This article originally appeared in the April 2003 issue of GammOnLine.|
Thank you to Kit Woolsey for his kind permission to reproduce it here.
"DON'T MAKE THE ACE POINT".
I can still remember myself screaming this at the captain in the chouette many years ago in the following position:
He finally went along with me and left the ace point slotted, playing 18/15. It all went like clockword. They left a shot, and we hit. They rolled an ace, entering and recirculating our lead checker. After much jockeying around we finally forced them to jar a second checker loose, hit it, and won the game. After it was over, the captain said: "Well, I sure learned something that game".
Was not making the ace point correct? I really don't know. It was probably wrong. We could certainly use the second checker, but it would have been quite possible to win the game without it. Meanwhile, there were a lot of bad things which could happen. White might have rolled a six after entering instead of another ace, and now we would have to hit the checker on the way around or likely lose a gammon (or even a backgammon). Even if we did force White to play ace and leave another blot, we would still have to hit the blot. If we failed to do so and White closed up his ace point, his racing chances would improve. If we had the full prime then certainly making the ace point would be wrong, but with our bar point open it is far from clear. Despite the favorable result, I now believe we should have completed the closeout.
The above scenario occurred back in the 70's, when all the experts were very purity oriented. Making the ace point was always considered to be a bad thing to do. The checkers were thought to be out of play. No longer could a pretty prime be established. Flexibility was lost. In fact, the ace point was called "the guff", and making it was the mark of a beginner.
The following position (or something close to it) was from a problem set which Paul Magriel had constructed in that era.
When Kent Goulding, who was a student of Magriel's at that time, showed me this position, I thought about it for a while and then said: "I hate to admit it, but I think I would make the ace point". It turned out that I was ahead of the times. Most players of that era played 13/7(4) in an effort to remain pure and go for the priming game. That is very wrong here. This is a tactical battle, with both sides having one checker back. Also, the 6-6 roll puts Blue well ahead in the race, which means that he figures to come out second best in a priming battle. White has many rolls which make his bar point or five point, and it is vital for Blue to get White on the bar and prevent him from continuing to develop his position. 13/7(4) simply follows the wrong game plan. Today, with the help of the bots, I have no doubt that almost all experts would make the ace point without giving it a second thought, but in those days everybody tried to play purely and making the ace point was considered awful.
Riddle: What is worse than finding a worm in an apple you are eating. Answer: finding half a worm. The same is true of the ace point. Making the ace point may be bad, but making half an ace point is worse. This was not understood by the players of the 70's. Interestingly enough, many of them had no aversion to hitting loose on the ace point in the early stages of the game -- if anything, they tended to overdo it. They understood the importance of tempo and keeping the opponent off-balance. However, their followup was not correct. Their idea was to leave the ace point slotted, so that eventually the checker will be hit and get recirculated. For example:
Playing B/23, 13/7 would have been considered automatic several years ago. Why make the ace point when you can make the bar point? Today, this sort of position is seen in a different light. That blot on the ace point is a thorn in Blue's side. If Blue doesn't cover it, it will sit there waiting to be hit at exactly the wrong moment. Not only will being hit on the ace point result in 24 lost pips in the race, but it will put Blue on the bar which could be disastrous in the future. In addition, 13/7 strips the midpoint. Blue is better off playing B/23, 7/1.
The following position gives us a good feeling about the philosophy of making the ace point. It occurred in the 1996 World Cup final between myself and Malcolm Davis:
At the table, I made the ace point. I felt it was rather clear, but mentioned to Malcolm that the commentators probably wouldn't like the play (this was on closed-circut TV). Malcolm agreed. That was correct. Kent Goulding was one of the commentators, and in his writeup in Inside backgammon he said that he thought long-term purity in this sort of complex position was more important than the short-term tactical gain of making the ace point. He showed the position to Nack Ballard and Mike Senkiewicz, two of the top players in the world, and independently they both preferred the "purer" play of 13/11*, 6/1*. It should be mentioned that at the time Malcolm and I had already had much experience with Jellyfish, since we had copies of it before it became public, while other players did not have this experience since the product had just come out on the market. It looks like this is going to be a long complex game, and the question is whether or not making the ace point will be an albatross hanging over my head. In my opinion, then and now, the spare checker on the three point was far more of an albatross then the made ace point, and a blot on the ace point was worse than making the ace point. My guess is that our improved understanding of the game would have making the ace point much more popular, and that almost all experts would do so now. At the time there was still the carryover of avoiding making the ace point in complex games, and that stopped the good players from making the play.
So, when should making the ace point really be avoided? The main time is during a priming battle. If you have an advanced anchor, then the ace point becomes just another inner board point. However, if you are possibly stuck behind a prime and trying to make a prime of your own, then owning the ace point can be very bad.
Making the ace point on White's head is thematically wrong. This is a priming type of game, not a blitzing type. It is correct for Blue to play the pure 13/8, 13/7 and try to build his four and five points.
The other main time when making the ace point may be thematically wrong is when playing a deep anchor or back game. Ideally, when you hit your shot you would like to trap the hit checker behind a blockade and jar another checker loose. However, you have to have the time to do this. Many players overdo this concept. For example, another problem from Magriel's problem set (or something close to it):
Back in the 70's, there were many players who would have played 6/2, 6/1 with this roll. Their idea was to avoid making the ace point at all cost, even if it meant piling a bunch of spares on the other points. The idea was that if the ace point remained uncovered and they hit a shot, they could eventually recirculate the checkers one by one as White was forced to enter with an ace. This concept would be fine if you could call the rolls of the dice, but you can't. The correct play is the obvious one of making the ace point, with the plan of collapsing the board from the back and hoping to hold onto as much as possible while keeping a smooth position. Keeping the ace point open in the hopes of recirculating checkers simply isn't going to work.
This is the type of position where it is correct to avoid making the ace point. Blue will probably be able to escape one of the checkers on White's three point next turn, and after that he will have plenty of time to maneuver and build a blockade. He cannot afford to waste one of his checkers to make the ace point. He simply leaves the point uncovered, and if he eventually hits a shot and White rolls an ace Blue won't mind one bit -- that will get the checker back into play so Blue can use all 15 checkers to contain the hit checker and jar a second checker loose.
In the old days, experts would play 8/5, 3/2. The idea, of course was to avoid making the ace point at all costs. In their minds, after they hit their shot White would enter with an ace and recirculate a checker. Then they would roll an ace, hit loose on the ace point, White would hit back, and another checker would get back into play. In the meantime, maybe White would get trapped on Blue's ace point and be forced to jar another checker loose. All this is too much dreaming. It could happen, but there are far too many things which could go wrong. Blue should play the simple 8/5, 2/1, and try to win by hitting just one checker and keeping it on the bar. This isn't at all farfetched. If White were down to just three checkers on the two point for a potential coup classique then the approach of not making the ace point would probably be correct, but in this position it is carrying things too far.
When it comes to blitzes, the ace point is just as powerful as any other point for keeping the opponent on the bar. In fact, it is often better to make the ace point than a higher point. The reason is that a higher point is easier to reach with the outer board builders. No self-respecting student of the game would get this one wrong:
You can throw the pure 9/7, 8/7 out the window. This is blitz, not a prime, and gammons do count double. It is clear for Blue to play 3/2*, 3/1. This puts a second checker on the bar and gives White only good twos. If White enters with a three or a four, he will probably be attacked again. Making the bar point gives White both aces and twos to survive. Note how the shift to the ace point activates the builders on the eight and nine points so they are now aimed at more of the key inner board points Blue needs to make. This theme is quite common.
One never knows when the ace point may come in handy. Even if an opponent isn't on the bar now, he may be there in the future, and owning that ace point may deprive him of a key roll. So often have they laughed when I made my ace point, only to have that laughter turn to groans when they roll 1-1 from the bar.