Open Side Menu Go to the Top
Register
2 cube decisions 2 cube decisions

03-23-2009 , 04:44 AM
These are from a tournament match i just played and wanted to get some feedback on my cube decisions. Taking a cue from the poker strat forums I won't say what actions I took. I did run these through GNU but wanted to see what others would do in these situations. I am the player being offered the cube in both.




2 cube decisions Quote
03-23-2009 , 06:05 AM
I assume Black is offering in both situations.

Number 1 might at most be a thin take, but you only have one 6 left before you must leave your anchor... I would rather drop

Number 2 is a tough one for me because I'm not too familiar with match play. But if you drop, Black's match equity will be 89%. That's so hopeless that I would take here - you should be better than 11% here. Maybe you can hit a fly shot, or you roll a large double to come back in the race...
2 cube decisions Quote
03-23-2009 , 10:06 AM
1) Easy take- Lots of ways to win this game

2) Pass
2 cube decisions Quote
03-23-2009 , 11:40 AM
1) I believe this to be a close decision. Red is 21 pisp behind but black hasn't made his 4-point yet which usually makes these positions more takable for red. If I should make up my mind I'll probably take but only by a small margin.

2) Seem like a very easy take based on the macth score. I wouldn't even double here.
2 cube decisions Quote
03-23-2009 , 11:46 AM
Doh! Didn't see the checker on the bar in position 2. Then I believe it's a pass, sorry.
2 cube decisions Quote
03-23-2009 , 12:36 PM
Both are passes.

1) Pass. Even without the 4 Point, Black has a strong enough home board to get home reasonably safely and may even send a red blot to the Bar. That combined with a 21 pip lead even before the coming roll means pass to me.

2) Pass - not even close IMO. Even if Black somehow doesn't make another point, red may dance a few times. Most Black rolls are at the very least "safe" for Black and many are down right good for him. Worst he can roll is 66, but just about everything else at worst leaves an indirect shot from the bar with a four point home board into which Red is rolling.

Last edited by Pot Odds RAC; 03-23-2009 at 12:46 PM.
2 cube decisions Quote
03-23-2009 , 01:12 PM
Quote:
Originally Posted by Pot Odds RAC
Both are passes.

1) Pass. Even without the 4 Point, Black has a strong enough home board to get home reasonably safely and may even send a red blot to the Bar. That combined with a 21 pip lead even before the coming roll means pass to me.

2) Pass - not even close IMO. Even if Black somehow doesn't make another point, red may dance a few times. Most Black rolls are at the very least "safe" for Black and many are down right good for him. Worst he can roll is 66, but just about everything else at worst leaves an indirect shot from the bar with a four point home board into which Red is rolling.
a) Will white win from this position more than 33% of the time....if the answer is yes, than it is a take. Black is no bargain to get home safely, in fact the roll 54 leaves a direct shot, any non double leaves an indirect shot, black could still win the race from here by rolling well. IMO this is an easy take....in fact i think a double by black is a mistake.
2 cube decisions Quote
03-23-2009 , 01:36 PM
Quote:
Originally Posted by buckwand
a) Will white win from this position more than 33% of the time....if the answer is yes, than it is a take. Black is no bargain to get home safely, in fact the roll 54 leaves a direct shot, any non double leaves an indirect shot, black could still win the race from here by rolling well. IMO this is an easy take....in fact i think a double by black is a mistake.
One roll that leaves a direct shot and still doesn't ensure a White win even if we do hit (Check me on this but I think we are only about a 36% chance to hit the direct IF it is presented). It may be closer than I am thinking, but I don't think white wins 33% of the time from this position.
2 cube decisions Quote
03-23-2009 , 05:10 PM
Quote:
Originally Posted by Pot Odds RAC
One roll that leaves a direct shot and still doesn't ensure a White win even if we do hit (Check me on this but I think we are only about a 36% chance to hit the direct IF it is presented). It may be closer than I am thinking, but I don't think white wins 33% of the time from this position.
I think the more interesting question here is ....what does black do with 5-4.....anybody??

I normally only play cash games with no cube.....so my cube calculations may be a tad rusty, but i would think if you add together whites running chances, his indirect shot chances( they are coming) ,his direct shot chances (maybe)...put them in a pot and stir....you come up with take.

Black to play 5-4?????
2 cube decisions Quote
03-23-2009 , 06:11 PM
Well, 54 sucks, but this roll is only 1/18 to fall. If it doesn't you still have to prove that your board doesn't crack. And if it does, you still have to hit.
2 cube decisions Quote
03-23-2009 , 06:20 PM
Quote:
Originally Posted by Viikatemies
Well, 54 sucks, but this roll is only 1/18 to fall. If it doesn't you still have to prove that your board doesn't crack. And if it does, you still have to hit.
I already think it's a take. Not to hijack the thread but if you are Black...what move do you make with 5-4
2 cube decisions Quote
03-23-2009 , 06:56 PM
54 is a 1/18 shot. And if I roll it I play it 13/4 - Red'll have 2s duplicated. And I don't think that one bad roll is enough on its own. Most of the time Black'll just get his midpoint checkers home safely.

The question is now whether Red's anchor leaves him with enough play to take even though the midpoint checkers most likely come home safely. I feel it's close, but I would drop.
2 cube decisions Quote
03-23-2009 , 07:08 PM
Quote:
Originally Posted by buckwand
I already think it's a take. Not to hijack the thread but if you are Black...what move do you make with 5-4
Playing as black on a 5-4 after a take I would play 5-1, 13-8.


You lose an attacker from the the 5 point but we have a spare and while making the 4 point would be nice it is not crucial and I would rather make the play the keeps us ahead in the race most often rather than the one that iincreasese the chances of making the 4 point. Also, the chances of red covering one of its 2 blots are lessened by having to hit on the 10 point.
2 cube decisions Quote
03-23-2009 , 07:45 PM
A quick rollout from GNU says 13-9, I3-8, and as usual my play was 3rd best

1. Rollout 13/9 13/8 MWC: 57.49%
0.588 0.094 0.007 - 0.412 0.114 0.003 CL 58.73% CF 57.49%
[0.002 0.001 0.001 - 0.002 0.001 0.000 CL 0.06% CF 0.08%]
Full cubeful rollout with var.redn.
1296 games, Mersenne Twister dice gen. with seed 820846360 and quasi-random dice
Play: 0-ply cubeful prune [expert]
Cube: 0-ply cubeful prune [expert]
2. Rollout 13/4 MWC: 57.45% ( -0.04%)
0.594 0.088 0.005 - 0.406 0.113 0.003 CL 58.83% CF 57.45%
[0.002 0.001 0.000 - 0.002 0.002 0.000 CL 0.06% CF 0.08%]
Full cubeful rollout with var.redn.
1296 games, Mersenne Twister dice gen. with seed 820846360 and quasi-random dice
Play: 0-ply cubeful prune [expert]
Cube: 0-ply cubeful prune [expert]
3. Rollout 13/8 5/1 MWC: 56.98% ( -0.51%)
0.583 0.079 0.004 - 0.417 0.113 0.004 CL 58.33% CF 56.98%
[0.002 0.001 0.001 - 0.002 0.001 0.001 CL 0.05% CF 0.08%]
Full cubeful rollout with var.redn.
1296 games, Mersenne Twister dice gen. with seed 820846360 and quasi-random dice
Play: 0-ply cubeful prune [expert]
Cube: 0-ply cubeful prune [expert]
2 cube decisions Quote
03-25-2009 , 03:18 PM
Just thought I would post what GNU had to say now that I have gotten some unbiased answers. I think it will be interesting to discuss the reasoning behind gnu's answers.

Putting these in spoiler tags in case anyone still wants to contribute without their answers being biased by one of the bots.

Position 1

Spoiler:


Cube analysis
3-ply cubeless MWC 60.00% (Money: +0.549)
0.758 0.058 0.003 - 0.242 0.027 0.001
Cubeful equities:
1. Double, take 62.81%
2. Double, pass 62.87% ( 0.06%)
3. No double 62.09% ( -0.72%)
Proper cube action: Double, take

Ok, 1st I need a little help interpreting this. Originally I thought that the proper cube action recomendations were for both me and my opponent. i thought that It was saying it is proper for both my opponent to double and for me to take.but since I tried the evaluation from both the double and take/drop position and the numbers did not change, I assume the numbers and recomendations are from black's perspective because obviously I am not 75% to win this game. Is Gnu evaluating black's cube actions if he either recieves or offers a cube here? If so, it is easy enough to reverse the numbers there.


The cubeful equities also confuse me a bit. Since I am trailing in the match at this point I have to believe these numbers are again from black's perspective, which would mean that it would be a pass from my perspective (although about as close a decision as you can get at just .06% difference.)

I did end up passing in this situation but knew it was a very close decision. The deciding factor for me was that I was trailing in the match and figured I would get better chances against this opponent in the future. If I had been ahead in the match and could afford to gamble a bit I would have accepted.


Position 2
Spoiler:




Cube analysis
3-ply cubeless MWC 80.99% (Money: +0.758)
0.871 0.017 0.000 - 0.129 0.001 0.000
Cubeful equities:
1. Double, take 87.10%
2. Double, pass 89.02% ( 1.93%)
3. No double 86.00% ( -1.09%)
Proper cube action: Redouble, take


So if I am interpreting Gnu's recomendaions correctly here. This is a correct double for black because it increases his MWC ~1% if red takes and 2% if they pass. It is also a correct take for me because my chances of winnning this game are ~2% better of coming back to win this game than dropping and winning 4 consecutive games (assuming no gammons.)

i ended up accepting this cube thinking it was a wrong decision because I was so pissed after he hit a miracle shot with a 6-5 as I was about to come home.So i actually made the "correct" play but actually thought it was wrong
2 cube decisions Quote
03-25-2009 , 04:32 PM
Quote:
Originally Posted by Pocket Trips
... because I was so pissed after he hit a miracle shot with a 6-5 as I was about to come home...
Ah, poker will never be able to beat backgammon when it comes to putting people on tilt

As for your solution - I've got it right: number one was really as thin as I thought and I was correct on number two as well.

This forum is really helping me to regain some of my backgammon practise that I lost with all these pokers in my mind...
2 cube decisions Quote
03-25-2009 , 05:27 PM
Quote:
Originally Posted by Viikatemies
Ah, poker will never be able to beat backgammon when it comes to putting people on tilt

As for your solution - I've got it right: number one was really as thin as I thought and I was correct on number two as well.

This forum is really helping me to regain some of my backgammon practise that I lost with all these pokers in my mind...
It would seem as though i am right out to lunch on cube analysis. I am not sure i understand how in situation 1 red will win 24% of the games and it is a pass......and in situation 2 red will win 13% of the time , but that is a take. What am i missing here ?
2 cube decisions Quote
03-25-2009 , 05:34 PM
Quote:
Originally Posted by buckwand
It would seem as though i am right out to lunch on cube analysis. I am not sure i understand how in situation 1 red will win 24% of the games and it is a pass......and in situation 2 red will win 13% of the time , but that is a take. What am i missing here ?
Maybe you've missed that it was match play. If Red drops in the second game he will only have 11% match winning chances which is less than the 13% in the game - and if Red takes and wins, he wins the match!
2 cube decisions Quote
03-25-2009 , 05:47 PM
Quote:
Originally Posted by Viikatemies
Maybe you've missed that it was match play. If Red drops in the second game he will only have 11% match winning chances which is less than the 13% in the game - and if Red takes and wins, he wins the match!
Gotcha....so, same position , cash game, it's a pass.
2 cube decisions Quote
03-25-2009 , 05:49 PM
A rather huge pass, I would think (-:
2 cube decisions Quote
03-25-2009 , 06:03 PM
Quote:
Originally Posted by Viikatemies
A rather huge pass, I would think (-:
Yes indeed.....i was looking at the problems from a cash game perspective.
2 cube decisions Quote
03-25-2009 , 06:20 PM
Apparently match score is also a pretty big factor in position 1 as well. I set up the same position but just reversed the match score with red leading 3-2 instead of trailing and that changed this decision from a pass to a take.

Cube analysis
3-ply cubeless MWC 47.10% (Money: +0.549)
0.758 0.058 0.003 - 0.242 0.027 0.001
Cubeful equities:
1. Double, take 49.86%
2. Double, pass 50.00% ( 0.14%)
3. No double 49.24% ( -0.62%)
Proper cube action: Double, take
2 cube decisions Quote
03-25-2009 , 07:10 PM
Quote:
Originally Posted by Pocket Trips
Apparently match score is also a pretty big factor in position 1 as well. I set up the same position but just reversed the match score with red leading 3-2 instead of trailing and that changed this decision from a pass to a take.
It was already a take in the original position wasn't it?
2 cube decisions Quote
03-25-2009 , 07:19 PM
Quote:
Originally Posted by mute
It was already a take in the original position wasn't it?
No. It is a bit confusing becaue the original position is from a match file I imported into gnu and the numberss are actually meant to represent my opponent's point of view. so it is actually a pass ( but only by .06%)
2 cube decisions Quote
03-25-2009 , 07:40 PM
As usual I am confused
2 cube decisions Quote

      
m