hearts

2S-raise, often 3 trump

2N-natural

3C-clubs

3D-limit+, 3 trump

3H-limit+, 4 trump

3S-mixed raise

 

This is tangential to the OP, but in that case why not play:

 

X: Negative

2♥: NFB

2S: Raise

2N: Natural

3♣: Clubs

3♦: ♥, possible FSJ

3♥: ♠ raise

3♠: ♠ raise

[mike777]

I never understood fit jumps on bbo.......

 

 

 

They seem to play it means

 

 

whatever I say it means

 

if you need rules then you are novice/lousy player....

[benlessard]

The problem with FJS is that it sometimes help the opponent to sacrifice too easily. (1H)--1S--(3m) (FSJ) with shortness in the 2nd suit and a small fit you can get a good sacrifice since you will be favorite to also have a double fit and you know your shortness is not ruffing partner values. FSJ with high requirement and no slam aspiration (or that doesnt set up FP) are simply not worth it IMO. I prefer loose requirements like 3trumps Hxx + a side suit with 2/3 or 3/5. Maybe it should depend on vulnerability.

[Phil]

I have found a fit jump over a minor into a major tends to be pretty pointless in a strong NT structure.

[whereagles]

Fit-jumps appear to better than WJS in theory. But if you only allow for pure fit jumps, as in say

 

1H (dbl) 3C

 

xx

Kxxx

xx

AQJxx

 

then you're gonna wait a LOOOOOONG time to use them. In accordance to it, I play fit-jumps as simply "willing to go to the 3 level, with fit and a side suit". That would mean something like

 

xx

Kxx

xxxx

AQJx

 

certainly qualifies.

[nigel_k]

I tend to agree with Adam. With a FJS, a lot of things have to fall into place for the FJS to lead to a better result than you would have got without it. Obviously when you're writing a book it's not hard to come up with a couple of examples but you may wait a while for one in real life.

 

A WJS seems to me to be more likely to gain when it comes up. I also think they are more common.

 

However, it may be a good compromise to play FSJ at the four level (or only when a double jump) and WJS at the three level. If you have something like xx Kxxx KQxxx xx and it starts 1♥-2♣ then maybe it could be right to sell out to 3♠ or 4♣ but probably it's tactically better to just bid 4♦.

 

Also, I really don't like the idea of FJS on a four card suit. If you have AQJx and partner has Kxx he is going to like that holding but really it is pretty neutral in respect to offensive vs defense. Even Kxxx opposite AQJx is not necessarily anything to get excited about.

[JLOGIC]

I hate fit jumps in most auctions. Weak jumps get a bad rap, they are very useful. Especially in the auction awm mentioned, 1H (2C), I cannot imagine playing fit jumps over weak jumps there. Showing a good long diamond suit without that many points is not only useful, there is no other way to bid it. Showing diamonds with a heart fit is not only not that useful imo, but there are other ways to bid it where you can at least show one good feature, and maybe end up showing both (cuebid to raise, or bid 2D and raise hearts next). I'm not saying fit jumps are useless, but they handle a hand that has many reasonable ways to bid it, and they are not as useful as weak jumps in my opinion.

[JLOGIC]

And frequency is way less important than:

 

A) Is there another bid for this hand type?

B) How useful is it to show this hand type?

 

Weak jump shifts crushes fit jumps in those categories to me. Not to mention that it is more important to actually preempt the opponents when you have a weak hand and a long suit.

[mycroft]

Justin, if I played with you, I'd be happy to play WJS. But as I said in the other thread, most "we play WJS by default" players around here can't wait to use their toy - and that scares me.

 

"it's [] important to preempt the opponents when you have a weak hand and a long suit" is fine, but my partners catch me with a 6-card major and 18 high (and they have a rag doubleton in my suit), or I can't tell if AJx is enough to roll 3NT. Or, they catch me with a flat 13, and nobody can make anything. And we've announced a(t least a partial) misfit, the points are "basically" shared; what to do is pretty highlighted. FJS says "keep bidding, opps, lots of shape", sure, but WJS says "take your plus, opps, everything's breaking badly".

[straube]

I hate fit jumps in most auctions. Weak jumps get a bad rap, they are very useful. Especially in the auction awm mentioned, 1H (2C), I cannot imagine playing fit jumps over weak jumps there. Showing a good long diamond suit without that many points is not only useful, there is no other way to bid it. Showing diamonds with a heart fit is not only not that useful imo, but there are other ways to bid it where you can at least show one good feature, and maybe end up showing both (cuebid to raise, or bid 2D and raise hearts next). I'm not saying fit jumps are useless, but they handle a hand that has many reasonable ways to bid it, and they are not as useful as weak jumps in my opinion.

 

Do you think 2N should be natural or some kind of mixed raise or limit raise+ with four trump a la Robson-Segal?

 

What do you think of FSJs at the 4-level (or 3S when our suit is hearts)? How would you amend the following?

 

1S (2D)

 

.....2H-H, f

.....2S-raise, mixed raise

.....2N-natural?

.....3C-natural

.....3D-LR+, 3 or 4 trump

.....3H-WJS?

.....3S-weak?

.....3N-natural

.....4C-FSJ?

.....4D-splinter

.....4H-FSJ?

[JLOGIC]

Fit jumps to 4 of a minor are fine in competitive auctions imo, mainly because "weak" becomes less useful/frequent. I am also fine just playing splinters though, but if RHO has bid a suit it becomes less likely that you have a splinter in a different suit obviously. I am pretty indifferent about it in that case.

 

I think playing 1S 2D 4H as a fit jump would be extremely horrible though, jumps to game in competitive auctions should not be fit/splinter/exclusion, they should be to play. I would be surprised if many felt otherwise.

 

Regarding all of the various raises, I don't have a strong opinion. I will say that until maybe 2 years ago I had a strong opinion that the world was too raise happy, and that natural 2N bids were really important, but I have come around, mainly because Meckwell play all of those raises, so all of my friends started playing them too because they are epic heroes obv.

 

Currently in that auction with Joe I would play that:

 

2N=4 card limit+

3D=3 card limit+

3H=mixed raise

3S=preemptive.

4C=splinter

 

If the auction was 1S 2H I would play

 

2N=limit+

3H=mixed

3S=preemptive

4x=splinter

 

If the auction was 1S 2C I would play

 

2N=4 card limit+

3C=3 card limit+

3D=weak

3H=mixed raise

3S=preemptive

4m=splinter

 

I don't have a strong opinion on whether all of these raises are necessary, it does seem to help knowing partner has 4+ vs 3+ trumps though, and joes favorite bid seems to be showing a mixed raise so I guess having it available always is important. Missing the natural 2N bid has not yet cost me that I can remember, usually you can double or bid 3N or pass reasonably. It's possible that playing strong club and having a limited 1M opening makes a natural 2N less useful.

[JLOGIC]

 

 

What do you think of FSJs at the 4-level (or 3S when our suit is hearts)? How would you amend the following?

 

 

I do not think that 3S over 1H 2m is analagous to 4 of a minor. I would always want to play 3S as weak there.

[JLOGIC]

It is also possible to me that it's best to play 1H (whatever) 4m as a fit jump, but 1S (whatever) 4m as a splinter, not that I do this with anyone.

[straube]

So the basic idea is that you have WJS to 3D when they overcall clubs and you have WJS when we open hearts and partner has spades. You always have a preemptive raise and a mixed raise (one below 3M) and you sometimes are able to split LR between 3 and 4 trump and sometimes not.

 

So 1H (2D)

 

2N-LR+ 3 or 4 trump

3D-mixed raise

3H-weak raise

3S-WJS

[quiddity]

And frequency is way less important than:

 

A) Is there another bid for this hand type?

B) How useful is it to show this hand type?

 

Weak jump shifts crushes fit jumps in those categories to me. Not to mention that it is more important to actually preempt the opponents when you have a weak hand and a long suit.

 

Playing devil's advocate (since I don't see much love for fit jumps in this thread):

1. Fit jumps tend to help with high-level decisions where more imps are at stake.

2. Re point A, when your side has a fit the opponents often have one too, and they are more likely to interfere with any multi-step method of showing the hand.

3. Re point B, it's not clear to me how useful it is to advertise your holding when the hand frequently belongs to the opponents. Of course it's nice to be able to preempt them but it has to be somewhat less nice now that partner has opened the bidding and RHO has been able to show a suit and some values.

[straube]

In competition with FSJs at 4m, there is Meckwell's approach. Not sure if I remember this right, and they may have changed their treatment but...

 

1M (overcall) 4C=slam try without control of their suit and 4D=slam try with control of their sit (these could be vice versa)

 

1H (bid showing or implying spades) 4C=universal slam try, 4D says I need help deciding what to do if they bid 4S

while 4H says I know what to do if they bid 4S.

[straube]

So how does the calculus change after 1M X? Now we're less likely to want to make a weak jump shift in a suit for which opponent has shown support and more likely to want to make a fit jump.

 

We play transfers and 2N is a LR+ and 3L are FSJs. Would it make sense to sacrifice a 1-under bid for a mixed raise? E.g. 1H dbl 3D would be mixed. Currently, we show a good raise with 2D and we have no mixed raise.

[JLOGIC]

I never learned calculus :P

[PrecisionL]

For those still interested in FIT JUMPS, see Dimiter Ziatanov's article on the Bridge World site:

 

http://www.bridgeworld.com/default.asp?d=editorial_dept/esoterica&f=zlat.html

 

Warning, it is very complicated. Also his Universal Splinter is mentioned.