Poll
After 5/2 5/3 the following shakes force us to leave a blot: 61, 64, 51, 54, 41, 42, 43
After 5/off, we blot with: 65, 64, 54, 43
after 3/off 2/off we blot with 64, 61, 54, 51, 43, 42, 41,
Looks pretty clear to me. 5/off.
quote moved inside spoiler - Zorba
spoiler tag added - Zorba
Question: 
Option 1: 5/2 5/3
Option 2: 3/off 2/off
Option 3: 5/off
Title: Fibsboard Forum Match 17 Game 2 Move 44 Herd to play
Post by: stog on January 16, 2014, 01:51:46 PM
Post by: stog on January 16, 2014, 01:51:46 PM
3 2
Title: Re: Fibsboard Forum Match 17 Game 2 Move 44 Herd to play
Post by: ah_clem on January 16, 2014, 05:15:31 PM
Post by: ah_clem on January 16, 2014, 05:15:31 PM
Spoiler
After 5/2 5/3 the following shakes force us to leave a blot: 61, 64, 51, 54, 41, 42, 43
After 5/off, we blot with: 65, 64, 54, 43
after 3/off 2/off we blot with 64, 61, 54, 51, 43, 42, 41,
Looks pretty clear to me. 5/off.
[close]
Title: Re: Fibsboard Forum Match 17 Game 2 Move 44 Herd to play
Post by: stog on January 16, 2014, 11:30:44 PM
Post by: stog on January 16, 2014, 11:30:44 PM
thx clem.. any other opinions or should i move on..? ;)
Title: Re: Fibsboard Forum Match 17 Game 2 Move 44 Herd to play
Post by: Zorba on January 16, 2014, 11:45:10 PM
Post by: Zorba on January 16, 2014, 11:45:10 PM
Spoiler
It wasn't so clear to me, although the 5/off move could well be right for all I know.
But there are few more things to consider here, apart from how many shots leave a blot next turn. The problem of leaving either 3 or 2 checkers on one of your highest point (especially the sixpoint; here it's the fivepoint), is known as double jeopardy. Whereas leaving 3 checkers is safer for the short term, you often end up moving the spare later and then you end up with the 2 checkers situation later on, so you may actually be taking more risk overall, compared to going for clearing the point right away. One way of solving this problem in general, is to look at how bad it would be to get hit soon, rather than get hit later.
Here, it's clear that el_micho's board and outfield coverage, is likely to improve over the next few turns. So that actually points a little towards stripping the fivepoint, trying to get this thing over with as soon as possible.
For this particular position, there are other things to note: el_micho's chances to win are very low to start with. If he hits a blot, then what? His sixpoint will be hard to make anytime soon, if at all. The four- and twopoint are slotted for now, and it's unlikely they both get made anytime soon. A four point board with the sixpoint open, is about the best el_micho can hope for, over the next few moves. It's hard to estimate how we are doing should we get one blot hit after our next turn, but we should still be massive favorites I think.
But what if two blots get hit, either in one go or one by one? It looks to me like that is what could give el_micho some real winning chances here. Not only do we need to enter both blots before we can run around again, giving el_micho some time to improve his board or outfield; we also run a much higher risk of being hit again when we need to run 2 men around the outfield.
Since 5/off leads to double blotting with a triple shot after both 64 and 54 next turn, whereas 5/3 5/2 is unlikely to ever lead to more than one blot, I actually voted for 5/3 5/2. However, the shot count is still a pretty strong argument, so I'm far from sure. It's just that the quality of shots could also be important here.
But there are few more things to consider here, apart from how many shots leave a blot next turn. The problem of leaving either 3 or 2 checkers on one of your highest point (especially the sixpoint; here it's the fivepoint), is known as double jeopardy. Whereas leaving 3 checkers is safer for the short term, you often end up moving the spare later and then you end up with the 2 checkers situation later on, so you may actually be taking more risk overall, compared to going for clearing the point right away. One way of solving this problem in general, is to look at how bad it would be to get hit soon, rather than get hit later.
Here, it's clear that el_micho's board and outfield coverage, is likely to improve over the next few turns. So that actually points a little towards stripping the fivepoint, trying to get this thing over with as soon as possible.
For this particular position, there are other things to note: el_micho's chances to win are very low to start with. If he hits a blot, then what? His sixpoint will be hard to make anytime soon, if at all. The four- and twopoint are slotted for now, and it's unlikely they both get made anytime soon. A four point board with the sixpoint open, is about the best el_micho can hope for, over the next few moves. It's hard to estimate how we are doing should we get one blot hit after our next turn, but we should still be massive favorites I think.
But what if two blots get hit, either in one go or one by one? It looks to me like that is what could give el_micho some real winning chances here. Not only do we need to enter both blots before we can run around again, giving el_micho some time to improve his board or outfield; we also run a much higher risk of being hit again when we need to run 2 men around the outfield.
Since 5/off leads to double blotting with a triple shot after both 64 and 54 next turn, whereas 5/3 5/2 is unlikely to ever lead to more than one blot, I actually voted for 5/3 5/2. However, the shot count is still a pretty strong argument, so I'm far from sure. It's just that the quality of shots could also be important here.
[close]
Title: Re: Fibsboard Forum Match 17 Game 2 Move 44 Herd to play
Post by: stiefnu on January 17, 2014, 12:14:46 AM
Post by: stiefnu on January 17, 2014, 12:14:46 AM
Spoiler
I like ripping two more off 3-off, 2-off. May leave a couple more shots later but his board ain't too scary and we're miles in front.
[close]
Title: Re: Fibsboard Forum Match 17 Game 2 Move 44 Herd to play
Post by: stog on January 17, 2014, 09:35:40 AM
Post by: stog on January 17, 2014, 09:35:40 AM
thx everyone--
i have been asked by someone to slow down (even leave votes for 2 days) but some or many of the previous moves might have been 'decided' by the facilitator of the match-- sometimes, so i have often made the call to move on with a quick vote that is 4 up
the forum matches are 'individual' and unique and are much affected by who is facilitating -- how busy they are etc..
as a participant it is still possible to comment on previous vote-finished moves, and question perhaps the tactics at a later date..
indeed often a rollout appears at a later date (thx clem and z)
i liked diane's 1 day voting pattern with the occasional extension for difficult moves ..
it is a feel on when/how to move the match on -- and it is also affected by work/life commitments---and routines (and time zones of course)
the tail end of many matches has often occurred on fibs chat, then been reported back -- but voting has continued here in this game while there has still been 'contact'.
we will have a couple of new facilitators helping us officiate future matches as well...
i may have been moving this match a little fast but 2 days for each move is i feel too long -- or that may be me ;)
it's definitely better as a decision for the individual facilitator rather than a 'prescriptive' working practice i think..
this move for example will remain open as it is close and i must go out... bye
i have been asked by someone to slow down (even leave votes for 2 days) but some or many of the previous moves might have been 'decided' by the facilitator of the match-- sometimes, so i have often made the call to move on with a quick vote that is 4 up
the forum matches are 'individual' and unique and are much affected by who is facilitating -- how busy they are etc..
as a participant it is still possible to comment on previous vote-finished moves, and question perhaps the tactics at a later date..
indeed often a rollout appears at a later date (thx clem and z)
i liked diane's 1 day voting pattern with the occasional extension for difficult moves ..
it is a feel on when/how to move the match on -- and it is also affected by work/life commitments---and routines (and time zones of course)
the tail end of many matches has often occurred on fibs chat, then been reported back -- but voting has continued here in this game while there has still been 'contact'.
we will have a couple of new facilitators helping us officiate future matches as well...
i may have been moving this match a little fast but 2 days for each move is i feel too long -- or that may be me ;)
it's definitely better as a decision for the individual facilitator rather than a 'prescriptive' working practice i think..
this move for example will remain open as it is close and i must go out... bye
Title: Re: Fibsboard Forum Match 17 Game 2 Move 44 Herd to play
Post by: Julia_H on January 17, 2014, 11:37:55 AM
Post by: Julia_H on January 17, 2014, 11:37:55 AM
Spoiler
In addition to the likelihood of a shot, and of a double shot, there's the question of what happens if we are hit. If we've ripped the man off the 3, then coming on with 21 or 41 is a bit embarrassing. There will be other possibilities for elmicho to winkle out another shot as well. However, if we rip two then we are better positioned to race should we have difficulty escaping to the point where elmicho leaves our board. Finally, leaving a shot on the acepoint is an all-or-nothing proposition, while leaving one high in the board and being missed doesn't leave us gin, particularly if we haven't cleared the five. I don't think that outweighs the immediate reduction in shot numbers by ripping from the five, it just demonstrates how many factors can come into play!
[close]
Title: Re: Fibsboard Forum Match 17 Game 2 Move 44 Herd to play
Post by: KDP on January 17, 2014, 12:37:21 PM
Post by: KDP on January 17, 2014, 12:37:21 PM
Spoiler
QuoteLooks pretty clear to me. 5/off.first we have to LEAVE a shot, then he has to HIT shot, then he has to CONTAIN the shot. i would spend less time worrying about things that MIGHT happen and just take 2 off. with his broken board im not too worried.
[close]
quote moved inside spoiler - Zorba
Title: Re: Fibsboard Forum Match 17 Game 2 Move 44 Herd to play
Post by: Cigale on January 17, 2014, 12:58:44 PM
Post by: Cigale on January 17, 2014, 12:58:44 PM
Spoiler
Hi!
Not much to think about ... every checker off is a checker off that counts. 3/off; 2/off and nothing in that position would scare me away from that.
Not much to think about ... every checker off is a checker off that counts. 3/off; 2/off and nothing in that position would scare me away from that.
[close]
spoiler tag added - Zorba
Title: Re: Fibsboard Forum Match 17 Game 2 Move 44 Herd to play
Post by: ah_clem on January 17, 2014, 04:02:12 PM
Post by: ah_clem on January 17, 2014, 04:02:12 PM
Before doing the rollout I'd like to say that ripping off two is a very reasonable play and may well be best. It's probably what I'd do OTB if I was too lazy to count shots.
On the topic of match pace, I'll put in my $.02 - there are a few people who check fibsboard constantly throught the day, and for them three or four moves a day works well. But I don't think there are that many in that camp. Many more participants only check once a day or so and for them a more leisurely pace will work better. I'm thinking of someone who flips by every morning as part of their daily routine - leaving a vote open for 24 hours means they can just check once a day and not miss anything.
I'd suggest the following:
Leave the vote open for 24 hours unless
Anyway, that's how I did it when I facilitated. But that doesn't mean it's the only right way, I'm happy to let whoever facilitates do it whatever way they want to.
On the topic of match pace, I'll put in my $.02 - there are a few people who check fibsboard constantly throught the day, and for them three or four moves a day works well. But I don't think there are that many in that camp. Many more participants only check once a day or so and for them a more leisurely pace will work better. I'm thinking of someone who flips by every morning as part of their daily routine - leaving a vote open for 24 hours means they can just check once a day and not miss anything.
I'd suggest the following:
Leave the vote open for 24 hours unless
- it's clearly a "next", in which case simply make the move
- the vote is overwhelmingly lopsided (say 5-0)
- it's a no-contact race, in which case finish it on fibs chat or let the bot play it out automatically
- after 24 hours it's tied or very close with discussion still ongoing
Anyway, that's how I did it when I facilitated. But that doesn't mean it's the only right way, I'm happy to let whoever facilitates do it whatever way they want to.
Title: Re: Fibsboard Forum Match 17 Game 2 Move 44 Herd to play
Post by: ah_clem on January 17, 2014, 04:28:43 PM
Post by: ah_clem on January 17, 2014, 04:28:43 PM
rollout
XG says clearing the 5 point is ever so slightly better than ripping off two, and that taking off one checker is a .017 error.
I think what I missed is that after 5/off 64 and 54 leave two blots, so while 5/off leaves fewer blot-leaving rolls, four of them are particularly awful.
XGID=-aDCaDa----------b--babad-:0:0:1:32:4:0:1:5:10
X:Player 1 O:Player 2
Score is X:4 O:0 5 pt.(s) match.
+13-14-15-16-17-18------19-20-21-22-23-24-+
| O | | O O O O O |
| O | | O O O |
| | | O |
| | | O |
| | | |
| |BAR| |
| | | |
| | | X X |
| | | X X X |
| | | X X X |
| | | O X O X X O |
+12-11-10--9--8--7-------6--5--4--3--2--1-+
Pip count X: 37 O: 106 X-O: 4-0/5 Crawford
Cube: 1
X to play 32
1. Rollout¹ 5/3 5/2 eq:+0.864
Player: 93.21% (G:63.40% B:9.98%)
Opponent: 6.79% (G:0.00% B:0.00%)
Confidence: ±0.003 (+0.862..+0.867) - [98.7%]
Duration: 3 minutes 21 seconds
2. Rollout¹ 3/Off 2/Off eq:+0.859 (-0.005)
Player: 92.96% (G:67.60% B:9.33%)
Opponent: 7.04% (G:0.00% B:0.00%)
Confidence: ±0.004 (+0.856..+0.863) - [1.3%]
Duration: 5 minutes 00 second
3. Rollout¹ 5/Off eq:+0.848 (-0.017)
Player: 92.38% (G:67.05% B:9.40%)
Opponent: 7.62% (G:0.00% B:0.00%)
Confidence: ±0.003 (+0.845..+0.851) - [0.0%]
Duration: 2 minutes 31 seconds
¹ 1296 Games rolled with Variance Reduction.
Moves: 3-ply, cube decisions: XG Roller
eXtreme Gammon Version: 2.10, MET: Kazaross XG2
Spoiler
XG says clearing the 5 point is ever so slightly better than ripping off two, and that taking off one checker is a .017 error.
I think what I missed is that after 5/off 64 and 54 leave two blots, so while 5/off leaves fewer blot-leaving rolls, four of them are particularly awful.
XGID=-aDCaDa----------b--babad-:0:0:1:32:4:0:1:5:10
X:Player 1 O:Player 2
Score is X:4 O:0 5 pt.(s) match.
+13-14-15-16-17-18------19-20-21-22-23-24-+
| O | | O O O O O |
| O | | O O O |
| | | O |
| | | O |
| | | |
| |BAR| |
| | | |
| | | X X |
| | | X X X |
| | | X X X |
| | | O X O X X O |
+12-11-10--9--8--7-------6--5--4--3--2--1-+
Pip count X: 37 O: 106 X-O: 4-0/5 Crawford
Cube: 1
X to play 32
1. Rollout¹ 5/3 5/2 eq:+0.864
Player: 93.21% (G:63.40% B:9.98%)
Opponent: 6.79% (G:0.00% B:0.00%)
Confidence: ±0.003 (+0.862..+0.867) - [98.7%]
Duration: 3 minutes 21 seconds
2. Rollout¹ 3/Off 2/Off eq:+0.859 (-0.005)
Player: 92.96% (G:67.60% B:9.33%)
Opponent: 7.04% (G:0.00% B:0.00%)
Confidence: ±0.004 (+0.856..+0.863) - [1.3%]
Duration: 5 minutes 00 second
3. Rollout¹ 5/Off eq:+0.848 (-0.017)
Player: 92.38% (G:67.05% B:9.40%)
Opponent: 7.62% (G:0.00% B:0.00%)
Confidence: ±0.003 (+0.845..+0.851) - [0.0%]
Duration: 2 minutes 31 seconds
¹ 1296 Games rolled with Variance Reduction.
Moves: 3-ply, cube decisions: XG Roller
eXtreme Gammon Version: 2.10, MET: Kazaross XG2
[close]