November 4, 2014 – How to Play Like Deep Finesse
Deep Finesse, for those of you who don’t know, is a computer program that analyzes how many tricks each side can win in any contract at bridge, for a given hand. The ACBL uses Deep Finesse for its hand records at tournaments, listing the potential winning contracts each way. All of this analysis is double-dummy, the tricks that could be taken knowing the location of all 52 cards. In real bridge, we only see 26 of those cards during the play, and our decisions are never anywhere near perfect.
To play like Deep Finesse, we need to construct the other two hands from the clues available. Bridge is really a pretty easy game, once we can figure out where all the cards are located! The defenders have a real advantage here, since they can use additional clues from the bidding, the opening lead, and partner’s signals. How declarer chooses to attack the hand will also reveal quite a lot about the hand, and good defenders are often defending double-dummy. I heard a wonderful story about the great Italian duo, Belladonna and Garozzo, playing in a US tournament against a low-seeded team. Early in the play, the declarer asked about their signals, and Garozzo replied, “We don’t signal. By trick three, we figure we know more about your hand than you know!”
Of course, declarer can also use many of these clues, watching the signals, and listening to the defender’s bidding. Like Garozzo, by trick three, you should often have a pretty good picture of the unseen hands.
Two back-to-back hands from Sunday reinforced these thoughts. On both, declarer failed, but would have succeeded fairly easily, by reconstructing the other hands from the clues available. Try them both out as play problems:
With North-South vulnerable, North dealt, and the auction proceeded:
West, having doubled you, starts off with the diamond six, to the seven, eight, and your ten. Plan the play, but first, write down all four hands.
All right, let’s see what we know.
- West made a take-out double, so should have support for both spades and diamonds, with a decent hand, maybe ten or more points. Since we have 11 and dummy has 14, there are only 15 points missing, and West will have most of those.
- Over the redouble, East passed, rather than bid one spade. That strongly suggests that East lacks four spades, so West will hold four (or maybe five) spades.
- East doubled 3C, so must have very good clubs. East will certainly hold four, and likely five clubs to the king-jack-and ten.
- If East has four points in clubs, then West must have almost everything else.
- West led a fourth-best diamond six. Using the wonderful Rule of Eleven, we subtract 6 from 11, and get five. That means there are five cards higher than that six in the other three hands, and we have seen every one, our AQ10, East’s 8, and dummy’s 7. West must hold the diamond KJ9.
Putting all of this together, here is my tentative prediction for the full hand:
Now, for the play: We have three sure spade tricks, and the lead has given us two diamond tricks. We will lead up to our club queen, and win two club tricks – that’s seven. And we can lead up to the heart jack and king to develop two more tricks. Game made. The play is very easy – lead a heart to the jack now, and then a club. We’ll get back in with the club queen to lead another heart, and, eventually, return to hand with the spade king to cash our diamond, if needed.
The actual hand was:
With hearts three-three, we’ll land an overtrick, +950. Isn’t that sweet.
By the way, West’s double fared quite poorly, but I approve. If East holds as much as the diamond ten, the diamonds will develop, and the heart AQ will often be sitting over the king. The defenders will have six tricks set up well before declarer can develop much of anything. Here, the heart king showed up in dummy – unlucky, and partner couldn’t produce anything better than the diamond eight – really unlucky.
Actually, the double worked out quite well, because declarer “mis-guessed” hearts, running the heart nine to the ten. That sort of guess would be quite understandable had the opponents never bid, but, here, the heart honors were marked, and there was no guess to this at all.
Then, we had:
Here, both sides were vulnerable, and the bidding went:
West led a fourth-best heart seven. Contracts like this are always very tricky and delicate, and the best plan is never very clear. Here, declarer started well by ducking the heart, and East played the king on dummy’s three. Interesting! East shifted to the diamond three, and South put in the queen, which held, with West playing the nine. So far, so good. What do you know about the hand? How would you play from here?
- First off, West must have the heart queen. Why? Well, West led the heart seven, fourth best, and, using that Rule again, East will hold one more high heart. If that heart is the queen, then West started with the 10987 in hearts, yet led the seven. No way.
- How are the hearts divided? Well, West jumped to two hearts, so will hold at least four hearts, and East made a take-out double, promising support for both majors. So hearts are either 5-3 or 4-4. West appears to have led the seven from something like Q1087 or Q1087x, and East failed to insert the nine. Also, West might have overcalled 1H the first time with five of them, so I am going to assume that hearts are 4-4, not 5-3.
- What is going on in diamonds? That is less clear, but East shifted to a low diamond, so should have three or four diamonds. If four, then West started with a singleton nine, yet didn’t lead it, and East shifted to the diamond three holding KJ103, also unlikely. At this point, I am inclined to place East with three diamonds to the K10 or KJ.
- Who has the club ace? Still unclear, but East is a passed hand, and has already shown up with two kings, and possibly another jack. It feels like West will hold the club ace.
- Who has the spade king? Also unclear. What was this jump in hearts? Was West inviting a game? If so, West will probably hold the spade king. Or was West simply jacking up the bidding? That is my guess, so I’ll place East with the spade king.
There is a lot of guess-work here, but my first stab at the layout is this:
We appear to have two heart winners, two diamond winners, and a club, so we’ll need to win three trump tricks. We’ll try trumping a club or two in dummy, and scoring some trumps in hand. It feels like eight tricks.
I don’t really want to give up my hearts yet, so I’d simply play clubs out of my hand next, and see what happens. Let’s say you lead a club honor next. West wins the ace and plays another heart, maybe trying to talk you out of that finesse, but you know the queen is onside, so you finesse, and discard the diamond six on the heart ace. You play a club from the table, but East ruffs that, and exits with the diamond ten to your ace (and West’s jack). Have you learned anything more?
Sure. Clubs were 6-1. We know that West started with the J9 in diamonds, and at least four hearts, so West can’t possibly hold more than one spade!
Have I said this before? Maybe, but,
COUNT, COUNT, COUNT, COUNT, COUNT!
Here is the ending, and we need to win four more tricks:
We can guarantee those four tricks by leading the spade three next. If West can win an honor, we’ll trump a club with the nine, which East can over-ruff, but we’ll have the rest easily.
It is a bit trickier if East holds all the high spades, making this the ending:
East will win that spade, but then? If East plays a trump, we finesse, and trump a club. East will over-ruff, but our hand will be high. More likely, East will play the diamond king. We trump, and trump a club, overruffed. East can either play a trump, letting us draw trumps, or the heart, where we throw the good club, trump in dummy, and score the spade ace-queen for the last two tricks. You can follow the play here:
Our declarer took a different route, but arrived at this ending. However, South tried to trump a club next. East over-ruffed, and led the diamond king. When West could over-ruff South’s ten, the defenders had three winners, not two. Down one.
Here was the full hand:
The defenders lost their way, but the biggest error occurred at trick one. For starters, holding Q109x, West should lead the ten, not low. However, East also failed to apply that amazing Rule of Eleven: Seven from eleven is four, and East could see four cards higher than the seven. South could not hold a card that would beat the eight (or the seven). Playing the king was a major blunder.
I may be a bit too harsh on East – As Michael Rosenberg pointed out, perhaps West led the heart seven from something like 10762 or 9762. In that case, playing the king won’t cost a trick, and may gain a useful tempo. Since West should always lead the ten from that Q1097, then East made the correct play. West earned most, perhaps all, of the blame.
As an aspiring intermediate, I thoroughly enjoyed this article – thank you for the post.
A somewhat unrelated comment: I prefer Double Dummy Analyser to Deep Finesse. It has a much nicer interface and it is free.
I’ve had Deep Finesse for years, and am used to it, but, of course, the ACBL uses it on their hand records, so others are also used to it.
As a weird aside, I recently ran a hand through it, which Deep Finesse got wrong! In one position, it said declarer would succeed by winning the next eight tricks. After clicking on any card, it now said, the contract can’t be made anymore. I checked with William Bailey, and he said the bug was known, and fixed years ago. Apparently my version is that many years out of date.