Having talked about rollouts, PR's, etc. recently,
I played a few 17-point matches against Ex-Gee
(at XGR++ level) and in this position It strangely
offered to resign a single game:

XGID=-C-CA-------a--------bedc-:2:-1:-1:00:12:0:0:17:10

GNGbg ID:5wIAAPe+AQQAAA:AgEgAgAAYAAE

Score is X:0 O:12 17 pt.(s) match.
+13-14-15-16-17-18------19-20-21-22-23-24-+
| X | | O O O |
| | | O O |
| | | O O |
| | | |
| | | |
| |BAR| |
| | | X |
| | | X X |
| | | X X X | +---+
| | | X X X X | | 4 |
| | | X X X X | +---+
+12-11-10--9--8--7-------6--5--4--3--2--1-+
Pip count X: 47 O: 16 X-O: 0-12/17
Cube: 4, X own cube
X on roll, cube action

I (i.e. O) still had a slight chance of winning a
gammon and the match (bots say about 1%).
But Ex-Gee offered to resign a single game.

I declined, won the game and then the match. I
thought could go back to that position and see
if it would do it again but resign/decline actions
weren't in the game record and I couldn't figure
out how to make the bot duplicate the decison.

Is it because the analysis settings said: "Do not
calculate resign errors"?

I don't remeber if Noo-BG includes declined resign
actions in its game records. Does it?

Occasionally I decline even correct resign offers
by the bots just to see the next roll, etc. and never
worried about going back to locate/analyse them.

Most/all of you here may not see nor even ask for
any reasons to include those in the game records,
but I do have at least a couple of them. However,
I'm not going to disclose and talk about them just
yet, in order to give you all a chance to offer your
views/arguments on this subject...

MK

On 1/11/2024 2:58 AM, MK wrote:
> Having talked about rollouts, PR's, etc. recently,
> I played a few 17-point matches against Ex-Gee
> (at XGR++ level) and in this position It strangely
> offered to resign a single game:
>
> XGID=-C-CA-------a--------bedc-:2:-1:-1:00:12:0:0:17:10
>
> GNGbg ID:5wIAAPe+AQQAAA:AgEgAgAAYAAE
>
> Score is X:0 O:12 17 pt.(s) match.
> +13-14-15-16-17-18------19-20-21-22-23-24-+
> | X | | O O O |
> | | | O O |
> | | | O O |
> | | | |
> | | | |
> | |BAR| |
> | | | X |
> | | | X X |
> | | | X X X | +---+
> | | | X X X X | | 4 |
> | | | X X X X | +---+
> +12-11-10--9--8--7-------6--5--4--3--2--1-+
> Pip count X: 47 O: 16 X-O: 0-12/17
> Cube: 4, X own cube
> X on roll, cube action
>
> I (i.e. O) still had a slight chance of winning a
> gammon and the match (bots say about 1%).

What bot says 1%? XG says it's more like 0.03%.

Even without a bot you can see that it has to be a lot
less than 1%. O has to roll doublets twice (at least one
of which has to be 33 or higher), and X has to roll at
most 7 pips in two rolls.

---
Tim Chow

On 1/11/2024 3:57 PM, Timothy Chow wrote:

> On 1/11/2024 2:58 AM, MK wrote:

>> Having talked about rollouts, PR's, etc. recently,
>> I played a few 17-point matches against Ex-Gee
>> (at XGR++ level) and in this position It strangely
>> offered to resign a single game:

>> XGID=-C-CA-------a--------bedc-:2:-1:-1:00:12:0:0:17:10

>> GNGbg ID:5wIAAPe+AQQAAA:AgEgAgAAYAAE

>> I (i.e. O) still had a slight chance of winning
>> a gammon and the match (bots say about 1%).

> What bot says 1%?

Ex-Gee and Noo-BG.

> XG says it's more like 0.03%.

Only at some plies.

> Even without a bot you can see that it has to
> be a lot less than 1%. O has to roll doublets
> twice (at least one of which has to be 33 or
> higher), and X has to roll at most 7 pips in
> two rolls.

I couldn't even begin to guess what percent. I
just knew that I had a chance and I asked bots
after the fact.

As long as I had >0% gammon, you should have
said that the piece of shit extra-terrestrial
bot (that, unlike the freeware garbage Noo-BG,
doesn't even know how to resign properly when
it should), shouldn't have offered to resign
a single game.

Your comment indicates there is a difference
between 1.00% and 0.03%, which is a terrible
argument to make. What's going on with you?
You seem to be falling apart lately... :(

But don't you and your venerated garbage bots
expect mercy from me. I will make guacamole
out of you all and your bots. ;)

Ex-Gee 2-ply says 0.055%, 3-ply says 0.063%,
4-ply says 0.090%, XG-Roller+ and XG-Roller++
both say 0.029%, rollout says 0.033%

Rounded up to 1 decimal, half say %1

Noo-BG 0, 2, 3 and 4-ply say 0.0% but 1-ply
says 0.1%

Noo-BG's 0.0%'s are rounded down to 1 decimal.

For Noo-BG, price is right. But 1-1.5 million
dollars for Ex-Gee's IPR..? Hah haaa! My ass!
I don't think he will be able to give it away
for free.

And you mentally ill gamblegammon players bet
your money on these garbage bots? Sigh... :(

MK

On 1/12/2024 2:47 AM, MK wrote:

>> What bot says 1%?

> Ex-Gee and Noo-BG.

> I couldn't even begin to guess what percent. I
> just knew that I had a chance and I asked bots
> after the fact.

> As long as I had >0% gammon.....

> Your comment indicates there is a difference
> between 1.00% and 0.03%, which is a terrible
> argument to make.

> Ex-Gee 2-ply says 0.055%, 3-ply says 0.063%,
> 4-ply says 0.090%, XG-Roller+ and XG-Roller++
> both say 0.029%, rollout says 0.033%

> Rounded up to 1 decimal, half say %1

> Noo-BG 0, 2, 3 and 4-ply say 0.0% but 1-ply
> says 0.1%

> Noo-BG's 0.0%'s are rounded down to 1 decimal.

Sometimes I reread recent posts, including my
own, to refresh my memory on arguments made.

What a scatterbrain my above post was. Wow!
My mind and fingers were going in the same
direction but not in step... :(

What I was trying to say is that I never try
to estimate my winning chances in numbers and
I probably couldn't come within 10% if I did;
surely not in three decimals. My smallest unit
greater 0% would be 1%.

Looking at the bots' various estimates while
typing, I was seeing 0.1% as 1%. My bad. :(

But my argument that XG shouldn't have tried
to resign a single game when I could gammon
>0% is valid regardless.

And if the the bots claim to be accurate to
three decimals, then their numbers should be
consistent at least in simple, last few rolls
positions and more importantly, they should
be able to resign correctly.

MK