Last week, while looking up opening roll related
articles, I had come accrosss a very old thread
titled: "Who goes first in Matches". See:

https://groups.google.com/g/rec.games.backgammon/c/JEMDGIhr4Sw/m/u9iIapMoTaYJ

In it, many ways of determining who goes first in
various different flavors of classic backgammon
and gamblegammon, starting with alternating in
match play as suggested by the original poster.

It had truely saddened me at least for the sake of
backgammon, (since I really don't care that much
about gamblegammon), to see "Chuck Bower" say:

"Probably since match backgammon evolved out
"of money backgammon, the individual games in
"the match were just kept close to money play

Who knows how many westerners who first learn
cubeful gamblegammon must think that cubeless
backgammon is a variant of a game invented by
the mentally ill American/arithmetician gamblers.

My point here is that winning the opening roll and
rolling the opening dice are two separate events,
even if they seem to be combined and happening
simultaneously.

Although while playing backgammon matches, the
winner of the opening roll may be the roller of the
high or low die, winner or loser of the previous game,
alternating, etc. gamblegammon players around the
world must be going by the American/western rule.

Since no buggy, garbage generating gamblegammon
bot offers any of those alternatives, (even while they
offer cubeless match play similar to backgammon),
let's just focus on the rule of the roller of the high die
going first by having to play the "opening dice" rolled.

What about if one (or both) of the players say that he
wants to double instantly if he wins the opening roll..?

Since he can't know the resulting dice before they roll
a die each, this should be technically acceptable. Any
logical arguments to the contarary..?

I intentionally keep my brain uninfected by the match
equity table bullshit and thus don't know if there would
be scores where doing this would make sense. Maybe
in post-crawford, etc. situations of gamblegammon..?

In match play, no mentally ill gamblers worshippers of
bots would do this because bots say that winning the
opening roll gives you only 52.5%(?) winning chances.

But Axel's Murat mutant bot would, as it would double
in any position with > 50% winning chances. :)

So, what if Axel runs an experiment on this and finds
out again that after 4 billion games, (okay I'll settle for
10,000 games but not extrapolated), the Murat mutant
wins 52.5% and shoves the so-called cube skill bullshit
up your mentally ill arithmetician asses...??

MK

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

On Thursday, 19 January 2023 at 08:58:03 UTC, MK wrote:

What about if one (or both) of the players say that he
wants to double instantly if he wins the opening roll..?

Since he can't know the resulting dice before they roll
a die each, this should be technically acceptable. Any
logical arguments to the contarary..?

MK

I'd be interested in seeing how GnuDung or XG would evaluate such a double.

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

On 1/20/2023 12:55 PM, Nasti Chestikov wrote:

On Thursday, 19 January 2023 at 08:58:03 UTC, MK wrote:

What about if one (or both) of the players say that he
wants to double instantly if he wins the opening roll..?

Since he can't know the resulting dice before they roll
a die each, this should be technically acceptable. Any
logical arguments to the contarary..?

MK

I'd be interested in seeing how GnuDung or XG would evaluate such a double.

I don't know that we'd need a bot to tell us that (except at certain
match scores) it would be a blunder to do that. The best opening roll
(31) only gives that player a 55% chance to win.

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

On January 20, 2023 at 10:55:44 AM UTC-7, Nasti Chestikov wrote:

On Thursday, 19 January 2023 at 08:58:03 UTC, MK wrote:

What about if one (or both) of the players say that
he wants to double instantly if he wins the opening
roll..? Since he can't know the resulting dice before
they roll a die each, this should be technically
acceptable. Any logical arguments to the contarary..?

I'd be interested in seeing how GnuDung or XG
would evaluate such a double.

Here are the opening positions with a game/match
technically started but without anything happened
yet. So, you can paste them and double right away
to experiment with whatever you want.

Gnubg ID: 4HPwAyDgc/ABMA:AgEAAAAAAAAA

XGID=-b----E-C---eE---c-e----B-:0:0:1:00:0:0:1:0:10

Gnubg 3ply says:
=============================================
Win W(g) W(bg) L(g) L(bg) Equity Cubeful
static: 52.1 15.4 0.8 13.0 0.8 +0.067 +0.084

Cube analysis
0-ply cubeless equity +0.067
52.1 15.4 0.8 - 47.9 13.0 0.8

Cubeful equities:
1. No double +0.084
2. Double, pass +1.000 (+0.916)
3. Double, take -0.191 (-0.276)

Proper cube action: No double, beaver (23.1%) !! =============================================

XG Roller++ says:
=============================================
Player Winning Chances: 52.84% (G:15.11% B:0.73%)
Opponent Winning Chances: 47.16% (G:12.61% B:0.53%)

Cubeless Equities: No Double=+0.084, Double=+0.168

Cubeful Equities:
No double: +0.120
Double/Take: -0.180 (-0.300)
Double/Pass: +1.000 (+0.880)

Best Cube action: No double / Take =============================================

I spent some time looking at temperature maps, etc.
but couldn't really make any sense out of it for sure,
since the it's hard to tell if the buggy bots handle the
opening position properly in equity calculations, etc.

Perhaps the only thing that may be useful is that the
bots don't redouble immediately and Gnubgs beaver
becomes simple take and takes by both bots become
drops at some match scores.

Between two humans, if a players declares that he will
double regardless, the other player has to respond also
before the roll whether he will take/drop/beaver, etc.

It would be interesting to see how a mutant bot would
do with this, (vs. waiting to have access to the cube at

50% after the opening roll). Actually, anyone can do a

short experiment by pasting the opening positions and
selecting "play from position" as human vs human first
and then changing one player to bot and forcing games
to be played out to the end. It sure would be tedious (vs.
a nice person like Axel making his mutant bot do it... ;)

MK

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