1. Rollout 13/11 10/5 Eq.: -0.250
0.447 0.125 0.007 - 0.553 0.175 0.008 CL -0.157 CF -0.250
[0.001 0.001 0.000 - 0.001 0.001 0.000 CL 0.002 CF 0.005]
Truncated cubeful rollout (depth 11) with var.redn.
1296 games, Mersenne Twister dice gen. with seed 817112407 and quasi-random dice
Play: 0-ply cubeful prune [expert]
Cube: 0-ply cubeful prune [expert]
2. Rollout 24/22 11/6 Eq.: -0.269 ( -0.018)
0.446 0.115 0.007 - 0.554 0.180 0.006 CL -0.173 CF -0.269
[0.001 0.001 0.000 - 0.001 0.001 0.000 CL 0.003 CF 0.006]
Truncated cubeful rollout (depth 11) with var.redn.
1296 games, Mersenne Twister dice gen. with seed 817112407 and quasi-random dice
Play: 0-ply cubeful prune [expert]
Cube: 0-ply cubeful prune [expert]
3. Rollout 24/22 13/8 Eq.: -0.311 ( -0.061)
0.435 0.118 0.008 - 0.565 0.187 0.008 CL -0.200 CF -0.311
[0.001 0.001 0.000 - 0.001 0.001 0.000 CL 0.003 CF 0.006]
Truncated cubeful rollout (depth 11) with var.redn.
1296 games, Mersenne Twister dice gen. with seed 817112407 and quasi-random dice
Play: 0-ply cubeful prune [expert]
Cube: 0-ply cubeful prune [expert]


I was hoping someone could explain to me why 13/11 10/5 is best here, and why 24/22 11/6 is so much better than 24/22 13/8 which was my first instinct in this position.

I'm basically an enthusiastic beginner but I'm gonna try answering anyway.

13/11 10/5 slots the 5-point and you've got a good chance of covering it next turn. I'm mildly surprised it's better than 24/22 10/5 though, since 13/11 strips the midpoint, costing you some flexibility, to make a somewhat irrelevant point on the 11, which likely won't stand for long. 13/8 strips the midpoint as well, while accomplishing even less.

As for the merits of 11/6, I can't say. I suppose it cleans up a blot, but it wasn't in any particular danger, and you lose a potential builder for the 5-point. Instinctively, I think I would have preferred 13/8 to that one.

I think perhaps the problem with moving up the 24 blot is that it puts in range of of a lot of dice rolls next turn (7,8,5,3 all hit) and at this point this is no advantage to moving it. It is relatively safe right now (gnubg is not gonna hit/cover the 24 point) and it is more important to take teh offensive and start trying to build a prime.

So the question remains as to what to do with your other checkers. Moving 13/8 isnt particularly good as sure it gives you an extra checker to hit from the 8 point but nothing else. Slotting the 5 point is good because you can cover with rolls of 6,8,3,1 next time and gives you a wide range of rolls to hit back if your opponent gets a 4 and hits.

Moving 13/11 I suppose is the best of a bad bunch, keeping 2 checkers 6 points from your blot on the 5 point is useful as this is the most likely distance you will be able to move thus increasing your chances of covering next turn. Moving any other checker 2 points adds very little to your overall position.

FWIW, its a lot easier to justify moves once you have done the rollouts but i think some of the logic above is ok.

I think that 13/11 10/5 is a nice balancing move that builds the front position against reds two back checkers. Note that there is no way for red now to hit the slotted checker on the 5-point and at the same time make good use of his four strong builders.

The monster roll 31 e.g. is duplicated now because it would also hit on black's 5-point and would also make the bar point.

When red makes one of his key points (bar and 5) black is in a good position to cover his 5-point. creating a decent blocking positon as well - against the two red back checkers (compared to our single straggler).

Now let's have a look at the alternatives. 24/22 11/6 doesn't build the front position and creates some flexibility problems because of the checker piles. At the same time the black straggler is now in reach of red's builder. If red points on the stragglers's head black will have serious trouble because red has still builders to extend his prime. If red does not achieve this he is very likely to make on of his key points. Now the threat will be even stronger and if the straggler doesn't escape right now trouble is looming again. Notice how black can afford to concentrate on his attack because red threatens absolutely nothing (except 31 or 11 or similar flukes...)

24/22 13/8 has the same drawbacks as 24/22 11/6 but it leaves an additional blot in the outfield. If the straggler now gets under attack there are two additional checkers to be picked up by black.

I don't believe stripping the midpoint is such a major issue, because red needs to build a front position very quickly now.

To this I'd say, of course the blot is "safe" on 24 because it's not an asset and there's very little to be gained by hitting it. Moving it to 22 gives a bit better coverage of his outfield and is the first step to extricating it before it gets trapped worse than it already is. Unless he's lucky enough to make a point on it with 53, getting hit is only a minor setback, and if it turns into a blot hitting war, it would be to your advantage. There's really not much risk in moving to 22. The risk is sitting on 24 for too long.

Brown benefits a lot from slotting the 5 because basically all of the rolls red could hit it with are monster rolls regardless of brown's play - the only one that doesn't make a point in red's prime is 5-4, which hits the blot on 10 anyway if you leave it there.

What I don't see is why 24/22 11/6 and 24/22 13/8 are better than 24/22 10/5 though. I suppose a double hit here with 5-4 or 4-3 could be really bad.

There are several things you want to notice about this position.

You’re three moves into the game, your position isn’t great, and you just rolled what’s probably your worst number. As yet you have no new points, some big stacks, and a bunch of blots. Meanwhile your opponent has made a good point (the 4-point) and is poised to make another good point somewhere on the next turn.

The big virtue of 13/11 with your deuce is that it makes a point, and it’s the only point you can make this turn. In the opening, all points are good, even seemingly innocuous outer board points. A point gives you control of more space at zero risk. That’s good.

Once you play 13/11, the only reasonable five is 10/5. But that’s OK. You’re slotting a key point, and many of the numbers which hit would make your opponent’s bar-point instead. He will hit if he can, but the net cost to you is lower since he had other good things to do with the roll.

What’s wrong with 13/8 24/22? A couple of things, as a matter of fact. First, it strips the midpoint and makes a bigger stack of checkers on the 8-point. Stacks are bad, because they mean your whole army is operating at low efficiency. The idea in the opening is to unstack your checkers on the 6-point and the 13-point, not to build new stacks somewhere else. Stripped points are bad, because you don’t have the same ability to build. If you want to build a point using a checker from the midpoint, you have to actually give up the midpoint.

Second, the 22-point is not a place where you want your back checker. He’s better off back on the 24-point, where White won’t attack him. Remember that when you have two checkers back on the 24-point, you split them in part to build an advanced anchor. With one checker back, that’s not possible. So you leave the checker back until you can run him out and hit something in the process, or until he’s in danger of being primed.

Third, after 13/8 24/22, you’ve still got three blots, which is a little too many for a position with no points to support. Your position isn’t critical yet, but it’s drifting in that direction.

Thanks for the analysis. Once you break it down like that 13/11 10/5 really does seem like the obvious move.

Well you also asked why 24/22 11/6 is so much better than 24/22 13/8.
Here is what I think:
They are both not that good, they both stack checkers and do the worthless 24/22.
But what makes 13/8 far worse then 11/6 is that
1.you keep a spare on the midpoint with 11/6
2. you cleaned up one blot with 11/6 , eventhough in a ugly stodgy way , but hey 13/8 is also ugly and stody, but has one extra blot.

Hope this helps, and that it was correct, offcourse you should play the GNU move here (13/11 10/5) , and that is waht I think i would have done over the board.
But I hope my explanation of why move 2 was better than move 3, was somehat educational, if not a little simple though.

Popeye!