[Leo LaSota]

If I had to guess, hands played in nt average slightly lower in terms of rank of card that wins trick 13 versus hands played in a suit contract.

[barmar]

Even without claims, I wouldn't expect aces and kings to be played frequently on the last trick. If you're running a suit, you usually start from the top, so the ace will be played early. And defense, honor sequences are common leads early in the play.

The low occurrence of low spot cards is also unsurprising. These are frequent early leads, played when you can't try to win, or easy discards.

[inquiry]

 JLOGIC, on 2012-August-20, 19:55, said:

Does excluding claims bias the results? Maybe bad players never claim so it doesn't matter. I'm not sure if it's relevant, but when you have aces to win the last trick with or high trumps a claim is more likely to be made.

FWIW I also thought it would be the ten of spades or something liek that.

I thought the original post request played to the "last trick" meant trick 13. What i do is output 1000's of hands from bridgebrowser in a format such as below, paste into excel, create filter and count rows that begin with pc|, and if the one before it began with pc, add one... so 13 plays in order would give the last row with 13... if no pc| at the start of row, then gets a "0"... then select only rows where the count was 13....

Doing this you could find most common lead, for example. Not sure either is useful.

To do the which card, by suit, I guess we will have to count a lot more hands... maybe 100,000 or 200,000. I will give it a try tomorrow.


mn|IMP-90 - 2009.04.01|pn|South,West,North,East|qx|o1,BOARD 1|rh||ah|Board 1|md|3S86HJ92DQ4CAT7632,SQJ743H64DT9CQJ95,SKTHKQ8DAKJ65CK84|sv|b|
em|NS 9.13|
sa|0|mb|2N|mb|p|mb|3N|mb|p|mb|p|mb|p|pg||
pc|H5|pc|H2|pc|H6|pc|HK|pg||
pc|D5|pc|D2|pc|DQ|pc|DT|pg||
pc|C2|pc|C5|pc|C8|pc|S2|pg||
pc|DA|pc|D3|pc|D4|pc|D9|pg||
pc|DK|pc|D7|pc|S6|pc|S3|pg||
pc|DJ|pc|D8|pc|S8|pc|S4|pg||
pc|D6|pc|S5|pc|C3|pc|S7|pg||
pc|H8|pc|HA|pc|H9|pc|H4|pg||
pc|HT|pc|HJ|pc|SQ|pc|HQ|pg||
pc|CK|pc|S9|pc|C6|pc|CJ|pg||
pc|C4|pc|H3|pc|CA|pc|C9|pg||
mc|10|pg||
pn|South,West,North,East|qx|o2,BOARD 2|rh||ah|Board 2|md|3S86HJ92DQ4CAT7632,SQJ743H64DT9CQJ95,SKTHKQ8DAKJ65CK84|sv|b|
em|NS 8.60|
sa|0|mb|1D|mb|1H|mb|1N|mb|p|mb|3N|mb|p|mb|p|mb|p|pg|

[JLOGIC]

Sure, you answered the question and the answer has no use, I guess I was trying to ask if it would be much different if people never claimed(which, as you imply, was not asked). I really had no point to my question, just was curious what people thought heh.

I would be curious as to the most common lead also. I would guess it's the CA (because people lead a lot of aces! and clubs is the least likely trump suit), or the CK (because many lead K from both AK and KQ). I find these thinkgs interesting just for the sake of them lol

[EricK]

 JLOGIC, on 2012-August-20, 23:23, said:

Sure, you answered the question and the answer has no use, I guess I was trying to ask if it would be much different if people never claimed(which, as you imply, was not asked). I really had no point to my question, just was curious what people thought heh.

I would be curious as to the most common lead also. I would guess it's the CA (because people lead a lot of aces! and clubs is the least likely trump suit), or the CK (because many lead K from both AK and KQ). I find these thinkgs interesting just for the sake of them lol

I was guessing maybe the ♠2 for a commonest lead. I have heard a number of people say, against NT, "If in doubt leads ♠", and I have heard a number of people say "If in doubt lead trumps." [I am not advocating for or against either of these rules of thumb]. You are most likely to be in doubt against NT if you have a choice of 4 card suits to lead (I reckon), and ♠ is the most likely trump suit.

[barmar]

 inquiry, on 2012-August-20, 21:40, said:

I thought the original post request played to the "last trick" meant trick 13.

Seems like a reasonable assumption, since the subject line specifically says "13th trick".

[Phil]

 JLOGIC, on 2012-August-20, 19:55, said:

Does excluding claims bias the results? Maybe bad players never claim so it doesn't matter. I'm not sure if it's relevant, but when you have aces to win the last trick with or high trumps a claim is more likely to be made.

FWIW I also thought it would be the ten of spades or something liek that.

edit; answered.

[inquiry]

Quote

I would be curious as to the most common lead also. I would guess it's the CA (because people lead a lot of aces! and clubs is the least likely trump suit), or the CK (because many lead K from both AK and KQ). I find these thinkgs interesting just for the sake of them lol

Here is some more statistics taken from BBO main room (each hand played 16 times) rather than tournaments (where hand might be played once or several hundred times). These were from sequential hands played in april 2009. There was no selection based on final contract. Talk about useless stats... but Justin was wrong, the most common lead card wasn't the ♣A, it was the ♦A. He was right about aces, and it was minor aces, then major aces, then minor kings then major kings (but the D2 snuck in before the SK). I wonder if "short club" was just enough to tilt the lead to diamonds over clubs? Anyway, none of this might be significant. After aces and kings, then 2's and 3's were popular.

For opening lead, from 112,383 played hands (more hands searched, but pass outs were ignored since there was no opening lead (cards sorted from most frequent to less frequent).


DA: 4714
CA: 4651
HA: 4247
SA: 4001
DK: 3125
CK: 2895
HK: 2787
D2: 2748
SK: 2695
H2: 2468
S2: 2413
C2: 2409
H3: 2311
S3: 2310
D3: 2297
C3: 2283
C4: 2205
H4: 2145
CQ: 2138
DJ: 2126
S4: 2097
DQ: 2039
HQ: 2013
HJ: 1991
D4: 1978
S5: 1970
CJ: 1953
SJ: 1907
DT: 1897
H5: 1895
SQ: 1889
ST: 1879
H6: 1862
D5: 1859
C5: 1852
S6: 1724
D6: 1724
H9: 1708
C6: 1669
HT: 1648
CT: 1640
C7: 1585
D9: 1532
D7: 1529
C9: 1509
D8: 1506
H7: 1483
S7: 1469
C8: 1437
H8: 1433
S8: 1378
S9: 1360

As for cards played on the last trick (note: there are 4 cards played on the last trick, so the total is much higher than the leads -- these were the same hands-- but a lot of hands didn't have a trick 13 and are excluded from the data....

Cards (general) played to the last trick.


A: 17168
K: 33349
Q: 42179
J: 42296 <<-- jack gets lead again, but really, no difference between Q or J
T: 38256
9: 31301
8: 24224
7: 18188
6: 12625
5: 8992
4: 5965
3: 3695
2: 2362

Cards specific, played to the last trick (unsorted).

SA: 4565
HA: 4235
DA: 4184
CA: 4184
SK: 8521
HK: 8323
DK: 8235
CK: 8270
SQ: 10616
HQ: 10479
DQ: 10296
CQ: 10788
SJ: 10455
HJ: 10562
DJ: 10397
CJ: 10882
ST: 9776
HT: 9442
DT: 9261
CT: 9777
S9: 7867
H9: 7931
D9: 7770
C9: 7733
S8: 6406
H8: 6263
D8: 5754
C8: 5801
S7: 4699
H7: 4714
D7: 4384
C7: 4391
S6: 3325
H6: 3176
D6: 3065
C6: 3059
S5: 2377
H5: 2232
D5: 2158
C5: 2225
S4: 1639
H4: 1489
D4: 1451
C4: 1386
S3: 990
H3: 967
D3: 892
C3: 846
S2: 628
H2: 602
D2: 545
C2: 587

[Zelandakh]

I think it would be interesting to divide the last cards between declaring side and defending side. My guess would be that declaring side's last card (if played out) will most often be a trump spot whereas defending side's last card will most often be an honour. The problem is that claims affect this quite a lot since declarer will often claim when only trumps in hand/dummy remain.