20 May 3/13: make the weaker suit trumps?

[John McLeod]

[Mike Tobias]

I'm always uncomfortable about choosing a 5 card suit over a 6, even in this situation. With 6 trumps, the most likely split against is 3-2 giving 3 surplus trumps, but with 5 it's 4-2 leaving only 1 spare trump which as Middlehand you may need as an entry. The strong spades may not be much use in this case.

I'd dither between spades and grand, possibly keeping the D10,8 in spades hoping for an extra trick.

[John McLeod]

Keeping Middlehand and the Skat the same, I used Kiebitz to deal the other cards 20 more times and analyse the results. I got:
10 deals where all contracts lose
6 deals where spades is the only making contract
2 deals where grand is the only making contract
1 deal where spades and grand both make, but not hearts
1 deal where hearts is the only making contract

So it looks as though the best chance is to make spades trump. Tom and I both got it wrong, but would have lost more as the cards lie by making the "correct" choice of spades. Kiebitz thinks that the best discard is always the 10 of diamonds and the 8 of clubs, but that's with perfect play. Keeping the 10 of diamonds might be worth a try, especially if your opponents are weak players.

[Mike Tobias]

Did you notice if the Jacks split the expected number of times?

When you do these deals in Kiebitz, do you discount those where you would obviously have been outbid?

[John McLeod]

I just checked: jacks were 1-1 ten times and 2-0 ten times.
No, I didn't try to eliminate hands where I would have been outbid. I suppose that doing that might raise the success rate somewhat.

[Mike Tobias]

Interesting that Grand only made in 3 of the 10 instances where they split. Unless they can make a fifth trick by overtrumping, the enemy can only get 65pts. Trumping a big one is therefore enough to win, and I would have thought that would happen more than 30% of the time.

[John McLeod]

Kiebitz finds the optimal play as the cards lie. In 7 out of 10 cases it found a line for the defence that didn't give the declarer the chance to trump an ace or ten. In real life, declarer would probably have won some of these deals. But then, I imagine declarer's success rate in hearts or spades would also be better in reality than against the optimum defence that Kiebitz finds, so I'm thinking that the conclusion that spades has a better chance than grand and hearts a worse chance is probably still correct. I can provide you with the actual 20 deals if you would like to analyse them by hand :)

[Mike Tobias]