This article originally appeared in the November 1999 issue of GammOnLine.|
Thank you to Kit Woolsey for his kind permission to reproduce it here.
As this article is being written, I am in Las Vegas at the Nevada Backgammon
Tournament. It is a great tournament, and I recommend it to everybody.
Howard Markowitz, the director, does a superb job, and the staff carries
out their work well. I'll be jotting down interesting situations as they
arise during the day, and then put them in this article at night while they
are fresh in my mind.
My opening event was the $750 jackpot tournament. The matches were 15 points. In my first match, I reached the following unusual position:
Blue is on roll. What should the cube action be?
This came out of a well-timed 1-2 backgame, after White rolled a couple of horror shots to jar three checkers loose. What's going on? I polled several good players, presenting it as a money problem in order to avoid confusing the issue even more than necessary. I got opinions ranging from easy take and no double, to double and pass. Of course at the match score White would be more inclined to pass, since his potentially juicy recubes to 8 lose a lot of their value at the match score.
In practice, I was White. My opponent did redouble to 4, and I chose to pass. The score was a definite factor; had this been the start of the match I would have taken. It should be noted that the pass/take decision may depend on how well Blue is likely to play the position. Blue's main strategy is to avoid letting White get an anchor at all costs. For example, if Blue rolls 5-2 I believe his correct play is not the seemingly obvious B/20, 24/22* but the unusual looking B/23, 7/2*. Was my opponent up to this? Hard to say. The only clue I had was that on the previous play he unhesitatingly broke his eight point to hit loose on the two point. This indicated to me that he had some concept of how to play the position, and that influenced my decision to pass the double.
Eventually we got to double match point, where I faced a series of agonizing pay me now or pay me later decisions almost in succession. Here is the first one. Things were going along quite smoothly, when I faced:
Obviously my choices are to pay now with 8/3*, 8/4 or to pay later with 6/1, 5/1. I felt that if I got hit and was hit back, I would still have a pretty good chance since he had only a four-point board and several checkers to extricate over my still decent blockade. If I waited, my next five could present real problems.
I played 8/3*, 8/4, and he hit back. I entered immediately, and on the roll after that was faced with another problem:
Again I could play now with 17/11, or I could wait and pay later with 6/4, 5/1. I chose to pay later. I felt he could have a difficult play to make. I'm not at all confident about my choice. The problem is that I lose my flexibility, and if he gets another checker into the outfield things could get even worse.
He did get another checker out, and I was forced to leave a double shot when I rolled 6-1. He missed the double shot, and I was then faced with another similar problem:
Will these nightmares never cease? Once again I could pay now with 10/5, 4/2* or pay later with 6/4, 6/1. At the table I chose to pay later and break my six point. I am now convinced that this was wrong. It doesn't gain all that much, since I am still leaving 5 shot numbers as opposed to the 11 shot numbers the loose hit leaves. I will still have the problem of getting the blot home safely next turn, and White will have a chance to move his men home and possibly improve his board. Most important, which I failed to see at the time, was that clearing the six point was very detrimental for my racing chances. Now every three I rolled would cause me to miss in the bearoff.
He moved his outfield checkers along, and once again the dice were there to taunt me:
Once again I could pay now with 10/3, or pay later with 5/1, 4/1. Being consistent, I paid later. I believe this is correct. Paying now doesn't have the big gains it had on the previous roll with the loose hit, while it has all the same losses.
White did nothing special, and I managed to get in safely. However he then rolled boxes, and the misses from my following twos and threes caused me to lose the race breezing. Live and learn.
Experts often challenge each other with interesting propositions. I saw this one being played between Nack Ballard and Mike Svobodny. Which side would you like to have (Blue is on roll)?
My first impression was that Blue had the better of it. He might get lucky and scramble everybody home without getting hit and win a gammon, and if he got hit White still has a lot of work to do. However after watching the position being played out for several games I changed my mind. Blue's numbers continue to be awkward, and he just doesn't get home safely. And once White hits a shot, the position plays very well for him. In addition White owns the cube, and can often rewhip at an ideal moment.
The open. 15 point match. I had a huge adverse swing on this position:
I had just rolled a 4-6 joker from the bar which hit two blots, and White was only able to enter one of them. My gammon chances weren't large, but it looked like a reasonable shot to take a roll. If I picked up the second blot or rolled some other crushing number, I would have good gammon chances and could play on safely for quite a while. Otherwise, I could just cash on my next roll. What could possibly go wrong?
What could go wrong? Well, I opened off with 3-3. Naturally I make my bar point. He fired back 6-4, entering and hitting. So much for cashing. After many flunks on my part a gammon was scored up, but unfortunately it wasn't in my favor.
Later on I cost myself the match in a very strange way, by blowing a bearoff play:
It is virtually always right to take two checkers off when you can, but there are exceptions. This position happens to be one of them. The problem with playing 6/0, 3/0 is if I then roll exactly two more threes in the bearoff that will cause me to miss twice, costing me a roll. If I correctly play 6/3, 5/0 I only take one checker off, but I will have 12 men left and two checkers on the three point so I won't miss at all if I roll two threes. You can work out for yourself how many threes I rolled in the later bearoff, and how many rolls I lost the race by.
Enter the consolation. My first match had this dynamic position:
I was Blue, on roll. Is it a rewhip? I have 22 rolls which cover the blot on the three point, and a few other good possibilities. If I do anything decent and he flunks, it looks like I will lose my market. The match score argues for relatively aggressive cubes, since he won't get much value out of holding a four-cube -- my take-point will be 10% if he rewhips to eight. Also since I am behind I can use the four points more than he can. So I sent it to four. He took, of course. I started with 6-5 making my five point, but he entered immediately. A few rolls later I was forced to drop the redouble, and he went on to win the match.
In the evening I tried another jackpot tournament. I've written a lot about match play strategy, so let's see some of it in practice:
White was on roll, and redoubled to four. Should I take?
The first question to answer is, if I take and White doesn't roll doubles
should I rewhip to 8? My gain-loss looks as follows:
Therefore, I would be risking 9% to gain 34%. A simple calculation shows that if he doesn't roll doubles I will win just over 1/4 of the time, so it is definitely correct for me to redouble if I choose to take and he doesn't roll doubles.
With this knowledge, we can calculate the equities:
Therefore, I am risking 22 1/2% in order to gain 76%. My mental estimates from this said I would have to win about 22 1/2% of the time to justify taking. I thought I remembered this position as having only about 21 1/2% winning chances, so I passed the double. Right or wrong, this is an illustration of how the calculations are done in the heat of battle.
I'm out of the main stuff, so not much to do but play in the blitz, doubles, and mini-matches. The blitz is all 5-point matches, played in brackets of 8. If you win 3 matches you qualify for the finals, which will be a bracket of perhaps 64 or less to be played tomorrow. If you lose, you can of course reenter. The mini-matches are quickies -- two point matches. Will probably be a bracket of over 256. The doubles are consultation with a partner, and is a bracket of 32.
My partner Herb Avram and I have won two doubles matches so far. In addition I have won three mini-matches, but of course there is a long way (and a lot of luck) to go there. In the blitz, after more failures than I care to admit to I finally managed to win my bracket. Here was a crucial position in my final match:
So I'm alive in the minor events, and maybe I'll get lucky in one of them. If so, there should be more interesting stuff tomorrow.
There was no good news for me, as I bowed out of everything quickly. Here is a deceptive position from our doubles match:
White is on roll. What should the cube action be?
Our opponents doubled, and I had to twist my partner's arm to get him to take. Actually I think it is pretty easy take. Once Blue enters White's five-prime doesn't figure to be much of a problem, since Blue has plenty of checkers to play with. Even if White escapes a couple of checkers, Blue will have excellent chances in a priming battle. If Blue doesn't enter, White will have to overcome a couple of roadblocks before he can extricate his back men, and if he fails to do so and his board starts to crunch Blue will become the favorite. Even if everything goes White's way, Blue will almost certainly have a very well-timed ace point game. Actually I think the double is marginal in theory. However when I presented this position to several good players, many of them initially thought the question was whether White should be playing for a gammon! In practice our opponents were forced to break their six point, and we wound up winning the game but losing the match.
Here is an interesting game from one of the jackpots between Paul Magriel and Rick Barabino. Magriel is ahead 6 to 1 in a 15 point match.
Magriel Barabino 1. 6-5 24/13 3-1 8/5 6/5 2. 5-3 8/3 6/3 4-2 8/4 6/4 3. 5-4 13/8 13/9 6-1 13/7 8/7 4. 6-4 8/2 6/2 4-3 13/9 13/10Certainly correct to try to extend the prime rather than put a back checker where it can be attacked.
5. 4-3 9/5 8/5 1-1 10/8 9/8 6/5
7. 4-3 13/9 8/5This is a nice well-controlled play. Magriel doesn't want to have too many blots exposed because his back checker may come under attack. The builder on the five point is just where he wants it to be -- safe, yet aiming at the critical four point.
7. 13/9 13/10
8. 3-2 24/22 13/10Magriel is counting on his stronger inner board to see him through a blot-hitting contest. I do not agree with this approach. Moving up to the 22 point just asks to be attacked, and he still has those blots in his outer board to worry about. I think Magriel should be concentrating all his efforts on making the four point or the bar point, and keep his back checker out of harms way. Even if Barabino completes his prime, Magriel won't necessarily be lost. Magriel has the better timing, and if he can make another blocking point he can win the priming battle by forcing Barabino's prime to crack. But Magriel can't improve his board if he gets stuck on the bar. In addition, there is another reason why Magriel should play more passively:
I think Barabino missed a nice double here. Behind 6-1 in the match, this is just the type of cube you should love to send over. A clear advantage, volatility in the sky, and gammon potential looming for both sides. A great chance to put a bunch of points on the plus side of the card. In addition Magriel would have to be very cautious about redoubling, since things could get out of hand if the cube came flying back to 8. Of course had Magriel made my more conservative play, Barabino would have much less incentive to double.
8. 1-1 10/9 5/3* 3/2 9. 5-4 -- Double
Now the double is clear. If Barabino springs a back checker, he will lose his market by a mile. It looks like Magriel has to take it. From his point of view, if Barabino doesn't spring that back checker the prime will break, and Magriel will be right back in the game. Even if Barabino does escape, Magriel may have time to enter, construct a prime, and win the priming battle by containing the remaining back checker. There is some gammon danger if Barabino picks up one of the outfield blots, but that has to be risked. As usual, both the doubler and the taker properly look at the cup as half full rather than half empty, and envision the good things which might happen rather than the bad things.
10. Take 5-1 8/2 11. 2-1 B/24 10/8Magriel properly stays back. He doesn't make the same mistake twice.
11. 5-4 9/4 9/5
12. 5-3 13/8 9/6Magriel properly safties a blot rather than playing 13/8, 13/10, even though the latter play would leave him more builders. His back checker is likely to be attacked, and he won't be able to use those builders if he is on the bar. Instead of being builders, they will become targets.
12. 5-1 8/3 4/3 13. 2-1 6/4 5/4 5-5
Magriel has a very clear redouble, even at the match score. He has first crack at rolling a six, and if he does roll the first six he will have a huge advantage. If he stays on the bar, Barabino may fail to roll a six and be forced to crunch his board. In addition maybe the double should be passed, or maybe Barabino will pass.
It looks like a clear pass for money. At this score, however, it might be another matter. Barabino can use a big game very much, and if he takes he will be potentially weilding a powerful 8-cube. If he wins the six-rolling battle he won't have to escape his second back checker. Magriel simply won't be able to take the recube. If both sides roll a six and Barabino winds up with a direct shot he can redouble on the come, and if he hits the shot he might even win it all by gammoning Magriel with the cube on 8. That outfield blot of Magriel's is a crucial factor. Even if Magriel wins the six-rolling contest the bar point remains open, which means that Barabino will have chances to roll boxes or hit a fly shot, and if he does either of these over comes the 8-cube. In practice Barabino did pass, but I thing he should have shot it out.
No good news for yours truly in the last chance. Here are the winners of the major events:
Open Main Winner: Rick Barabino Finalist: Mika Lidov Semi-finalist: Dirk Schiemann Semi-finalist: Leo Fernandez Open Consolation Winner: Johannes Levermann Finalist: Dennis Culpepper Semi-finalist: Tom Zarrinam Semi-finalist: Harry Cohn Open Last Chance Winner: John O'Hagan Finalist: Bob Glass Super Jackpot Winner: Paul Magriel Finalist: Ray Fogerland Semi-finalist: Rick Barabino Semi-finalist: Mike Senkiewicz Intermediate Main Winner: Sho Sengoku Finalist: Lise Howard Semi-finalist: Joe Moore Semi-finalist: Guy Thurber Intermediate Consolation Winner: Ted Bougton Finalist: Mark Damish Semi-finalist: David Orandle Semi-finalist: Hugo Spagenburg