the dorbel daily

Sunday, 20 June 2010

Split Or Slot

At the moment, Fibsboard is running one of its Guest v. The Rest matches and yours truly is in the hot seat. The team opened with a 2-1 and there was some intelligent debate which covered almost all the main points in the slot or split debate. The two plays 24/23, 13/11 and 13/11, 6/5 are thought to be more or less equal, with many players switching between the two depending on score and/or the quality of the opposition. I like to split when leading in the match and at dmp, while usually slotting when trailing to catch the small extra gammon chance. The theory also goes that the games are likely to be more complex after the slot, so should be chosen by the stronger player to increase his chances of outplaying his opponent. I believe this to be true, as the weaker player sometimes mishandles the blot hitting contest which often follows the slot and also does less well when the point gets made and a priming contest results. This can be so marked that I sometimes choose the slot even when leading against some opponents, but whether I gain by this is hard to say.
What can the beginner/intermediate player learn from this? I would say that if you are not already comfortably playing at an advanced level the split is best for you and in fact you should probably be splitting with almost everything that doesn't make a point. The reason I say this has little to do with the type of game we can expect though. The decision about when to split the back checkers is one of the hardest decisions that we have to make in the early part of the game and it rarely gets easier as the game proceeds. Spare yourself a tough play later by making a habit of splitting early, on the opener if you get it and on the response if not.
If asked, I would say that this is certain to improve your win rate. This is how you should be playing the opening roll to suit your skill level if you are not yet an advanced player. This probably means everybody who is normally below 1750 on Fibs.
6-5...24/13 of course.
6-4...24/14 for simplicity.
6-3...24/15 rarely tried but simple and as good as anything else.
6-2...24/18, 13/11.
6-1...13/7, 8/7 of course
5-4...24/20, 13/8
5-3...8/3, 6/3
5-2...24/22, 13/8
5-1...24/23, 13/8
4-3...24/20, 24/21 rarely seen, but always my choice when leading and at dmp.
4-2...8/4, 6/4
4-1...24/23, 13/9. At dmp 24/20, 24/23 is fun and reasonable.
3-2...24/21, 13/11
3-1...8/5, 6/5
2-1... 24/23, 13/11 (NEVER 13/10)
Stick to these, see if you win more matches! I'll be interested to hear from anybody who tries these. Enjoy The Game!

Thursday, 10 June 2010

Waving The Flag

In a comment on my last post, Ah_Clem points out the omission of positions where unstacking is wrong. This is of course fair comment. However, I think that if challenged he could easily produce a list of occasions where other priorities are greater for himself. 6-5 in the opening usually stacks a sixth and even a seventh checker on the midpoint. Hitting a blot, or making a useful point, or safetying a blot can and usually are important enough to leave the stack in place for the moment and of course we often roll a number that can't unstack without leaving a vulnerable blot anyway. What the flag is doing is serving as a reminder that we have a glaring positional weakness that needs to be dealt with urgently. It can't replace thought! Usually, more important goals like the above will be obvious, but the flag will help us when the answer isn't clear, not infallibly I agree but even when it is wrong it won't usually be wrong by much.

This is the start of a three point match with Red to play 6-1. 16/9 is an obvious play, but I liked 13/7, 8/7, unstacking and making a useful point. Actually, it is probably slightly inferior, losing a few more gammons than the alternate, but there is no way to know that over the board, so following the flag saves you wasting mental energy trying to puzzle out the best play, energy that may be better expended later. Even if I could play God for a moment and know that 16/9 was very slightly better theoretically, I might still choose to make the bar. Why? Partly because making the bar gains a clear asset, partly because I think White's play will be slightly more difficult next turn. In general, I like to complicate things, although as White is a bot here, that's probably not a good idea!

Because stacks have this need to unstack themselves, they should also carry a second flag warning the opponent, "Danger! Unstable structure. Do not approach". Stacks are very bad at building primes but they do just fine for hitting and blitzing. Here's a case in point. What would you do as Red here?

White's position has a non-obvious weakness here, which is the stack on the 6pt. Those checkers are just dying to get into the game and as the 3pt is already made, they really would like to go on the 5 or the 4pt. If Red, as actually happened in the game, makes the mistake of moving to either (or both) of those points, White will be delighted to hit. If Red stays where he is, White will have to try and develop them more naturally, something that stacks are just not good at. I'm not sure what the best play is here. 13/9, 8/5 is feasible and probably outranks the similar 13/10, 8/4 because it slots a better point. 8/4, 7/4 could be best, swapping the bar point for the 4pt and of course strengthening the board. However, I do know that anything that splits at the back is a big blunder. Don't give stacks the chance to fall on you!
More to come on this subject, but until I find the time to write the next bit, Enjoy The Game!

Friday, 4 June 2010

More About Flags

Well the last post engendered an overwhelming response of precisely zero comments. No doubt the fact that it was dull, badly written and featured a not-very-interesting position had something to do with it. Let's try and do a bit better on the same subject. The title by the way, "Do You Have A Flag?" was inspired by Eddy Izzard, go to to get the idea.
So, flags for stacks, take a look at this one.

It's 0-0 in a match to 7 points and Red has a 5-2 to play. He is a beginner and chose 20/13 and it was my job to explain to him why this was not the best play. "It stacks an extra checker on the midpoint, which is bad" was my response.
"Yes, but why is it bad". Reasonable question, why is it? Stacks are bad because they concentrate a high proportion of your checkers in one place, making your position less flexible and harder to improve next turn. On roll here and trailing by 40 pips, Red has four points made with spares on three of them. If he plays 20/13, he now has spares on only two points, thus giving him fewer choices next turn. Clearly making the bad stack on the midpoint worse is not the way to go.
The point of having our "Unstack Me Now" flag on the midpoint is to save a lot of thinking. That's the place that we need to play from and 13/6, or conceivably 13/8, 6/4 is the way to go. While we are at it, how about 13/8, 3/1? That leaves us spares on four points, which is definitely good, but buries a checker on the ace point, which is definitely bad. Why is it bad? Because it reduces our playable checkers from 15 to 14 and reduces choice in the same way as a stack does, but worse, because that checker can never go anywhere. You can turn it into a small asset by making the point of course, but that cuts us down to 13 playable checkers!
Experienced players will look at positions like this and see that stacking another checker or burying one is bad just by instinct, but it is a learned instinct. The point of the flag on the midpoint is to give you a short cut to the place from which your play should usually come. It doesn't replace thinking of course, but it may save a bit of mental energy.

The same player faced some more unstacking problems in the next game. here's one of them.

In the last game, Red was getting the worst of a positional struggle of the type known as a mutual holding game, where both sides held high anchors and improving position was the over-riding aim. Here Red has doubled early and has White struggling to survive, but Red must still look for positional improvement and the "Unstack Me Now" flag still flutters from the top of the midpoint stack. Red will hit with the 5 of course, but which 4? He played 20/15*, 20/16, getting those back checkers moving and of course playing with complete safety for this turn. It's not a bad play, there aren't many bad plays when your opponent has two men on the bar against a four point board, but he can do better and 13/9 is the best 4. Not only does it unstack the midpoint, but we can now find a use for two more auto signals that I find very useful. I like a neon arrow, flashing and directed at the point that I want to make next, which here is clearly the 6pt. Obviously a checker in the outfield to hit there if White manages a 6 from the bar is good. Not only that, but the entire outfield needs a large VACANT sign. Red has no points there and this is always a weakness. Making any point there will improve Red's chances of coming home safely, both to act as a safe landing point and to provide a reservoir of checkers to hit on and/or make 6pt.
Flags, neon arrows, VACANT signs, all help me to keep my attention on the essentials. Helpful to you? Let me know. Until the next time, enjoy the game!