I have just played my first Axelisation game against XG.
As in the title, I didn't cheat by consulting the paper during the game. Furthermore, my settings don't give the pipcounts.
So, quite similar to OTB.
I know it's just one game but I'm really impressed by it all.
The position below was the first cube decision after breaking contact,
and I held the cube. I Axelised correctly for a hold.
And all simple methods hold too. The decision is somewhat close, though.

XGID=--ABCCB-AC--a----bbcbaca--:1:1:1:00:12:9:3:0:10
X:Daniel O:eXtremeGammon

Score is X:12 O:9. Unlimited Game, Jacoby Beaver
+13-14-15-16-17-18------19-20-21-22-23-24-+
| O O | | O O O O O |
| O O | | O O O |
| | | O O |
| | | |
| | | |
| |BAR| |
| | | |
| | | |
| X | | X X | +---+
| X | | X X X X | | 2 |
| O X X | | X X X X X | +---+
+12-11-10--9--8--7-------6--5--4--3--2--1-+
Pip count X: 82 O: 86 X-O: 12-9
Cube: 2, X own cube
X on roll, cube action

Analyzed in 4-ply
Player Winning Chances: 69.52% (G:0.00% B:0.00%)
Opponent Winning Chances: 30.48% (G:0.00% B:0.00%)

Cubeless Equities: No Double=+0.390, Double=+0.781

Cubeful Equities:
No redouble: +0.635
Redouble/Take: +0.591 (-0.044)
Redouble/Pass: +1.000 (+0.365)

Best Cube action: No redouble / Take
Percentage of wrong pass needed to make the double decision right: 9.7%

eXtreme Gammon Version: 2.10

The below Axelisation is closer. Again I Axelised correctly for a hold.
But it's now extremely marginal. The thinking goes:
Pip counts are 74 and 78. Add 2 for the opponent's two extra crossovers
to get 74 and 80. 1/6 of 74 must be more than 12. So the difference must
be more than 6. So, if we are being Axelish, we would hold even if
the cube was centred. But this is probably an Isight error. With it being this
close with me owning the cube, it's probably initial double/take.

XGID=--ABCDCA-A-------bbccaca--:1:1:1:00:12:9:3:0:10
X:Daniel O:eXtremeGammon
Score is X:12 O:9. Unlimited Game, Jacoby Beaver
+13-14-15-16-17-18------19-20-21-22-23-24-+
| O O | | O O O O O |
| O O | | O O O |
| | | O O O |
| | | |
| | | |
| |BAR| |
| | | |
| | | X |
| | | X X X | +---+
| | | X X X X | | 2 |
| X X | | X X X X X | +---+
+12-11-10--9--8--7-------6--5--4--3--2--1-+
Pip count X: 74 O: 78 X-O: 12-9
Cube: 2, X own cube
X on roll, cube action

Analyzed in 4-ply
Player Winning Chances: 71.27% (G:0.00% B:0.00%)
Opponent Winning Chances: 28.73% (G:0.00% B:0.00%)

Cubeless Equities: No Double=+0.425, Double=+0.851

Cubeful Equities:
No redouble: +0.684
Redouble/Take: +0.675 (-0.009)
Redouble/Pass: +1.000 (+0.316)

Best Cube action: No redouble / Take
Percentage of wrong pass needed to make the double decision right: 2.7%

eXtreme Gammon Version: 2.10

The final Axelisation is really a thing of beauty. 66 + 11 is 77 for me.
For the opponent, 70 + 2 for the extra crossovers.
So we reach the threshold of 5 exactly. So it's an ultra-marginal redouble. And this doesn't disagree much with XG. XG sometimes holds and
sometimes redoubles, depending on the settings. XG didn't disagree with
my redouble in the immediate analysis. But the coin toss went the other
way in a rollout. The simple 10% and 8/9/12 methods don't work well
here, because of the lack of a crossover penalty.

XGID=--ACCEC----------aadcacaa-:1:1:1:00:12:9:3:0:10
X:Daniel O:eXtremeGammon

Score is X:12 O:9. Unlimited Game, Jacoby Beaver
+13-14-15-16-17-18------19-20-21-22-23-24-+
| O O | | O O O O O O |
| | | O O O |
| | | O O O |
| | | O |
| | | |
| |BAR| |
| | | X |
| | | X |
| | | X X X X | +---+
| | | X X X X | | 2 |
| | | X X X X X | +---+
+12-11-10--9--8--7-------6--5--4--3--2--1-+
Pip count X: 66 O: 70 X-O: 12-9
Cube: 2, X own cube
X on roll, cube action

Analyzed in Rollout
No redouble
Player Winning Chances: 72.44% (G:0.00% B:0.00%)
Opponent Winning Chances: 27.56% (G:0.00% B:0.00%)
Redouble/Take
Player Winning Chances: 72.41% (G:0.00% B:0.00%)
Opponent Winning Chances: 27.59% (G:0.00% B:0.00%)

Cubeless Equities: No Double=+0.449, Double=+0.896

Cubeful Equities:
No redouble: +0.733
Redouble/Take: +0.724 (-0.009)
Redouble/Pass: +1.000 (+0.267)

Best Cube action: No redouble / Take
Percentage of wrong pass needed to make the double decision right: 3.3%

Rollout:
1296 Games rolled with Variance Reduction.
Moves: 3-ply, cube decisions: XG Roller
Confidence No Double: ± 0.003 (+0.730..+0.736)
Confidence Double: ± 0.004 (+0.719..+0.728)

Double Decision confidence: 100.0%
Take Decision confidence: 100.0%

Duration: 5 minutes 50 seconds

eXtreme Gammon Version: 2.10

Paul

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

"[email protected]" <[email protected]> writes:

I know it's just one game but I'm really impressed by it all.

Finally, thanks. You needed quite a bit of hand-holding ...

By the way, one advantage of GNU Backgammon while getting familiar with
the Isight method is that it is implemented. So you can do the math
yourself and then compare with the results given in the "Race Theory"
dialog (at least version 1.06 is needed).

Good luck!

Axel

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On Tuesday, April 19, 2022 at 3:54:31 PM UTC+1, Axel Reichert wrote:

"[email protected]" <[email protected]> writes:

I know it's just one game but I'm really impressed by it all.

Finally, thanks. You needed quite a bit of hand-holding ...

....

I don't agree with this assessment at all.
My key error was that I misread the algo for stacks on the lower points,
so that I (wrongly) thought that sufficiently high stacks all got penalised by the same
amount, rather than the wastage being per checker.

That single misconception carried through to many wrong postings and wrong scores.

Axel

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

"[email protected]" <[email protected]> writes:

On Tuesday, April 19, 2022 at 3:54:31 PM UTC+1, Axel Reichert wrote:

Finally, thanks. You needed quite a bit of hand-holding ...

....

I don't agree with this assessment at all.

I meant this statement rather long-term: I took about 1.5 years from
your first feedback on my article to (perhaps) adopting it in your "over
the board" repertoire. No offense, I am happy about every "customer".

Best regards

Axel

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On Tuesday, April 19, 2022 at 7:51:59 PM UTC+1, Axel Reichert wrote:

"[email protected]" <[email protected]> writes:

On Tuesday, April 19, 2022 at 3:54:31 PM UTC+1, Axel Reichert wrote:

Finally, thanks. You needed quite a bit of hand-holding ...

....

I don't agree with this assessment at all.

I meant this statement rather long-term: I took about 1.5 years from
your first feedback on my article to (perhaps) adopting it in your "over
the board" repertoire. No offense, I am happy about every "customer".

Best regards

Axel

I would like to adopt it in my OTB repertoire when possible.
I will always do this against XG.
In live play, I'm not sure whether I can do this at the required speed.
I excel in pure mental arithmetic in the sense of square root of this, multiply this by that etc.
But combining a set of rules is something else. I can do it but I'm not sure how fast.
I will do it in live play if I'm fast enough.

Paul

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)