[quiddity]

---

♥4 led to East's Jack.

This is a matchpoint hand from a thread in the novice forum; I'm moving some discussion here to avoid further distraction in that thread. I found it interesting to compare the possibility of taking a third-round ruffing finesse in hearts vs trying to ruff out East's QJx.

1. E-W were robots. I assume they're probably programmed to always play the lowest of equals as third hand. If they were human, do you think restricted choice applies here or do you think humans would/should always play the jack from QJ? Is this field-dependent?

2. How much of an edge would you need to take the finesse? Again, is it field-dependent?

[smerriman]

quiddity, on 2016-September-16, 15:05, said:

1. E-W were robots. I assume they're probably programmed to always play the lowest of equals as third hand.

Without answering your actual question, according to the system notes, GIB is aware of restricted choice and will randomise its play from an equal doubleton, but will play lowest from equals in other cases.

[MrAce]

quiddity, on 2016-September-16, 15:05, said:

1. E-W were robots. I assume they're probably programmed to always play the lowest of equals as third hand. If they were human, do you think restricted choice applies here or do you think humans would/should always play the jack from QJ? Is this field-dependent?

Due to the blast auction and E having no clue about the heart position, Q from QJx is extremely unlikely. I do not think any good player would do this.

But I have seen many weak players in weak fields who has no clue which card to play from touching honors in 2nd seat in 3rd seat or even which one to lead.

[kuhchung]

Yeah, I wanted to reply that I would never Q from QJx since it feels like it's going to mess up partner all the time. Even if it's right technically, seeing the stiff in dummy with 5 trumps (and is it even right then?) in real life I am going to mindlessly follow the jack.

I would ruffing hook but I don't actually know what's percentage and I don't math

[kenberg]

I was involved in the NB discussion.


Although it was posted there as playing against robots, I think the general discussion was not confined to that. In particluar we spoke some about how to play under the assumption that the lead from either xxxx or Qxxx would be third best. We assumed that this agreement could be trusted.

My thoughts after the lead of the 4 to the J and the K. Top two trump making sure to keep low sots in the board for transportation (this transpotation issue was the point of the original post). Not play K and and T, watching the spots before playing from dummy. Assume, of course, the Q has not appeared. Lho has played 4(the lead) and then spot spot. If the two spots x,y are higher than the 4 then he did not start with Qxy4 and lead third highest. So the Q is on the right. Now suppose W has played 4,3,x where x is one of 678. This is the only problem continuation. Even if W held spots x>y>4>3 he had to play the 3 so as not to give the show away by playing xy.

So it comes down to
Qx43 opposite Jyz
or
xy43 opposite QJz
where x,y,z all come from 6,7,8.

These two seem equally likely with three ways to do either.


Now I do not know whether the bots actually lead 3/5 including from spots.
Gib system notes:

Quote

In suit contracts, GIB's opening lead is frequently a side singleton or doubleton, to try to get a ruff. When it leads a suit bid by the opponents, this is almost always the reason. Read the book Winning Suit Contract Leads for insight on the way GIB leads against suits.

If it leads an honor that's part of a sequence, it uses standard honor leads (K from AKx, A from AK doubleton). If it leads from a long suit, it leads 4th best (but see above: it doesn't always lead its long suit). When leading from 3 small, it leads top of nothing against NT, low against a suit contract.

My choice was to ruff the T. I acknowledge that perhaps E migjt have played the Q from QJx but I did not think it likely enough to sway me.

I do think it is somewhat field dependent, or at least opponent dependent. It has become popular, including at high levels, to lead the second highest from four spats. Hey, I can work that one out.

Side note: At first glance Restricted choice in spots might suggest playing W for the Q. After all, suppose W plays the 7 under the T. If he started with 7643 he had the option of playing the 6 instead of the 7 while if he had the Q743 he had no option. But this is countered by possible restricted Choice for E and it balances out. Easier:to avoid RC: count the ways of putting one of the 678 on the left, together iwith Q43, and the ways of putting two of 876 on the left with 43. Three ways for either.

So I think anyway.

[quiddity]

The question of whether to take the finesse is complicated because there is an additional possible "line": some portion of the field might not get a heart lead or might just woodenly knock out the diamond and always take 11 tricks. The ruffer would beat this field when the queen ruffs out, and otherwise would tie them. The finesse would lose to the field if it loses. That field is good for the ruff and bad for the finesse. So we definitely should not finesse if it's just even money.

I assume there's a point where it must be right to finesse. We could look at the case where the entire field takes the wooden line. The finesser's matchpoint expectation would be x(M) where x is the percentage of hands (among the subset which matter) where the finesse wins and M is the number of tables. The ruffer's expectation would be approximately x(M/2) + (1-x)(M). He gets half a matchpoint per table when the finesse is better and a full matchpoint per table where the ruff is better. Set these to be equal and solve for x.. x = 2/3. So if the finesse is a 2:1 favorite it looks like it must be right to finesse. So I guess the answer to "how much edge do we need" is somewhere between even money and 2:1.

[1eyedjack]

Can't say that I am completely certain about all of the circumstances under which GIB will randomise from QJ, but from experience I believe this to be one of those non-Random occasions where it will always follow with J from that choice. So RC does not apply.

At least, RC does not apply to the 3rd seat Jack.

But it may very well apply to the opening lead.

By which I mean that GIB is somewhat averse to leading from holdings such as Qxxx.

These factors could point more in favour of ruffing out RHO Queen. Whether they tip the scales ...

[kenberg]

I considered this last point, that perhaps a non-heart lead could have been chosen but wasn't, but I did not really know what t make of it. Presumably W does not have Ak or QJ of clubs and probably not JT either. With Qxxx in clubs and Qxxx in hearts maybe he would still prefer the heart. I went around with this for a bit and couldn't really get anywhere with it.

My thoughts were of humans since I rarely play with the bots.

[The_Badger]

hi quiddity,

I looked at this hand and didn't reply to this on the Beginners and Novice's Forum as it was posted for less advanced/expert players to reply to.

However, I personally felt that the hand was misplaced on that forum: in my opinion, it was more of an intermediate/lesser advanced problem.

Unblocking trumps, ruffing suits good, playing for restricted choice, ruffing finesses, etc. are all included: personally I thought unblocking trumps would catch out quite a few intermediate players too.

[kenberg]

T1: Heart to the J and the A.
T2-3: Draw trumps, they are 2-1.
T4. Heart K, W playing the 3, E playing a spot, one of {8,7,6}. Call it x
T5. Heart T, W plays a spot, one of {8,6,5}. Call it y. The remaining cards are the Q and the remaining spot, call it z, from {8,7,6}


Time to consult their cc to find their lead agreements.

If the lead from yz43 would be the lower of yz, we can rule that out.

What do they lead from y43?

And how about Qy43? Do they lead the 4 or the 3?

If they lead the 3 from Qy43 then it seems E has the Q.

There are quite a few variations here, we might hae to call a recess in the play while we work through all of them.

With one partner I have the agreement that we lead second highest from three or more spots. He likes this, I am skeptical. Against a suit contract I then lead the 4 from Q74 and I lead the 4 from 743. But we lead the 3 from Q743. So, if declarer sees me lead the 4 and then follow later with the 3 and the 7, I will not have the Q. . And if I lead the 4 and later follow with the 7 and the 8, he can rule out 874 and 8743 (the lead would be the 7) so he can play me for Q874.

It's the old give and take. We want our leads to be descriptive so that partner and I are on the same wavelength but declarer is listening in and that can cost.


About posting it in the B/N Forum.
The OP wanted to illustrate the necessity of planning, in this case by not blocking the trump suit. My own view is that this is ok for B/N. Early on players can be encouraged to count winners and losers (old joke, if they don't add up to thirteen, then count your cards). Here there are 11 winners after losing to the diamond ace. And two losers, the minor suit aces. But maybe the clubs on the board can all be pitched. If so, we have to watch for transportation.
Sure this is ot the Day 1 lesson, but early on is suitable, at least for the more serious minded B/N. And, presumably, it is the more serious minded B/N who is reading the Forum. But for a discussion of whether to gamble with the ruffing finesse, possibly lowering 650 to 620, I think that's a bit, at least a bit, more advanced and so this reincarnation here seems right.
In my comments on the B/N I did go where the discussion led, but I included a apology/warning that many B/Ns may want to just ignore it.

I am often hesitant to respond on B/N since I am not (I claim) a B/N. It can be a problem but I think if we are all aware of this, and trust the B/N player to skip over something that he feels is more than he is up for, it's ok.

[alok c]

Deleted..

[alok c]

Trump unblocking may not be required when you are assuming East has♥Qxx. Cash ♠A,ruff a♥,come to hand by ♠K, play♥K.