Dummy Squeezes are Fun

So, the second place streak continued. Joel and I were 2^nd in the D7 Flight A North American Pairs. Of course, that’s a much tougher event than the sectional events the week before. Almost all of the top players from the district showed up.  It’s exactly 6 years after David and I qualified (barely) for a trip to the NAPs in flight B. The spring 2007 NABC was in St. Louis as well and as I recall it was one of my least favorite NABC venues, although that may be just because it was very cold and rainy.

We played pretty well all day but weren’t getting any really great boards until the second half of the evening session. We bid 4♠ with a shapely 13 opposite 7 hcp, got doubled and made an overtrick. They tried to steal from us by making a light 2♥ overcall and a natural 2NT from advancer. On the companion board, they tried to steal from us again and more good things happened for us.

Dealer: E Vul: NS	North ♠ 5 ♥ AJ76 ♦ KJ ♣ AKT432
West ♠ J3 ♥ QT8 ♦ AQ32 ♣ J965		East ♠ AQ9862 ♥ 9543 ♦ 74 ♣ Q
	South ♠ KT74 ♥ K2 ♦ T9865 ♣ 87

East	South	West	North
2♠	Pass	3♠	X
Pass	3NT	Pass	Pass
Pass

The 3♠ raise on Jx may have shut out some players or gotten an uncomfortable 4♣ overcall but for us, it just steered us right into 3NT. If west passes, we may wind up in 3♦. And if east doesn’t open because of the 4 hearts, we likely would have wound up in 3♣ making. So, a lot of matchpoints were on the line in 3NT. The opening lead is the ♠J. East ducks, south wins the K, plays a top club, comes back to the ♥K, plays a club to the T, ♣A, and then a club to west’s J. How should east discard?

Seeing all the hands, it’s easy enough to see that east should discard spades, so that after taking the ♠AQ, he can squeeze/endplay dummy. The defense would come to 1 club, 2 spades, and either a diamond and a heart or 2 diamonds depending on how dummy discarded. However, it must be tempting to keep all the spades, especially at matchpoints where a 2^nd or 3^rd undertrick could be valuable. That would mean discarding potentially safe diamond exit cards or discarding hearts, potential allowing a long heart in dummy to be good.

Think about declarer’s line of play. Even after the Q fell under the ace, the only way to pick the suit up for no losers was to play for QJ doubleton so declarer was always willing to concede a club trick. Would he do that without a second spade stopper and with a source of tricks in hearts? Furthermore, would he have taken the first trick with only Kxx? Probably not.

Is it really important to hold heart length with dummy? Maybe. If declarer has the Q, it doesn’t matter. The critical situation is when declarer started with Kxx, meaning partner’s Q will fall under the A and letting declarer score 4 heart tricks would be devastating.

Yes, we normally would expect partner to have 3 spades to raise and NS did bid 3NT under pressure so south may not have as much as he would like opposite a t/o double. That swings things the other way a little bit but most of the evidence points toward declarer having a second spade stopper, in which case east has no reason to hold more than the AQ.

Discard a discouraging diamond first, then 2 spades. Partner will win the club and lead a spade to east. Dummy sheds a heart first and then is squeezed when east cashes the next spade. Here is the end situation with east on lead at trick 8 and 2 tricks in the bag.

Dealer: E Vul: NS	North ♠ ♥ AJ ♦ KJ ♣ 43
West ♠ ♥ QT ♦ AQ32 ♣		East ♠ A6 ♥ 954 ♦ 4 ♣
	South ♠ T7 ♥ 2 ♦ T98 ♣

On the ♠A, dummy cannot retain both red tenaces without discarding a club winner so he lets go of a diamond. East can’t afford to play a spade because it would give declarer his T and can’t afford to play a heart because it gives declarer a finesse he can’t take on his own so on a diamond lead west takes the ♦AQ for -1. As it happened, east had discarded 2 diamonds and a spade and was forced to lead a major suit at trick 8 so we wrapped up 5 clubs, 3 hearts, and a spade for the +600 and a tie for top. 

It’s at this point when I really started to feel like we could win the thing. Patty and Kevin, however, followed up their 60% with a 59.5% to beat us by 3/4 of a board.

Sometimes bridge is an easy game. That’s kind of how it felt last weekend. We made a couple of defensive errors but every bid worked out well and kept taking extra tricks as declarer. Perhaps some of it was a superior bidding system but perhaps more importantly, we were paying attention to everything a little more because we were playing a new bidding system.

Sean and I played only 2 out of the 4 days of the Charlotte sectional, didn’t win an event but got over 26 points – 2^nd in each of the Saturday pair games and 2^nd in the Sunday Swiss. Very few Fantunes-relevant things came up except for the 2 bids, which were fairly frequent. I must say, we had a lot of good auctions.

On one of the more exciting deals, I held Q, Jx, AKQxxxx, AKQT and see my partner, with both sides vulnerable, open 2S, showing 10-13 points, an unbalanced hand with 5+ spades and fewer than 4 hearts. Thankfully my long suit was diamonds and not clubs else I probably would have forgotten than 3C is an invitational or better spade raise. 3D, however, is natural and forcing so I tried that, intending to follow it up with 6C as there doesn’t seem to be a reasonable way to find out if partner has a heart control but odds seem to be with me that he does. LHO bid 3H, passed back to me. Perhaps I should hit this but that would be a poor result if LHO has at least 7 heart tricks, especially if we have a slam. I try 4H, which is doubled back to me. So it sounds like partner has no heart control so I bid 5C and got to play it there opposite the awesome dummy: AKJxx, xxx, J, Jxxx.

I guess I was wrong about there being no exciting football games last weekend. Tech now has a pretty good chance to win the coastal division with a record of 6-6 overall, 5-3 in the ACC. They definitely have to beat Duke this Saturday. Then they need Duke to beat Miami, which would put GT alone on top or tied with ineligible UNC. If Miami beats Duke, GT needs UNC to win its last 2 games against Virginia and Maryland to force a 3 way tie, and then they need Virginia to beat Virginia Tech so that GT will win the 2^nd tie-breaker of record against the next top team in the division.

Florida squeaked by lowly Louisiana-Lafayette but went up to #2 in my computer rankings. That’s strange. Alabama slipped from a virtual tie for 1^st to 3^rd. K-State is 4^th and Oregon is a very distant 5^th. Notre Dame is a clear #1 while my computer still rates UGA, LSU, and South Carolina much lower than the polls – they’re 12-14. Georgia Tech is #62, right in the middle, and Duke is #58, but my computer has Tech by 9 this Saturday. My picks against the spread took a hit, going only 7.5-12.5 in the top 25 games.

Fantunes and college football through 10 weeks

Recently I’ve been looking at the Fantunes bidding system in depth and trying to decide if it is something I would like to play. In case you are unfamiliar, this system, developed and used by world champions Fulvio Fantoni and Claudio Nunes, features very sound, natural, and forcing 1 level openings and 2 level openings that are natural and usually 10-13 hcp. The biggest advantage to this is that it puts you in better position to describe your hand if they interfere, having already shown strength and a real suit (except that 1C could be just about any strong NT). That is worth a lot, given how people compete on air so much nowadays. However, 4441 hands are more of a problem because there’s no real natural way to bid them. Their system puts the 12-14 hcp 4441 hands in with 1NT but the ACBL (at least in general convention chart events) frowns upon that. Sean and I are trying to work around that and try this modified Fantunes system this weekend.

In other news, college football rankings after 10 weeks are out. After beating LSU, Alabama is still solidly #1 in the BCS but Notre Dame still has a slight edge over them in my computer rankings. I expect that by the time they demolish Georgia on December 1, they will rise to #1 in my computer.

								Strength of
Rank	Prev	BCS		W	L		Team	Schedule	Total Rtg.
1	1	4		9	0		NOTRE DAME	8	94.180
2	2	1		9	0		ALABAMA	20	93.329
3	4	6		8	1		FLORIDA	5	88.660
4	3	2		9	0		KANSAS STATE	40	86.593
5	5	11		7	1		OREGON STATE	27	83.872
6	7	3		9	0		OREGON	78	79.820
7	6	NE		10	0		OHIO STATE	70	78.126
8	9	5		8	1		GEORGIA	53	77.566
9	11	12		6	2		OKLAHOMA	16	77.071
10	19	15		7	2		TEXAS A&M	21	75.651
11	14	16		7	2		NEBRASKA	14	74.802
12	16	13		8	1		CLEMSON	84	73.692
13	8	14		7	2		STANFORD	33	72.895
14	10	7		7	2		LSU	24	72.874
15	13	10		8	1		FLORIDA STATE	93	71.230
16	15	8		7	2		SOUTH CAROLINA	22	70.900
17	23	17		7	2		TEXAS	43	69.275
18	25	18		7	2		UCLA	48	68.779
19	20	9		9	0		LOUISVILLE	118	68.661
20	12	22		6	3		TEXAS TECH	12	66.833
22	33			6	2		CINCINNATI	82	64.309
21	21	23		7	1		RUTGERS	112	65.170
23	22			7	2		San Jose State	87	64.213
24	32			6	3		MICHIGAN	29	63.990
25	28	25		8	1		TOLEDO	113	63.848
26	26	24		7	2		Northwestern	75	63.097
27	34	20		8	1		Louisiana Tech	120	62.929
28	31			8	1		Kent State	122	62.380
29	18			7	2		BOISE STATE	100	62.267
31	42			7	2		UCF	115	61.820
30	27	21		7	2		Mississippi State	86	62.059

Georgia Tech really needs to beat UNC this week and Duke next week to get to a bowl for the 16^th straight season because they don’t stand much chance against Georgia. My computer picks UNC by 9. It looks like there aren’t many interesting games this week, which is fine because I don’t have time to watch any football this weekend.

In case anyone is wondering, my computer prediction program is 61.1% against the spread this season in games involving top 25 teams. That's pretty darn good. It was 53.2% in 2011, 52.9% in 2010, and 47.5% in 2009.

Its predictions for this week:

Florida State 37, @Virginia Tech 17

@Alabama 31, Texas A&M 25

Notre Dame 33, @Boston College 13

Georgia 35, @Auburn 17

@LSU 29, Mississippi State 19

@Stanford 24, Oregon State 21

@North Carolina 35, Georgia Tech 26

Wake Forest 25, @NC State 24

BAM BAM

Only recently did I notice that the ACBL masterpoint formula awards a 10% premium for board-a-match events. Yesterday we ran a BAM at the Warner Robins club with a respectable 7 tables and it awarded 1.29 MP per player for the 100% sectionally-rated game. One should not be able to win such an obscene number of points (3.12) for 1^st out of 7 at a club. You’d have to be 1^st out of 17 in a sectonally-rated pair game or 1^st out of 9 in a sectional Swiss game to get that many points (1^st in a 9 table MP or Swiss game is worth 3.17).

There are no match awards for BAM so that takes away a little bit of the bonus that BAM gets compared to Swiss but this still ought to be a good argument for making more team games BAM. If I ever get to be in charge of a tournament schedule, there will definitely be some main events that are BAM, specifically the Sunday event will be flight A BAM and flight B Swiss, but I’ve mentioned that several times before.

There were a number of interesting deals last night. I was put to the test in the first round in this not-so-great 3C contract.

 Jxx

8xxx

xxxx

Ax

Tx

AQ9x

Qx

KJ9xx

The bidding is crucial here, although if I paid no attention to the bidding, I would have surely gone down 1 by simply taking the finesses and cashing winners. However, the fact that RHO doubled after 1C-P-1D and LHO bid 2S directly over my 1NT rebid led me down a very wrong path. Yes, I have a crappy 1NT rebid but it’s BAM and I can sometimes be a hand hog. Anyway, when 2S came back to me, I competed to 3C and got it there.

LHO led the HJ which I took with the Q. That looks a lot like a singleton. I then played a club to the ace, RHO dropped the T, then played low on the second club. This made it appear that she had a doubleton (consistent with the bidding) so I went up with the K and led another club. It seems pretty likely at this point that RHO started with something like Kxxx, KTxx, AKx, xx so taking a losing club finesse would expose myself to 2 heart ruffs plus the natural trump trick. As it turns out, QTx was on my right. She then cashed the SA, DAK, and led the DJ. So, RHO is 3-3 in the minors. Which major does she have 4 of? If I figure this out, I can still escape for -2 and win the board against a likely 2S making at the other table.

RHO could have started with Axx, KTxx, AKJ, Qxx. That gives LHO KQxxx, J, Txxx, xxx.

If RHO started with Axxx, Kxx, AKJ, Qxx, that gives LHO KQxx, Jx, Txxx, xxx.

Of course, what if RHO has no 4 card major? Axx, Kxx, AKJx, Qxx. That gives LHO KQxxx, Jx, Txx, xxx

The first one is definitely more likely so after ruffing the 3^rd round of diamonds, I prepared to endplay RHO by leading a spade, expecting LHO to win and continue spades. I could ruff that with my last trump and lead a heart to the 8. RHO would win that but have to lead a heart into my A9 tenace. I would come to 4 club tricks and 3 hearts for -2.

As it turns out, RHO had the 3^rd hand (and therefore an abnormal double), so when I led the heart toward dummy. LHO won the HT and cashed 2 spades for -4. Sure enough, the teammates were in 2S making so -2 would have been enough to win the board. FML.

New football rankings and Athens sectional recap

Few things went according to plan with last weekend’s trip to Athens, at least according to the plan in place on Friday morning. Even the other plans that were made as the weekend went along were often changed. Playing 2 sessions of bridge Saturday was the one thing that did. Ironically, that was clearly the worst part of the weekend. Winter, however, got to spent that time having tons of fun with our new friends and I was able to join them after bridge.

If you don’t play bridge, skip this paragraph. Bridge was easy Friday night. We won with ease as everyone was in a gift-giving mood. Saturday was kind of terrible. Twice I attempted to duck smoothly when declarer finessed into my K (behind the ace) to save my entry, cut him off from dummy later, or something like that. In neither case did I score my K later and the plays were costly.  The hand that still baffles me the most is when my partner held Qxx, void, KQxxx, KJTxx. I opened 1H, he responded 1NT, and I rebid 3H. He raised this to 4H. What the heck does it take to bid 3NT? I mean, yes, ideally, the spades would be a tad better but this falls well within the expected spade length/strength for this bidding. Then my LHO doubles 4H and partner still doesn’t bid NT. Obviously now our 6-0 trump fit is getting s 5-2 trump split or worse and you have TWO good sources of tricks outside hearts! 1H-1NT-3H is not like 1H-2C-3H: the latter shows a self sufficient suit because 2H is still forcing with 6+ hearts. In the former (the auction that actually happened), 3H is the only bid you have available to show some extras with 6 hearts so you can’t expect it to be self-sufficient. It could be and if so I may well bid 4H over 3NT.

Non-bridge players who are football fans: start reading again. In other news, college football rankings through week 8 are out. Notre Dame is #1 and UGA is way down at #22 in my computer. Somehow, my algorithms don’t love the SEC as much as usual and certainly not as much as the voters. Here are the top 25 in the Asbury computer followed by some computer predictions for next week.

Rank	Prev	BCS		W	L		Team	Str. of Sch.	Rating
1	1	5		7	0		NOTRE DAME	14	93.881
2	3	2		7	0		FLORIDA	16	93.233
3	2	1		7	0		ALABAMA	52	89.851
4	4	7		6	0		OREGON STATE	34	89.829
5	5	3		7	0		KANSAS STATE	44	87.974
6	10	4		7	0		OREGON	67	81.699
7	11	8		5	1		OKLAHOMA	39	80.155
8	7	14		6	1		TEXAS TECH	26	78.863
9	8	NE		8	0		OHIO STATE	75	77.413
10	17	6		7	1		LSU	54	76.311
11	16	17		5	2		STANFORD	4	75.509
12	15	15		7	0		RUTGERS	117	74.352
13	6	19		5	2		WEST VIRGINIA	17	73.090
14	14	11		7	0		Mississippi State	115	71.423
15	20	12		7	1		FLORIDA STATE	99	70.399
16	13	21		6	1		BOISE STATE	101	70.251
17	12	13		6	2		SOUTH CAROLINA	24	70.159
18	9	20		5	2		TEXAS A&M	22	69.980
19	34	22		5	2		MICHIGAN	38	67.651
20	21	9		6	1		USC	98	67.518
21	28	18		6	1		CLEMSON	97	67.394
22	19	10		6	1		GEORGIA	100	66.975
23	29	23		5	2		TEXAS	25	66.448
24	35	25		6	2		WISCONSIN	51	65.366
25	25	24		7	0		Ohio	124	65.319

This seems odd: my computer has its top-ranked Fighting Irish losing 28-20 to Oklahoma (-9). Then again, it does try to account for home field and momentum from recent games and even adjusts a bit if a team is at the end of a particularly tough stretch of games. Other scores to expect (with the point spreads in parentheses taken from collegefootballpoll.com):

BYU 28 (+2),              @Georgia Tech 24

Florida 33 (-3.5),         Georgia 20

@South Carolina 35 (-14), Tennessee 24

Clemson 37 (-16),   @Wake Forest 24

@Florida State 36 (-26), Duke 24

@Alabama 30 (-24),   Mississippi State 21

@Kansas State 37 (-8), Texas Tech 28

Here’s another anomaly: @Vanderbilt 161, Massachusetts -114. This is undoubtedly because Massachusetts has a negative rating of -2.902 and apparently that throws the prediction algorithm out of whack. Geez, I need to add some home state bias to this program. This business of predicting GT to lose when they're officially the favorite is not cool. Neither is predicting UGA to lose by 13 when they're only 3.5 point underdogs.