Careful with the doubles when GiB is your CHO

Playing a 8-board robot tournament on BBO after a long-time, I got off to a rather poor start when my robot partner pulled my penalty double:

5H down 1 was a 38% board.  4Sx down 3 would have been a lot better of course.  But I had paid my 25c, and by golly, I was going to get my money's worth.  I decided to hunker down and finish the set.

By the 7th board, I was at 70% overall, and in first place.  This was the 7th board and after my RHO bid 4S, I had a decision to make.


Should I pass, double or bid 5H?  I recalled board 1, where my partner took out a clear penalty double.  What would the robot do with a cooperative one? I decided to bid 5H as insurance and when dummy came down, I thought I had blown it, and passing was the right call.  It turns out that 4S makes. Many of the other tables were in 4Sx, so my Center Hand Opponent would have passed the double this time, so 5H undoubled and down 3 was worth 82%.

The last board of the set was another double-or-not decision. This was the board:

After I doubled 3S for penalty and CHO took me out, I decided to not push my luck by doubling 4S for penalty.  It turned out that 4S down 3 was worth 95% anyway.  Doubling was not needed.

This sort of fielding partner's propensities is something that you always have to do with human players -- there are players who will sit and players who will run. It appears that GiB will take out his partner's doubles even at high levels.  And I can not figure what the robot's logic is regarding which doubles are for penalty and which ones are cooperative.  Any suggestions and/or pointers are appreciated.

Analyzing the common game

Finally, our club director signed us up for the Common Game, but I haven't been able to play much at the club.  But with a special Memorial Day game held on Monday afternoon, I got to try it out. Partner and I had a good game (63%), but another pair (who are often our teammates) had an obscene one (80%), so that was good only for second place.

For the most part, the boards fared about the same at the club as in the larger, nationwide field. The difference between the club and field was stark, however, on Board #3, I was South.  I opened the hand 1C.  West passed (!), partner bid 1H, I replied 1S and partner now bid 2NT.  He made 3 for +150:

This was worth 57% of the matchpoints nationwide but only 38% at the club. 

It appears that people at our club are bidding the two flat hands aggressively and ending up in 3NT (how?).  As it turns out, the defence to beat 3NT is extremely difficult.  The opponents need to lead spades, squeezing declarer before broaching either hearts or diamonds (depending on what declarer discards) to beat the contract.

When I bid aggressively,  on Board 15, it turned out to be a cropper.

I was West, and opened the hand 1H.  When partner replied with a semi-forcing 1NT, I should have realized that he might have a minor-suit bust and rebid a calm 2H. Instead, I got carried way by the good hearts and Aces and Kings and bid 3H.  Partner raised me to 4 and I got a diamond lead.

How do you play this thing? I decided to play for a doubleton heart honor with South.  So, I took the Ace and King of diamonds and played the third diamond.  Now, North won with the Jack of diamonds and switched to a club.  I took the winning club finesse, and discarded a spade on the Ace of clubs. I then took a heart finesse that lost to North's King.  Unwilling to give me one more club discard, North finally led a spade. When I got in with the King of spades, I plopped down the Ace of hearts, but unfortunately, South had a third heart and the queen did not fall.  4H down 1.

At the club, others stayed low, got a spade lead and easily chalked up 9 tricks in hearts.

The double-dummy contracts are always worth a think. On Board #17, the opponents bid 4S and made 7.

Double-dummy, they are supposed to make only six.  How? Assume that East is the declarer and South gets to lead. 

I led the King of hearts, but unfortunately, this gave declarer the entry she needed to run the 9 of spades and then a small spade to the queen and Ace of spades. She then came to hand with the Ace of diamonds and, after finessing clubs, was able to throw away her losing diamonds on the clubs. The way to beat the contract is to lead a diamond. This knocks out declarer's entry early.  If she leads the 9 of spades to finesse spades, she has no entry to her hand to take the club finesse (spades are blocked). If she leads a club to the Jack, the 10 is not a entry because I can ruff the third club. Leading the first spade low to the Queen fails because of the 2-1 split.

4S making 7 was a bottom, but I don't think others found the diamond lead. The declarer at our table just played it better than the field.

Estimating poorly

Playing in the Open Pairs game at the OKC regional, partner and I had a 62% game, winning our direction.  All through the session, though, I thought we were having about a 53% game.

Estimation is an important skill to have and I have been working on it. After every board, I estimate whether the board is a top (+2), a bottom (-2) or in between.  Because you have about 25 boards in a session, adding these scores can give a pretty good estimate of how you are faring.  Of course, it works only if your estimates (-2 to 2) are reasonable.

So, I went back and looked to see where and why my estimates were off.  In general, lots of boards I had marked as average (0) turned out to be Average+ (about 8 out of 12).  Essentially, getting to the right contract was worth an Average+.  Theoretically these ought to get washed out by boards where the opponents got to the right contract their way. Even though I consistently marked those also as average (0), they too turned out to be Average+ because the field was not taking all their tricks on defense.   Similarly, I would mark down a board where we dropped a trick as below average (-1) but it would turn out that the field was not picking up these tricks either.

Turns out that many of the stronger players were playing the first round of knockouts.  It may have been an "open pairs", but the field was relatively weak.

The only true mishap of the day occurred on a board where I made two bad decisions in the bidding:

I was West and had to decide whether to open this hand 1H or 1NT.  With a 15-17 balanced hand, I always open 1NT even if I have a 5-card major. However, I decided to upgrade the hand, open 1H and rebid 2NT.  That was mistake #1.  This is pretty clearly not an upgradeable hand.   Partner now bid 1NT (semi-forcing) and South overcalled 2S.  What's my bid now?  

I doubled (takeout) and partner bid 3C.  This is where mistake #2 happened. Forgetting that I had doubled (so 3C was to play), I bid 3S and put partner in a hopeless 3NT contract.  That was our only zero of the night.  Had I passed 3C on this board, we might have won the overalls too, not just the E/W direction, because 3C making 4 would have been Average+.

But I should not complain too much -- the other bidding misunderstanding of the night gave us a cold top (all 12 of the matchpoints).  It was on this board:

After North passed, partner (East) opened 1D and South overcalled 2S (weak).  I bid 3S.

Now, partner had a problem.  Was my 3S a limit raise of diamonds or a general-purpose force where his first priority is to bid 3NT with a spade stopper?  In any case, he had no spade stopper, so he bid 4D.  Back to me.  Well, if partner doesn't have a spade stopper, I can see a diamond slam if partner has a spade singleton.  So, I cue-bid 4H. Would partner cue-bid 4S? 

Partner was not on the same page, however.  Was 4H a cue-bid with diamonds agreed, or an offer to play? Partner passed my 4H bid.  Oops.

The Ace of spades was led and dummy came down.  What do you do?

"Thank you, partner," I said.  No one needed to know that I was in a 4-2 fit.

Ace of spades was followed by a low spade to South's Jack.  South then switched to his Q of diamonds. No surprise there.  Diamonds were 2-1 with the preemptor having only one diamond.  Chances were that his heart and club holdings were then 3-3.

Who rates to have the Queen of hearts?  North, of course.  Firstly, he has 4 hearts to South's three.  Secondly, with the KQJ of spades and Qxx of hearts, South would probably bid 1S, not 2S.  So, I took the diamond switch in hand, and played Ace of hearts and the Jack.

North didn't cover the Jack and I was home.  I ran the Jack, played a club to the Ace, pulled a round of trumps with the king and started running my diamonds. North could ruff in with the Queen of hearts, but I had the rest.

4H made on the 4-2 fit was a cold top.  The better pairs were in 5D while the rest of the field were in diamond part-scores or failing in 3NT.

Tell partner you want an uppercut

One of the most rewarding defensive maneuvers is achieving an uppercut.  You essentially manufacture a trump trick out of thin air.

In a club game, partnering a good player (and frequent teammate) for the first time, we have decided to play upside-down count and attitude, but nothing beyond that.  Still, we are doing well ( we would finish second, with 57%) when the opponents find themselves in a nice 3C partscore.

I lead partner's suit and the play proceeds up this point (hit Next to see the play) when the cards are as follows:

What card should I play?

From partner's point of view, I could have either the King of hearts or the Jack of clubs to defeat this contract.  If I have the King of hearts, he needs to lead a heart before declarer can discard his heart loser on the good spade. If, on the other hand, I have the Jack of clubs, then an upper-cut is the ticket to beating 3C.  My signal needs to tell him which line to take.

Discouraging hearts, perhaps with the 10 of hearts would have lead partner to finding the uppercut. He leads a diamond and if declarer ruffs with the Ace, my Jack is good ...

(Yes, declarer misplayed this.  If he pulls trump before playing spades, there is no upper-cut. Also, he finessed spades into the wrong hand.  If he finesses the spade into my hand, his hearts would be safe even if I held the Ace of hearts.  But then, this ain't the Bermuda Bowl.)

Sac and squeeze

The very first board of the very first match in a GNT game, we are playing one of the strongest pairs in the room. I pick up:

W

Me

♠

AJ9xxxx

♥

AJx

♦

x

♣

xx

And the bidding goes:

W

Me

N

North

E

East

S

South

—

—

—

1♣

1♠

Dbl

2♠

Pass

3♠

5♣

Pass

Pass

?

Do you agree with my bidding so far?  What's your bid?

From my point of view, 5C is making.  Is 5S a worthwhile sacrifice?

The best case scenario is that partner has Kxx of spades and shortness in hearts. In that case, I go down 1.  The worst case is that partner has KQx of spades and nothing else. I then go down 3. I rolled the dice and bid 5S, got doubled and went down ... five!  Turns out that the real worst case scenario is that partner raised with 10xx xxx Qxxxx Qx.

Needless to say, we lost that match.  But only by 7 imps.  Without my ill-advised sacrifice, we would have won.

We lost the next match too and found ourselves in a round-robin with two other two-loss teams.  "Why are they here?," sputtered our opponent to the director who patiently explained that the reason he had play us in the round robin was because we had lost our first two matches.  "I don't want to be here either," I mollified him, "I would rather be playing there."

Our opponent had just about settled down when this hand came up (I was West):

Once North  put in the King diamonds, I realized that I could squeeze South out of his king of spades. If the finesse worked, in other words, I didn't need to take it.  North, the danger hand, with his established hearts would be kept off lead ...  Here's the full hand (click Next to follow the play)



He wasn't happy about that either.  Playing in the last round of a Swiss teams having lost all your matches, you should not be subjected to a squeeze.

We ended up blitzing the round robin and ended up fifth. Not bad, but I had hoped to do better.

Classified ads for Team Matches on BBO

If you have ever played "pick-up" team matches on BBO, you know the routine. You show up with a partner, reserve seats for yourself and pard and wait for people to join your team match.  Ten to fifteen minutes later, the match picks up 8 players and you start playing.  After the first board, two people leave declaring that their partner is an idiot.  Finding substitutes is a royal pain and substitutes keep bailing in the middle of a game.  What you wouldn't give for reliable opponents!

Is there a better way? How about if you could come with a local team of players you know. Perhaps the team of four that you use to play the Grand National Teams.  What if other such "serious" players register online as a team, and specify their typical availability (weekends? weekdays? weeknights?).  You could private-message them, set up a match and play.  

The closest analogy would be an Inter-city league or classified ads for team matches.

A couple of objections raise themselves.  How do you know the other team is any good? Here, being part of a social circle where people use their real names is ideal, and Bridge Winners provides one for serious bridge players.

So, here's the way it works:

1. Register your team online.  Essentially, you provide your BBO and Bridge Winners user names. You also specify your availability and self-rate how good you are.
2. Look at the list of teams who are registered and choose a team that is around your level. To do this, look at their self-rating, the Bridge Winners name of the team captain and the best team that they have beaten in the past.  Send the captain a private message on Bridge Winners suggesting a match within the next week.
3. Plan out when you are going to play, how many boards and the format (Board-a-match? IMPs? ACBL GCC?).  Swap convention cards using Bridge Winners.  Do all of this over email.  Also swap the list of BBO user names so that one of the captains can create a Team Match on BBO.
4. After the match (or in the case of a no-show), the winner should report the results of the match.  This goes to update the Team List. Essentially, people looking to schedule future matches can learn when the team was last active (I will periodically delete teams that haven't been active for 3 months), whether their scheduling is reliable and whether their self-rating is accurate.

Use the "Test" and "Test1" team names to try out the forms and see the results.

A couple of caveats:

1. It's mostly on the honor system, but I hope that it will work and be of use. Obviously, if people start reporting bogus matches, the ratings cease to be of much use but it will probably still function well as a classified ad service.
2. I created the system because I have been wanting to learn the Google Apps API, and this was a small enough task.  There is one glaring problem that I hope someone familiar with the Google Apps API can help me with.  If there is a problem with a registration or match result (such as if the user specifies a team that doesn't exist), I ignore their submission.  I obviously need to alert them that they gave me a wrong team-name, but the documentation warns that "user-interface elements are visible only to an editor who opens the form to modify it, not to a user who opens the form to respond." and sure enough, it doesn't seem to be possible to display an error dialog to the user submitting the information.  So, how can the user be told his request did not go through?  Gah!  Until some kind person can tell me how to do this, wrong submissions will be silently ignored.

Right through the pack

Over Spring break, we made a family trip to the Pacific Northwest.  We spent one rainy afternoon in Portland immersed in Powell's World of Books, the largest used book store in the world or something close to it.  It's a Portland institution.

We went into the store as a family of four, but in a couple of minutes, we had all dispersed.  The wife went to the pottery section.  The daughter took down every knitting book in the store looking for cool animal-pattern hats.  The son made a beeline to origami books.  And me?  I went looking to see what books they had on card games.

For the most part, the used bridge book selection was terrible -- their buyer must not play bridge.  There were really dated books on Kaplan-Sheinwold and the like.  Among the dated books was a paperback with a publication date of 1947 with an introduction by Ely Culbertson. The bidding on a deal I saw had the dealer opening 1S holding 4 spades and 5 (much better) hearts. I was replacing the book on the shelf when the blurb on the front cover caught my attention.

"Rated one of the top three bridge books in a 1984 ACBL survey!", the book proclaimed. A 1947 book that people remembered 37 years later?  That piqued my curiosity and I started to read the first chapter. It seemed very weird, and definitely not likely to be rated best of anything. So, I flipped over to a middle chapter and started reading the chapter.  And was immediately hooked.

It is an awesome, awesome book.  The bridge deals are amazing and the writing crystal clear. They are mostly run-of-the-mill hands, but the depth is incredible.  Darvas and de Hart manage to capture the ineluctable beauty of card play, The stories are cute and the illustrations charming. How could I have not heard of Right Through The Pack before?  Forget Victor Mollo who gets repetitive after a couple of stories.  Robert Darvas never bores you.  Every story is different.

This was by far the most entertaining bridge book I have ever read.  If you have not read Robert Darvas' classic, run thee off to a bookstore and get thyself a copy.