On 1/14/2023 2:14 PM, ah....Clem wrote:
XGID=-BbBDBCa--A-b-a--a-c-bc--A:1:-1:1:51:0:0:3:0:10
X:Player 1 O:Player 2
Score is X:0 O:0. Unlimited Game, Jacoby Beaver
+13-14-15-16-17-18------19-20-21-22-23-24-+
| O O | | O O O | +---+
| | | O O O | | 2 |
| | | O O | +---+
| | | |
| | X | |
| |BAR| |
| | | |
| | | X |
| | | X X |
| O | | X X X X O X |
| O X O | | X X X X O X |
+12-11-10--9--8--7-------6--5--4--3--2--1-+
Pip count X: 87 O: 144 X-O: 0-0
Cube: 2, O own cube
X to play 51
The only real decision here is whether to enter on the ace or the 20
point. The difference between 6/5 and 4/3 after bar/20 is negligible,
and after entering with the ace 10/5 is the only reasonable play.
Why not enter with the five? Well, it exposes triple shot, plus the
shots at the outfield blot. Granted, threes are duped but X is almost guaranteed a return trip to the bar. The thing is, if X enters on the
ace point he's trapped behind a blockade which might be as bad as being
on the bar.
This came up in the facebook online "match" and Neil Kaz, Wil Snellings,
and I all opted for entering with the ace. Host David Potter had this to
say:
Let's look at the position and try to figure out what's going on and why
this is a bad play for us humans. We have a huge lead, the cube is
turned and we have a much stronger board at the moment. Our game plan
should be to race our checkers home but Black has a great outfield to
hit us when we move our back checker out. Bar/24 10/5 keeps us safe next
turn, but that still leaves us with the problem of getting that back
checker home through Blacks mine field. We would also have the problem
of rolling some bad low numbers which would force us to kill checkers to
the 1 and 3pts. Now let's look at coming in on the 20pt. First thing to
notice is that 3's are duplicated. Black's great numbers are
44,61,63,31,33&11 which all make the 5pt on our head. But they would be
great numbers anyway, even if we came in on the 24pt. There are however,
plenty of Black rolls that are not very good or that hit and leave blots
in hitting range. Bar/20 fits in with our game plan of running our back
checker home and keeping a strong board. I don't know if this analysis
will actually help anyone over the board as the best move is obviously
so anti-human. But maybe there is something to be learned here.
In the variant, I've moved O's outfield checker so that threes are not duplicated.
XGID=-BbBDBCa--A-b-a--a-c-bc--A:1:-1:1:51:0:0:3:0:10
X:Player 1 O:Player 2
Score is X:0 O:0. Unlimited Game, Jacoby Beaver
+13-14-15-16-17-18------19-20-21-22-23-24-+
| O O | | O O O | +---+
| | | O O O | | 2 |
| | | O O | +---+
| | | |
| | X | |
| |BAR| |
| | | |
| | | X |
| | | X X |
| O | | X X X X O X |
| O X O | | X X X X O X |
+12-11-10--9--8--7-------6--5--4--3--2--1-+
Pip count X: 87 O: 144 X-O: 0-0
Cube: 2, O own cube
X to play 51
1. Rollout¹ Bar/20 6/5 eq:+0.118
Player: 55.38% (G:27.66% B:1.64%)
Opponent: 44.62% (G:7.63% B:0.26%)
Confidence: ±0.009 (+0.109..+0.127) - [86.4%]
Duration: 2 minutes 17 seconds
2. Rollout¹ Bar/20 4/3 eq:+0.110 (-0.007)
Player: 55.11% (G:27.90% B:1.69%)
Opponent: 44.89% (G:7.79% B:0.26%)
Confidence: ±0.009 (+0.101..+0.120) - [13.6%]
Duration: 2 minutes 24 seconds
3. Rollout¹ Bar/24 10/5 eq:+0.044 (-0.074)
Player: 53.87% (G:21.57% B:0.83%)
Opponent: 46.13% (G:5.16% B:0.17%)
Confidence: ±0.008 (+0.036..+0.052) - [0.0%]
Duration: 2 minutes 46 seconds
¹ 1296 Games rolled with Variance Reduction.
Moves: 3-ply, cube decisions: XG Roller
eXtreme Gammon Version: 2.10
*///////////////////
*/ variant
*///////////////////
XGID=-BbBDBCa---Ab-a--a-c-bc--A:1:-1:1:51:0:0:3:0:10
X:Player 1 O:Player 2
Score is X:0 O:0. Unlimited Game, Jacoby Beaver
+13-14-15-16-17-18------19-20-21-22-23-24-+
| O O | | O O O | +---+
| | | O O O | | 2 |
| | | O O | +---+
| | | |
| | X | |
| |BAR| |
| | | |
| | | X |
| | | X X |
| O | | X X X X O X |
| O X O | | X X X X O X |
+12-11-10--9--8--7-------6--5--4--3--2--1-+
Pip count X: 88 O: 144 X-O: 0-0
Cube: 2, O own cube
X to play 51
1. XG Roller++ Bar/20 11/10 eq:+0.171
Player: 56.15% (G:27.65% B:1.55%)
Opponent: 43.85% (G:7.20% B:0.27%)
2. XG Roller++ Bar/24 11/6 eq:+0.055 (-0.116)
Player: 53.49% (G:20.39% B:0.79%)
Opponent: 46.51% (G:5.11% B:0.19%)
3. XG Roller++ Bar/20 4/3 eq:+0.032 (-0.139)
Player: 52.11% (G:25.92% B:1.48%)
Opponent: 47.89% (G:9.13% B:0.34%)
4. XG Roller++ Bar/20 6/5 eq:+0.030 (-0.141)
Player: 52.02% (G:25.80% B:1.45%)
Opponent: 47.98% (G:9.12% B:0.34%)
eXtreme Gammon Version: 2.10
Without the duplication of threes, XG still massively favors entering
with the five, but plays the ace as 11/10 to dupe threes instead of
shuffling homeboard checkers.
--
Ah....Clem
The future is fun, the future is fair.
--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)