In yesterday’s pairs game my partner and I twice found ourselves in four-three fits. In both cases we ended up in unmakeable contracts, but both results were slightly above average. In the first one my partner dealt at favorable vulnerability.

Hand #2: My hand (9 HCP; 7 losers):

♠A54 ♥T3 ♦AQJ72 ♣T64Partner’s hand (12 HCP; 7 losers):

♠T863 ♥A6 ♦95 ♣AKJ62 

Partner opened 1♣; I responded 1♦; after LHO overcalled 1♥, partner rebid 1♠. I seriously considered jumping to 3♠, but I downgraded my heart values and settled for 2♠, which was down one. 2♦ would have worked better.

On the second hand LHO dealt, and we were vulnerable.

Hand #25: My hand (12 HCP; 7 losers):

♠AKQ3 ♥KT75 ♦964 ♣62
Partner’s hand (12 HCP; 6 losers):

♠6 ♥AQ8 ♦AQ73 ♣T8753This time, if I remember correctly, the opponents were silent. Partner opened 1♦; I responded 1♥; partner bid his club suit. I jumped to 3NT, and partner corrected to 4♥. Down one again. This time we should have stopped at 2NT or 3♥.

My style for the last couple of years has been to support partner’s four-card suit whenever I held three cards in that suit plus a singleton or worthless doubleton in a side suit. I think that I picked this up from an article by Marshall Miles, but I could not subsequently find it. Yesterday’s result made me wonder if I should rethink this approach.

The first lesson to be gained from these hands is that Losing Trick Count does not work with Moysian fits. In the first hand we made three tricks less than LTC would predict. The second hand was two tricks short of the prediction. This was a known phenomenon. LTC can only be applied when an eight-card fit has been discovered.

Another thing that I had previously noticed was that Moysian fits work better if the trump honors are concentrated in the declarer’s hand. Otherwise, he/she may have to use valuable honors for ruffing. In the first case, my holding was perfect: A54. Could the problem have been that I had a void instead of a singleton? Yes, but only indirectly. If I had held five hearts, I would have bid them in preference to diamonds. So, the only possible Moysian distribution with a club void was 3=4=6=0.

I should definitely have rebid my six-card diamond suit. As long as partner had two of them, we should be better off in diamonds. This would have been much better than a no trump bid, even though I had an excellent holding in the opponents’ suit. So, from now on, I will amend my policy to prefer six-card rebids to three-piece support.

In the second case we found ourselves in a good strain. The problem was that we ended up just slightly too high. I don’t think that there is much more to learn from this hand. If the diamonds had behaved, we would have made it.