Match 1, Game 2, Move 2, Forum

[blitzxz]

6-2?

Score:
Forum(black) 1, Factotum 0, lenght 5 points

Pips:
Black 161, White 158

GnuID:
4HPwCSDg8+ABKA:cAmrAAAACAAA

[lewscannon]

Spoiler

Not a whole lot to do with this roll. 13-5 at least maybe gets you the 5 point. Moving one of the back pieces will probably only get you hit.

[close]

[Zorba]

Spoiler

Hmm! I'm far from sure on this one, quite a challenge how to make the best of an awkward roll. Nothing plays really well here. I used the method of successive elimination here:
Moving both backmen up is usually a bad idea. It makes attacking too easy and valuable for your opponent: he can now attack and slot a great point at the same time, maybe also the next move. With both moved up, there's an extra risk of an immediate double hit by opponent.

Running one backman with either 24/16 or 23/15 walks into a double direct shot and gives factotum a very easy path to cleaning up his outfield blot.

13/5 is a decent slotting play, not bad in itself but it risks getting a third men back against factotum's one and lose a lot of ground in the race, which is otherwise about equal. It may make factotum's safetying of his outfield blot easier. Still, this looks like a good candidate play.

24/18 is a pretty good split here, catering both for a defensive or waiting style holding game if we can make that barpoint, or an attempt to run a backchecker in a more race-oriented play. It also makes it harder for factotum to safety his outfield blot, creating superior coverage all over the board. It's a very connective play.

The bad part is the 2, which doesn't play well. 13/11 positions a builder well, but that extra blot and direct shot could be quite costly.

8/6 keeps things safe on our side of the board, but creates an ugly stack on our 6pt.

I'm far from sure, but I wasn't prepared to play our entire roll to slot the fivepoint and prefer the split, also to put more pressure on factotum's blot. So 24/18, and then I thought we could live with the ugly play for now, since it still leaves a builder on the eight point and the sixpoint stack will find its way eventually, hopefully. I'd jjust rather not get hit here, when there's a chance we may be hitting factotum next turn instead. I could see any of these three plays being right though. Will be interesting to see comments and bot stuff on this one!

[close]

[lewscannon]

Spoiler

Hmm! I'm far from sure on this one, quite a challenge how to make the best of an awkward roll. Nothing plays really well here. I used the method of successive elimination here:
Moving both backmen up is usually a bad idea. It makes attacking too easy and valuable for your opponent: he can now attack and slot a great point at the same time, maybe also the next move. With both moved up, there's an extra risk of an immediate double hit by opponent.

Running one backman with either 24/16 or 23/15 walks into a double direct shot and gives factotum a very easy path to cleaning up his outfield blot.

13/5 is a decent slotting play, not bad in itself but it risks getting a third men back against factotum's one and lose a lot of ground in the race, which is otherwise about equal. It may make factotum's safetying of his outfield blot easier. Still, this looks like a good candidate play.

24/18 is a pretty good split here, catering both for a defensive or waiting style holding game if we can make that barpoint, or an attempt to run a backchecker in a more race-oriented play. It also makes it harder for factotum to safety his outfield blot, creating superior coverage all over the board. It's a very connective play.

The bad part is the 2, which doesn't play well. 13/11 positions a builder well, but that extra blot and direct shot could be quite costly.

8/6 keeps things safe on our side of the board, but creates an ugly stack on our 6pt.

I'm far from sure, but I wasn't prepared to play our entire roll to slot the fivepoint and prefer the split, also to put more pressure on factotum's blot. So 24/18, and then I thought we could live with the ugly play for now, since it still leaves a builder on the eight point and the sixpoint stack will find its way eventually, hopefully. I'd jjust rather not get hit here, when there's a chance we may be hitting factotum next turn instead. I could see any of these three plays being right though. Will be interesting to see comments and bot stuff on this one!

[close]

Spoiler

I thought of these options too, but came to the conclusion that even if the 5 gets hit after the 13-5, it's no big deal. You have a good shot of making a point in his home board, since you're already on the 24 & 23. The split move (to me) doesn't seem to get anywhere. I'd much rather be aggressive. The 5 point, if made, looks like it will definitely come in handy this game, as I have a feeling there's gonna be some hitting going on

[close]

[NIHILIST]

I slot the 5 point in these situations.

Bob

[diane]

Spoiler

I didn't agree with Lews this time, so I am probably wrong    I went for the 13-7 and 23-21.  That just seemed more useful than 24-22 - which didn't either give us a chance at a more forward anchor, nor stay back in case it turns  out to be necessary. I gave lots of thought to slotting that 5 though, and may well do that as many times as I choose this one.

[close]

[ah_clem]

I vote 23-21 13-7

Usually it's a mistake to split and slot simultaneously. But in this case, our back men are already split, and there's no outfield bearing on the 21 pt, so why not move one up to a better point, ie one we actually want to make, instead of the useless 23? 

If we weren't already split, I'd say slot the 5.  But since we're already split, I don't like that play.

[Zorba]

Spoiler

If you're gonna slot a point, it should be the fivepoint. It is a clearly better point to have than the barpoint, especially against one man back. Also, there are less shots at the fivepoint slot than at the barpoint slot, so it survives more often. So, more gain and less risk with 13/5 as far as our side of the board goes.

On the other side of the board, 23/21 has some value in that it guards the outfield better and prepares for a (more) advanced anchor, but combined with the slot this also makes us vulnerable on both sides of the board (don't slot and split). 13/5 is also slot and split, however that split is minimal and therefore much less vulnerable to a succesful attack.

[close]

[diane]

Spoiler

One thing is sure, whatever happens to the checkers, the vote is definitely split  The slot on the 7 for me isn't really about slotting, it is moving the 6 somewhere because I have to. I would expect it to get hit, not to stay, and if I get hit on the 7, that is better for me next go, than the hit on the 5. Hence my choice.

[close]

[blotsalot]

I like  24-18 23-21 here for several reasons:
1 it creates no new blots.
2 it "forces" the action onto the far side of the board and tends to keep your opponents blots in place , and therefore targets for the future, if he is to attack you now.
3 it preserves chances to either build  the opponents bar point OR his 4 point if either blot gets hit or pointed on.
4 it maintains connectivity.

All that being said, I like 13-5 also as it pays long term rewards if you succeed in covering it in the very near future.


JD

[ah_clem]

The thing that finally tipped me in favor of 23-21 13-7 is what happens after you get hit.  With this position, you're more likely to make a useful anchor after entering. Since getting hit is likely, might as well anticipate it.

That said, I'm far from certain about this as the right play.

[Zorba]

I only have a quick and dirty 0-ply rollout for now, so it's not gospel, but 13/5 seems best, mostly because it wins more gammons than some of the best alternatives (over 13% instead of 11% for some of the plays that run a backchecker). For pure single winning chances, the rather surprising 24/18 23/21 wins.

        Full cubeful rollouts with var.redn.
        648 to 7776 games, Mersenne Twister dice gen. with seed 818332620 and quasi-random dice
        Play: 0-ply cubeful prune [expert]
        Cube: 0-ply cubeful prune [expert]

    1. Rollout          13/5                         Eq.:  -0,0959
        46,17  13,62   0,89 -  53,83  12,70   0,68 CL  -0,0633 CF  -0,0959
      [  0,07   0,09   0,06 -   0,07   0,07   0,03 CL   0,0023 CF   0,0058]
    2. Rollout          24/18 23/21                  Eq.:  -0,1108 ( -0,0149)
        46,70  11,38   1,01 -  53,30  13,39   0,57 CL  -0,0926 CF  -0,1108
      [  0,07   0,09   0,06 -   0,07   0,07   0,04 CL   0,0023 CF   0,0058]
    3. Rollout          24/18 13/11                  Eq.:  -0,1152 ( -0,0193)
        46,17  11,13   0,76 -  53,83  12,82   0,64 CL  -0,0959 CF  -0,1152
      [  0,09   0,10   0,07 -   0,09   0,08   0,04 CL   0,0027 CF   0,0070]
    4. Rollout          24/22 13/7                   Eq.:  -0,1365 ( -0,0406)
        45,26  12,45   0,77 -  54,74  12,42   0,61 CL  -0,0901 CF  -0,1365
      [  0,10   0,12   0,07 -   0,10   0,09   0,06 CL   0,0033 CF   0,0083]
    5. Rollout          24/18 8/6                    Eq.:  -0,1368 ( -0,0409)
        46,05  10,72   0,74 -  53,95  12,76   0,54 CL  -0,1026 CF  -0,1368
      [  0,10   0,12   0,07 -   0,10   0,09   0,04 CL   0,0031 CF   0,0072]
    6. Rollout          23/21 13/7                   Eq.:  -0,1368 ( -0,0409)
        45,49  13,12   0,92 -  54,51  12,97   0,70 CL  -0,0909 CF  -0,1368
      [  0,11   0,13   0,08 -   0,11   0,09   0,06 CL   0,0035 CF   0,0083]
    7. Rollout          23/15                        Eq.:  -0,1664 ( -0,0705)
        45,26  11,04   0,99 -  54,74  13,34   0,65 CL  -0,1275 CF  -0,1664
      [  0,23   0,28   0,25 -   0,23   0,21   0,14 CL   0,0081 CF   0,0191]
    8. Rollout          24/16                        Eq.:  -0,1909 ( -0,0950)
        45,27  11,37   1,11 -  54,73  13,72   0,67 CL  -0,1250 CF  -0,1909
      [  0,23   0,38   0,26 -   0,23   0,26   0,10 CL   0,0068 CF   0,0180]

[playBunny]

Rollout 1296 trials, 2-ply for 12 plies, 0-ply thereafter

Code Select Expand


# .. Move ......... Equity .. (Diff) .... Win ... WinG ... WinBg -- Lose ... LoseG . LoseBg
1 .. 13/5 .......... -0.089 ............ 46.4% .. 13.3% ... 0.8% -- 53.6% .. 12.7% ... 0.7%
2 .. 24/18 23/21 ... -0.126 (-0.037) ... 46.5% .. 11.3% ... 1.3% -- 53.5% .. 13.9% ... 0.5%
3 .. 24/18 13/11 ... -0.130 (-0.041) ... 46.2% .. 11.2% ... 0.7% -- 53.8% .. 13.3% ... 0.6%
4 .. 24/22 13/7 .... -0.135 (-0.046) ... 45.3% .. 12.8% ... 0.7% -- 54.7% .. 12.8% ... 0.7%
5 .. 23/21 13/7 .... -0.138 (-0.049) ... 45.5% .. 13.0% ... 1.1% -- 54.5% .. 13.0% ... 0.8%
6 .. 24/18 8/6 ..... -0.146 (-0.057) ... 45.7% .. 10.6% ... 0.7% -- 54.3% .. 12.8% ... 0.6%
7 .. 24/16 ......... -0.161 (-0.072) ... 45.7% .. 11.2% ... 0.7% -- 54.3% .. 13.9% ... 0.7%
8 .. 23/15 ......... -0.167 (-0.078) ... 45.5% .. 11.5% ... 0.8% -- 54.5% .. 14.0% ... 0.6%
9 .. 24/22 8/2 ..... -0.175 (-0.086) ... 44.4% .. 11.9% ... 0.8% -- 55.6% .. 13.7% ... 0.7%
10 .. 13/7 8/6 ...... -0.189 (-0.100) ... 44.1% .. 12.7% ... 0.9% -- 55.9% .. 13.5% ... 0.7%
Std Err .............. 0.011 ............. 0.1% ... 0.2% ... 0.1% --- 0.1% ... 0.2% ... 0.1%

        Full cubeful rollout with var.redn.
        1296 games, Mersenne Twister dice gen. with seed 818003339 and quasi-random dice
        Play: supremo 2-ply cubeful prune [world class]
        keep the first 0 0-ply moves and up to 16 more moves within equity 0.32
        Skip pruning for 1-ply moves.
        Cube: 2-ply cubeful prune [world class]
        Different evaluations after 12 plies:
        Play: 0-ply cubeful prune [expert]
        Cube: 2-ply cubeful prune [world class]


My chosen move was 24/18 13/11 which was intended to either prepare for making anchors and blocking points or, hopefully, make trouble for the escaping backrunner. I didn't care for the 13/7 moves much because they gave options to hit the 7-point blot and continue out to make an anchor, or hit it and run the other escapee to the outer table on his home side without there being much threat. I rejected the 13/5 for the same reason but, as the rollouts show, it has the greatest advantage of all the moves if the slot doesn't get hit. I saw the second choice 24/18 23/21 but dismissed it immediately as I pretty much automatically consider it a runner-type beginner kind of move. But it has its moments and, if not for the 13/5, this would be one of them!

[playBunny]

# .. Move ......... Equity .. (Diff) .... Win ... WinG ... WinBg -- Lose ... LoseG . LoseBg
 1 .. 13/5 .......... -0.089 ............ 46.4% .. 13.3% ... 0.8% -- 53.6% .. 12.7% ... 0.7%
 2 .. 24/18 23/21 ... -0.126 (-0.037) ... 46.5% .. 11.3% ... 1.3% -- 53.5% .. 13.9% ... 0.5%
 3 .. 24/18 13/11 ... -0.130 (-0.041) ... 46.2% .. 11.2% ... 0.7% -- 53.8% .. 13.3% ... 0.6%
 4 .. 24/22 13/7 .... -0.135 (-0.046) ... 45.3% .. 12.8% ... 0.7% -- 54.7% .. 12.8% ... 0.7%
 5 .. 23/21 13/7 .... -0.138 (-0.049) ... 45.5% .. 13.0% ... 1.1% -- 54.5% .. 13.0% ... 0.8%
 6 .. 24/18 8/6 ..... -0.146 (-0.057) ... 45.7% .. 10.6% ... 0.7% -- 54.3% .. 12.8% ... 0.6%
 7 .. 24/16 ......... -0.161 (-0.072) ... 45.7% .. 11.2% ... 0.7% -- 54.3% .. 13.9% ... 0.7%
 8 .. 23/15 ......... -0.167 (-0.078) ... 45.5% .. 11.5% ... 0.8% -- 54.5% .. 14.0% ... 0.6%
 9 .. 24/22 8/2 ..... -0.175 (-0.086) ... 44.4% .. 11.9% ... 0.8% -- 55.6% .. 13.7% ... 0.7%
10 .. 13/7 8/6 ...... -0.189 (-0.100) ... 44.1% .. 12.7% ... 0.9% -- 55.9% .. 13.5% ... 0.7%
Std Err .............. 0.011 ............. 0.1% ... 0.2% ... 0.1% --- 0.1% ... 0.2% ... 0.1%

        Full cubeful rollout with var.redn.
        1296 games, Mersenne Twister dice gen. with seed 818003339 and quasi-random dice
        Play: supremo 2-ply cubeful prune [world class]
        keep the first 0 0-ply moves and up to 16 more moves within equity 0.32
        Skip pruning for 1-ply moves.
        Cube: 2-ply cubeful prune [world class]
        Different evaluations after 12 plies:
        Play: 0-ply cubeful prune [expert]
        Cube: 2-ply cubeful prune [world class]

[socksey]

I like  24-18 23-21 here for several reasons:
1 it creates no new blots.
2 it "forces" the action onto the far side of the board and tends to keep your opponents blots in place , and therefore targets for the future, if he is to attack you now.
3 it preserves chances to either build  the opponents bar point OR his 4 point if either blot gets hit or pointed on.
4 it maintains connectivity.

I don't remember how I voted, but this was my last look and choice.   

socksey



"The best thing about the future is that it comes only one day at a time." - Abraham Lincoln

[playBunny]

How you voted is show in bold in the list of poll options.

[socksey]

I thought so, but thanks for the confirmation! 

socksey



"The best index to a person's character is how he treats people who can't do him any good, and how he treats people who can't fight." - Abigail Van Buren