BBO Discussion Forums: Balancing - BBO Discussion Forums

MPs, all green, how do you rate X?



And if it was IMPs?

I'd have a go, partner is marked with some values and we very likely have an eight card fit somewhere, so don't let them play in their fit at the two level.

"I'd rather be -1100 than -110" - Larry Cohen.

I don't like it, but I'm the short trumps, so I'm the one that is expected to balance.

All red, I might be more concerned about -200. White, I don't have to think.

mycroft, on 2020-November-09, 10:32, said:

Partner can reasonably be anywhere from a 3334 8 count to a 3244 14, I suppose it's an odds game now, simulation might be called for, but I'm not sure the odds are in favour of bidding. Also depends on how much credit I give opps to take 300 if it's available.

Do we play that, if I double, partner can bid 2N with 4-4 minors (or, indeed, any hand with 4-4 in unbid suits)?

This is a common expert treatment...2N as a 'scramble' or a 'grope', looking to avoid a 4-3 fit?

One would absolutely not want to play this in 3C if he holds, as an example, Jxx Qx Axxx Qxxx. Of course, one would prefer not to play the hand at the 3-level at all, but it would be more difficult for them to double 3D than 3C.

Personally, I pass. Move that spade King to clubs, so I was xx K108x K9xx Kxx, and I'd probably come in. More offence, less defence and less likely an opp has a good club holding with which to double 3C.

Don't over-read snippets of wisdom from great players. Mr. Cohen was no doubt exaggerating to make the point that, in his view, most players do not balance enough, but even the most aggressive players settle for -110 (or with luck -170) once in a while.


Edit: speaking of preferring to be -1100 rather than -110, I had that experience in a Bermuda Bowl match once. Different circumstances: we opened a 10-12 1N, the cards lay badly, our run outs got us to the second-best spot, with terrible breaks, and we went for 1100. Our teammates stopped in 3m, scoring 110...could have been 130��. I can assure you that my entire team would much prefer that we’d been -110.

somewhere between aggressive and crazy

X (L-word)

mikeh, on 2020-November-09, 11:30, said:

Of course we have this tool available to try to look for the best (or less bad) combined suit length. Partner is not expected to overfavor H if they have another 4-cd suit. The aim is to find sth playable. And better be +110 than -50!

My argument about the odds is: from experience, when I let them play 2 of a fit, I'm booked for 30%. When it makes exactly 2, it goes lower.

Trading 30% for 5% or 75%, I only have to be right once for every two times it's wrong to break even. All break-evens (they take the push to 3, it makes 140 all day) can be ignored.

And then, the opponents have to make the right guess, and the right guess is never "pass it out" (even if it's the safest, and frequently done).

Cyberyeti, on 2020-November-09, 11:06, said:

I ran a quick and dirty simulation, see below.
It assumes partner will take out to a minor, never NT or pass.
It wrongly includes some hands where partner would interfere in a suit.
It assumes opponents will never double...
But it still should be somewhere in the ball park I guess.
Unless something is wrong (which would be no surprise) the results look very favourable to double, more so than I expected.

#
# "I'd rather be -1100 than -110" - Larry Cohen.
#
predeal south SK5, HKT84, DK952, CT95
E_open1S = hcp(east)>11 and hcp(east)<=19 and spades(east)>=5
E_pass2S = hcp(east)<16
W_raise2S = hcp(west)>=6 and hcp(west)<=9 and spades(west)==3
N_pass = (hearts(north)<3 and hcp(north)<=14) or (hearts(north)==3 and hcp(north)<=8)
N_suit_C = clubs(north) >= diamonds(north)
N_suit_D = diamonds(north) > clubs(north)
produce 100
##### Choose one condition and tweak the suit in tricks and score below
#condition E_open1S and W_raise2S and N_pass and E_pass2S and N_suit_C
condition E_open1S and W_raise2S and N_pass and E_pass2S and N_suit_D
E_tricks= tricks(east,spades)
N_tricks= tricks(north,diamonds)
E_score = -1*(score(nv,x2S,E_tricks))
N_score = score(nv,x3D,N_tricks)
outcome = N_score == E_score ? 0 : (N_score > E_score ? 1 : -1)
##### unquote ONE action below at a time
#action frequency(E_tricks,5,10)
#action frequency(N_tricks,5,10)
action frequency(outcome,-1,1)

TAKEOUT TO DIAMONDS =======================================================
Better score than pass:

Frequency :
   -1	       5
    0	       0
    1	      95
Generated 101393 hands
Produced 100 hands
Initial random seed 1604947334
Time needed   41.502 sec

TAKEOUT TO CLUBS =======================================================
Better score than pass:

Frequency :
   -1	      30
    0	       0
    1	      70
Generated 51202 hands
Produced 100 hands
Initial random seed 1604948313
Time needed   74.662 sec

A 15 count with a spade stop overcalls 1N, a lot of hands with 13-15 overcall 2m with 5 or 6, a number of 33(34) hands will pass with 14 or less. Are you also assuming they never X ?

Cyberyeti, on 2020-November-09, 14:47, said:

Excluded from the calculation, one way or another.

Cyberyeti, on 2020-November-09, 14:47, said:

Wrongly ignored, as mentioned.

Cyberyeti, on 2020-November-09, 14:47, said:

That's in there.

Cyberyeti, on 2020-November-09, 14:47, said:

Yes as said. The 'score' function can't handle double. I can give you the frequencies of number of tricks if you like.

you're also ignoring when they take the push (or, in a really bad case, when they take the push and partner says "I have a little extra, too, 4♠"; and it always rolls. But I try not to play in games where people do that).

And given the term "Operation 'get them to the 3 level' succeeded", that happens - a lot. Frankly, it's the best result (even if it's not the best-scoring result) when you balance.

Am I reading this incorrectly ?

a number of 33(34) hands will pass with 14 or less.

That's in there.

N_pass = (hearts(north)<3 and hcp(north)<=14) or (hearts(north)==3 and hcp(north)<=8)

I think you're excluding 3 hearts and 9-14.

Cyberyeti, on 2020-November-09, 11:06, said:

Cyberyeti, on 2020-November-09, 16:04, said:

It's late , but I think I am not passing with any 4+ hearts or any 15+ or any 3 hearts 9+.
Doubling with 3 hearts and 9+ was an attempt to go close to your requirements above as I read them, my own inclinations are different but non standard.

Waouh, I didn’t think I would lead to so strong sims!

I though (as confirmed by opps who are our usual teammates, and mildly confirmed by partner) that I had been’too agressive but it wasn’t as scores now are 67% for X.

Partner forgot the 2NT 2 places to play and made 10 tricks in 3H with sth almost worth a X over 2S (?) actually, xxx Axxx JTxx AK. With H 3-2, SK offside as expected and DQ onside, and careful play.

The hand could have been a good question on the other side, what do you do over 1S pass 2S?

I think with M swapped I would probably say it is a little exaggerate, but I have 2 chances /3 to land well, including 1 at the 2-level, so I could X over 2H and pray. Over S, always harder.

pescetom, on 2020-November-09, 16:25, said:

You would double with a 3334 13 ? I would not, OK, we disagree.

3343 is even worse because you now play in a 3-3 club fit opposite a 4333 not worth 1N.

X for me. ♠K probably good. This is missing from discussion: east and west are both weak 1♠-2♠ lol so have they the points or distribution to X YOUR contract? only maybe. I would like to know system that E/W are playing, 1♠-2♠ weak or constructive,or have they option to raise ♠ weakly through forcing 1NT??

Knowing there methods would help me decide,but X looks normal not crazy.

Cyberyeti, on 2020-November-09, 18:22, said:

No I would not, but I misread your earlier post as saying that you would - sorry.
I detest doubling without 4 card in oM and my partner knows that, but I also recognise that such rigidity is considered excessive by many here so I was looking for a more orthodox guideline.
As soon as I have time I will rework the double criteria to be more restrictive and take at least the 6-card minor overcalls out of partner's pass.
Then if possible have a look at what happens if they take the push to 3♠.

I would suggest to the OP that you would always want to listen to MikeH. His record as a player is clearly superior to mine, and to most of the posters here.

However, having listened to Mike, I still would double here (with trepidation), at least partly because 2NT by partner is a scramble for us. The reason for that is that a) that 3442 nice 12-count is an auto-pass (I find it much scarier to double with that hand than with the balancer's, especially when opener hasn't passed yet); and b) I don't have stamina to waste on these kinds of borderline decisions; so I let the system carry me here, and save my thinking for when the system can't carry me (which could easily be in the play to 3♦x where I try to take the eighth trick). Is that a weakness? Of course it is. What if the system is wrong? Well, then we change it (and I will then follow that).

I'm better at judgement than I was. But it is still easier for me to memorize system and let it carry me than to have to judgement-think more hands. For others, the line is (sometimes very) different.