Hrqls: I'm not too familiar with crowded, but it occurs to me that with 5 opponent pieces on the bar, closing the 5 point is more important than hitting twice.

Hrqls: I would have made the same move in this position.

Hrqls: I think the move you suggest is a huge blunder and would love to try a computer analysis at some point. Your opponent has a great chance of running or blocking your 7 point etc etc.

I'm sure with that particular roll you must hit the opponent on your 7 point. In fact even with 6-4 you should play 13/7 24/20 and with 6-5 I would hit both (13/7 6/1).

Interesting problem: If you roll 6-6 do you play 13/7*/1* or 13/7* 24/18 ?

Hrqls: I'm sure in the situation I described there is no better move than the 2 I siggested. I think 24/22 is slightly better than 13/11 unless you really need a gammon.

What would you play? Anyone else have a theory?

Hrqls: If you don't hit and they block the 7 spot it's very good for them. High risk high return.

For example if you started with a 6-3 and moved 24/18 13/10, then your opponent rolled 6-2, his best move is to hit you with 13/7* 24/22 or 13/7* 13/11

alanback: I would agree if the universe of players was larger, however with a limited number of players who have that level of experience it doesn't seem too odd to me.
Thankfully the list includes the number of games, so we can easily look at it any way we want.

alanback: lol, i didn't even notice it was crowded. Just looked really quickly.

Hrqls: You had almost no chance to hit back, all he needed was a 7 or 8 on 2 dice, or a 1 or a 3 to block you entirely.

alanback: I disagree, I think it was too good to double.

Marfitalu: I believe the question was "why is this a draw and not a win and a loss?".

I think this is a question of definition on this site. A match is considered a single game no matter how many games are involved, dubious but probably easier to implement?

Hrqls: Taking the double would have been a major mistake.

Looks like he's going to win a single anyway, very little chance of a gammon. Some small chance you could come back and win though so doubling is probably correct.

playBunny: I agree it depends on the position, and what knowledge you gain, however I would err on the side of caution if you aren't sure.

pentejr Yes, the %age changes slightly depending on the match score, but not as much as you might think. You also have to account for gammons either way so a simple percentage is not useful until both players have borne off.

What you need to do is work out the chance of winning for each match score (there are tables) and figure it out from there.

So in a 9 point match, ignoring gammons, if I refuse my chance of winning from 0-1 is about 44%.

If I take the double I will either be at 0-2 or 2-0. The chance of winning from 0-2 is about 37% and, conversely, from 2-0 is about 63%.

So if my chance of winning the first game are currently x, then my chances of winning the match if I take are .37(1-x)+.63x = .37+26x

So to equalize to teh 44% chance if I drop, then .44 = .37+.26x
.26x = .07
x = .07 / .26
x = about 27%

Which, as you said, is a little more than 25%

:)

I think you shouldn't analyze until after the game (not the match, just the current game). If you analyze immediately after a move it may give you insight into the position that you wouldn't otherwise have that could be used next turn.

Pedro Martínez: Correct, on closer inspection one of the games is clearly rigged. That and the Asterix-like names I'd say the prosecution can rest.

Shall we petition for removal, and possible execution?

Pedro Martínez: Hard to say, the few games that aren't private look legitimate, the others at least took a few turns.

By the way can we please get rid of private games, they are not needed.

playBunny: Ah, but gammons and backgammons are counted, so you may have a 50% chance of winning but a positive equity if your chance of gammoning is better than your opponent. right?

Hrqls: Definitely a take in my opinion as well. You had an inferior position, but not that inferior. Control of the cube is important.

Hrqls: oops, i misread it, i thought it was 5-2 from the 8 point. In this case your move is probably best.

Hrqls: That's an interesting position. I'd probably make your 4 point instead, but putting both pieces on the bar with only one exposed may gain some time. I think it's probably a close call.

Hrqls: Here's and example of when you can play a double hit:

After opp rolls 4-3 and moves 24/20 24/21

Then you roll 3-2. Now you can play 8/5* 8/4*. The possibility of closing the 3 & 4 points is worth the risk. However hitting the 1 & 2 points is not helpful at all.

Hrqls: IN that example definitely not, too risky. I would play 24/20 13/8. Maybe 13/8 13/9 if you're going for a gammon. Problem is that with the split the piece on 9 is more vulnerable.

Which, by the way, is another reason splitting the back men is good, makes it easier to hit the opponent on the next turn. :)

Hrqls: I believe your question was meant to be "should i hit twice in my home?". The answer to this is not so clear. Chance of being hit is about 55%, but the advantage gained if not hit can be quite large.

I'd say it depends on how well developed your opponents home is (which would increase the risk) and how many builders you have ready to take advantage next turn.

Czuch Chuckers: pair up on 18, not even close I don't think. The anchor on opponents 7 is very good.

pentejr: 2-2, 3-3, 4-4 and 6-6 are indeed 4 rolls.

However: 2-1, 1-2, 2-3, 3-2, 3-4, 4-3, 5-6, 6-5 are 8 rolls!!

remember non double rolls come up twice as often as double rolls.

In addition to playBunny's point that a double can be used well in your own home, whereas a 4-3 might not be so good without the split.

pentejr:
possible reasons

1. Four pieces is better than three (if you roll a double for instance)

2. Splitting the back with 24/23 is good - if you roll 6-5, 4-3 or 3-2 next turn for instance you get an advanced anchor.

3.Rollouts say 13/8 24/23 is better, so who am I to argue?

pentejr: oops, my mistake. In that case I vehemently disagree.
Your move may be correct in gammon save situations, but I think 13/8 24/23 is correct in most cases.

pentejr: I agree with 13/8 24/23. I never play anything else.

playBunny: I can't imagine it's impossible actually. All it would take is a quick check at the end of the turn, if the next player can't move then roll the dice and move for that player (using the current code but calling it now instead of waiting for a player action).

I don't think it breaks the design so much so as to make it so very difficult to code. On the other hand I won't know for sure unless I see the code.

playBunny: Hear hear

The concern over "chatting" is ridiculous. What's the difference between making one move or ten and then chatting? I don't get it. I think your reason is likely the only one.

playBunny: I disagree with the 24/22. Here is the whole text for context.

"Split your back checkers (24/23) and slot your five-point. It was right to split your back checkers until recently, when modern computer rollouts showed that slotting is usually better, but you can vary this depending on the match score."

Clearly it is comparing splitting to slotting and assuming the 13/11.

In fact later in the site it has the two options:

"Opening 24/23, 13/11 13/11, 6/5"

Czuch Chuckers: I understand what it says, I am telling you it is a typo and means what I said it means.

Czuch Chuckers: 2-1 would normally be 13/11 24/23

I think it probably means 13/11 is correct no matter what, and then either 24/23 OR slot the 5-point. Slotting the 5-point is better than it looks, especially if you need a gammon.

alanback: ditto, it's much better with the cube. And my rating skyrocketted since it's introduction ;)

Hrqls: 24/20(2) is correct if saving a gammon it seems.

playBunny: hmmm, i'm going to amend after some thought. In that situation 24/22(2), 6/4(2) is better, although 24/20(2) is obvious in a gammon save situation.

24/20, 6/4(2) is probably only good if you hit a blot on 20.

playBunny: I think I made a mistake, but want to be clear:

After the opponent plays 4-2 (which has to be 8/4, 6/4 or it's a mistake) then I roll a 2-2, I'm playing 24/20, 6/4(2).

I think I confused you into thinking I played 4-2 first.

What do the talkative players here think?

6-6: always 24/18(2), 13/7(2) except after 6-1 opening, then 13/7(2), 8/2(2)
5-5: usually 13/3, but sometimes 8/3(2), 6/1(2) if you hit a blot.
4-4: as playBunny suggests, except consider 24/16, 8/4(2) or 24/16, 13/9(2) if you hit a blot on 16.
3-3: Far too many options to fit.
2-2: Usually 13/11(2), 6/4(2) but sometimes 24/22(2), 6/4 (e.g. after a 4-2 opening) and occasionally 24/20, 6/4(2) if you hit a blot on 20.
1-1: either 8/7(2), 6/5(2) or 24/22, 6/5(2) depending on the situation. Don't forget to hit a blot on 20 though.

playBunny: I understood, but I don't even understand why you'd play a cubeless game if the cube is available, gammons or otherwise.

Walter Montego: Why take the cube out? it's much more interesting with it.

Pythagoras: Hey, that looks close to what I thought!!

Hrqls: I was only comparing the other two openings since we were talking about gammon saves and gammon goes. Of course 24-18 13-9 is possibly best overall, particularly in light of playBunny's analysis.

playBunny: I would tend to make the 2-point in games where the gammons don't matter, I thought it was back in vogue as well.

Your numbers are not all that conclusive in the matter. It looks like they would favor making the 2 point when trying for a gammon though, any conclusion on that?

Pythagoras: With 6-4, making the 2 point is a good play in any situation, though 24-14 is just as good usually. While going for gammon the 2 point should be made and while saving a gammon 24-14 is correct, but the cubeless equity of either move is about the same I think.

Thad: Hear Hear. I guess it's part of the game engine for all games though.

playBunny: You'll need to improve your comedic routine before you impress me.

Walter Montego: Obviously 1-2 is different from 2-1 (though effectivey the same) so there is your 36.
Each player has their own set.

Hrqls: I think the removing of blocks is silly. The whole point is that by creating a block you get a certain type of advantage, but limit your options next turn etc.

I think the addition of the cube negates the point about this game being too much luck. A good 5 point game is pretty skillful.

Pedro Martínez: I believe that was the point. The U.S was planning to hold their own cup between their A and B teams and call the winner the world champion. Oh, wait, they already do that in some sports.

Czuch Chuckers: Since the cube came in my hyper rating has skyrocketted, which supports the view that single games are too much about luck. If you want a fairer rating play at least 5 point cube matches.

playBunny: But it would happen quite often if 100 people play 500 moves a day (once a week or so).