I think 99% would get this one wrong.

I think you are right!

This is a critical time for White. He is stripped on his 6pt, and stacked everywhere else. Anything that helps him play safely is probably a mistake.

Over the board, I would likely run a blot from the the 21pt to the 14pt. When you post this problem here, however, hinting that there is some QF (quiz factor) involved, I an inclined to suspect that giving White any target in the outfield is wrong. Make him his inside, where there will be many returns. In addition, running from the 21pt means that White will only face a single shot if he lands there.

Offering White the chance to make a tempo hit in the outfield could be fatal. If White can just squeeze by with a blot in the outfield for one turn, he may well convert that blot into a much needed point. If he can do that, his chances of playing safely on subsequent turns go way up.

It looks like Black needs his midpoint much more than his 20pt. This may be a fine time to break the 20pt and escape to the safety of the midpoint. That move also buys some more time for Black while he waits for the inevitable shots that White must cough up sooner or later.

Late thought: I wonder whether Black should be trying to make White's bar point. If so, then having three builders on different points might be better than owning the 5pt. Looks like I need to do some "play-from-here" practice with this one. If making the bar point is a theme, it should show up soon enough.

Nice position!

Mike

Thank you, Mike.

Thinking it over, I don't think the bar point is an issue here, because white has enough time to throw some double that will move him off his 13 point.

White's few escape points and a very weak vs strong board motivate black to increase his hitting chances to the maximum. Only a few doubles from white could be detrimental with his very weak board, but black has also some gammon potential if he hits.

Black needs to hit. The obvious move is 21/14, but possibly not the best. Moving 20/13 may seem risky at first, but as mentioned, there really isn't much to worry about. Doing so maintains 3 checkers with which to hit white instead of 2. 20/13 may be the best move.

In a fast-paced game I probably would have moved 21/14; if 20/13 is the better move, I would have been part of the 99%.

Yep, I'd automatically play 21/14. Feels natural, not really risky, but what everyone's saying makes sense. Leaving the blot on the 20 point is less risky than normal and you want to get a shot in.

I'd go 20-13, hard for white to cover any points.

Ok white has no builders but its hard to break the golden point. Id still play 21/14

Fllecha this is another one of those psychological plays - I have memories of making the riskier move in these situations only to get spanked by my opponent's subsequent doubles.

Trailing by 20 pips is a factor to consider. If running from the 20 point is right (if it is) maybe you can consider it a "jam play" that is a play that make some confusion in an inferior position. In chess is some how standard vs a not so strong opponent. Here white position is stacked with bad distribution in home but, in white shoes, if black break the anchor id Be very happy

20/13 isn't really all that risky. The only extra good rolls for white are 44, 33 and 11. I can live with that for the extra shots at the outfield.

Like several posters already said: 21-14 OTB but 20-13 because of QF and thinking it through.

Right, with an equity difference of 0.9

Here is another more clear leave of the 5 point which I just met in a game:

Was it too good to double? I would have, this seems to be a chronic problem in my game.

One important thing. If black misses, white easily scores a gammon; if black hits, he rarely will score a gammon because of the open six point. It's even not sure he will hit the second blot, if white enters soon enough.

True. But black does hit on 29 rolls, sending white back to a 5 point board, with another blot loose. I double for sure.

Wow, that 2nd position is fascinating for sure. I would have played 24/18* with the 6 without a doubt.

Sorry, rightly formulated: giving up the 20 point in order to increase the hitting chance of the second blot.

And how would you play this one?

I would not give up the 24 point. The 24 is arguably a better anchor than the 20 here, and there's better stuff to do with the 4 than 24/20. I think the main candidates are 20/14 and 11/9 5/1*. The attack seems a little thin, so I'd go with 20/14. Good outfield control and we have the ace point game to fall back on if things go bad. Getting the back men out if things go well doesn't seem like a big problem.

I think the position in Post #12 is no-double/take. It's a good position to do a little "backgammon math". Assume double and take. 29 shots, so 7 misses. White can win all the 7 games by redoubling, but he can do a little better by playing on for the gammon. So give White 8 wins in those 7 games. Of the 29 hits, most of them only hit one checker, and then White enters in about a third of those, so he gets one checker hit and then enters right away in about 7 or 8 games or something. He's probably a favorite in that variation on average. So he gets 5 more wins there.

We're already up to 13 wins and we haven't considered all the variations. I get that White loses some gammons, but this is looking like too big a take to redouble. I don't think it's a beaver, but if it was only 27 shots, like if the White checker on the 9 point was on the 8 point, I think it actually would be beaverable.

About the position in the OP -- 20/13 would be a great play! I agree with the 99% figure. I would have found the right play if I considered it, but I'd be quite unlikely to consider it. This position is a strong argument to slow down and think about all the possibilities.

I certainly would have doubled this one, and I would have been delighted if my opponent would have taken, and next there would be anxiety.

It is even a 0.1 loss not to make this play.

The next problem is somewhat related.

I've made lots of huge mistakes doubling (and not doubling) these types of positions. I find it tough to go by the visceral feel of these positions -- the no doubles don't look that different from the passes some of the time. So now I always try to put some numbers on things like I did above. My numbers aren't going to be super accurate, but generally they'll be in the ballpark enough to avoid a big blunder and help guide me a bit if the decision is close.

If I'm remembering right (I don't have it in front of me), Walter Trice's book "Backgammon Boot Camp" has a nice section on this type of decision. Excellent book for any level of player, by the way. Starts off very basic but then later gets into some advanced concepts.

That big stack on the 20 point just looks like it's desperate to release a checker. Unfortunately, you have to step into a triple shot and make a concession on the home front to move one. I kind of think it's worth it though. So I'd play 20/15 7/6. The two blots in White's board are key. If his board was even a little more together, I'd just play 18/12.

Congratulations, you have succeeded in the test.

Maybe someone could do an XG rollout for the doubling problem.

With a GNU untruncated rollout of 400+ at worldclass, the verdict is: NO REDOUBLE. Doubling and wisely taking would cost black 0.28.