You wouldn't do the calculation over the board to figure out winning percentage in the first position you posted, you should just memorize the percentages for 2, 3, 4, and 5 roll positions. Then you can use those numbers in simpler calculations or to help make an informed guess.

For instance, the second position you posted, you can do a little calculation and be 100% sure you have a take in a money game. My reasoning over the board would go like this:

White's clearly better than a 3 roll position, so he's more than 21%. How much better? If Black rolls 2 consecutive 2s, it turns a very likely loss into a very likely win. 2 consecutive 2s happens around .3 * .3 so like 9%. A lot of that 9% should get added to 21% probably pushing me above 25% right there. And there's other stuff going on, like double 1s and doubles 2s don't save a roll for Black, so surely this well above 25% and an easy take.

Yes, that's the correct reasoning. Similar reasoning applies whenever your opponent needs an even number of points post-crawford. If he needs an odd number of points, you should pretty much take everything, as losing 1 point is almost as bad as losing 2. If that's not obvious, think carefully about the 1-away 3-away score post-crawford.

It's obvious, but what is not obvious is that I would be a big dog after these two rolls. It's not even obvious to me that I am a dog after these two rolls. I guess he is ahead on the race and has escaped a checker and I failed to hit, but I also have a lot of builders.

Now that it's happened I'm guessing there are a lot of sequences like this that I need to be aware of and drop more often - I've been assuming these were automatic takes.

Awesome, this is what I was looking for to help me understand this chapter better, but I honestly have bigger leaks to fix.

All of these are from the same game:








I start thinking that I am making progress than I play a game like this.

You're losing mostly because he's on roll.

Join the club.

I have the feeling you aren't making multiple 0.1 and 0.2 errors in a single game.

Latest problem position I had. Actually had quite a few in this game, but this one I struggled with a lot.

So, it's a match to 3 and I've doubled.

I immediately look at 9/8* 8/4*, and this was my thought process over the board

Benefits:

1. Puts two checkers on the bar
2. Increases gammon chances
3. Slots a point in my home board

Cons:

1. Leaves 13 direct return shots
2. Of those 3 shots (44, 45) send back two men and I probably lose
3. His timing is good if he gets in and I probably lose a lot of games

Alternatives:

• 6/1 - retains my flexibility and gives me a few more rolls to come in safely. He's given up the 4 point and I can make that safely with 52, 32, 22. Not a lot, but something.
• 7/3 7/6 - Ugg. Gives up a landing spot for my checkers on the 9 and stacks up the 6 even more.

SO... I choose 6/1 after a significant amount of thought.

What to do?

Ok, so I'm thinking it through a bit more:

- I only really dislike the sequence where he hits me and I fail to get in which is ~ 1/3*1/2 = 1/6. Do I lose all of these? no, but I am in big trouble with my blot on the 9 and his strong home board.
- If he fails to get in (2/3) I'll cover with any 2,3, 5 which I think is 27 rolls, or 3/4 which is 3/6 games
- 2/6 games I probably have a ton of equity after the double hit and he fails to come in.

So, I win 5/6 and lose 1/6. I know this isn't correct but I'm trying to come up with a method I could use over he board to estimate the situation.

Now, I need to compare that to 6/1. Which I'm not sure what to do... but the double hit is looking better now.

Anyone playing a long time would pass in a second. Think about it - you can play this game for the match and win the match 45% of the time. Or you can pass, start a new game for the match which you win 50%.

Stick with it. These things come easier in time. It's a difficult game and don't expect everything to come together right away.