• Sweet but psychic

    From pepstein5@gmail.com@21:1/5 to All on Sat Apr 9 12:30:11 2022
    When I've finished a game with XG, I have a moment of considerable
    tension as I wait for its PR verdict.
    What I've found is that my guesses as to how I've played appear to
    be completely uncorrelated with the PR score. I might feel confident
    that I've made all the right plays, only to be given a PR of 24.
    Or maybe all my decisions seem dubious and I turn out to have
    played near-perfectly.
    What do others experience? Are others able to guess correctly
    how well they played?

    Thank You,

    Paul

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Nasti Chestikov@21:1/5 to peps...@gmail.com on Sun Apr 10 09:47:16 2022
    On Saturday, 9 April 2022 at 20:30:12 UTC+1, peps...@gmail.com wrote:
    When I've finished a game with XG, I have a moment of considerable
    tension as I wait for its PR verdict.
    What I've found is that my guesses as to how I've played appear to
    be completely uncorrelated with the PR score. I might feel confident
    that I've made all the right plays, only to be given a PR of 24.
    Or maybe all my decisions seem dubious and I turn out to have
    played near-perfectly.
    What do others experience? Are others able to guess correctly
    how well they played?

    Thank You,

    Paul

    Your failing is believing that what the bot tells you is gospel.

    Your first roll is 5-2, you play 13-8, 13-11. Well done.

    Your next roll is 6-1. If you play (the correct roll of) 13-7, 8-7 you'll get slaughtered because the "correct" roll is apparently 11-5, 6-5.

    And then AI Factory Backgammon *will roll* 6-6 or 6-5 and blow you out of the water.

    This "golden point" nonsense has a lot to answer for.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Timothy Chow@21:1/5 to Nasti Chestikov on Wed Apr 13 08:29:34 2022
    On 4/10/2022 12:47 PM, Nasti Chestikov wrote:
    On Saturday, 9 April 2022 at 20:30:12 UTC+1, peps...@gmail.com wrote:
    When I've finished a game with XG, I have a moment of considerable
    tension as I wait for its PR verdict.
    What I've found is that my guesses as to how I've played appear to
    be completely uncorrelated with the PR score. I might feel confident
    that I've made all the right plays, only to be given a PR of 24.
    Or maybe all my decisions seem dubious and I turn out to have
    played near-perfectly.
    What do others experience? Are others able to guess correctly
    how well they played?

    Thank You,

    Paul

    Your failing is believing that what the bot tells you is gospel.

    This is irrelevant. Even if the bot has completely mistaken ideas of
    good backgammon play, it may still be possible to predict with high
    accuracy what PR it will give you.

    ---
    Tim Chow

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From pepstein5@gmail.com@21:1/5 to Tim Chow on Wed Apr 13 06:18:45 2022
    On Wednesday, April 13, 2022 at 1:29:37 PM UTC+1, Tim Chow wrote:
    On 4/10/2022 12:47 PM, Nasti Chestikov wrote:
    On Saturday, 9 April 2022 at 20:30:12 UTC+1, peps...@gmail.com wrote:
    When I've finished a game with XG, I have a moment of considerable
    tension as I wait for its PR verdict.
    What I've found is that my guesses as to how I've played appear to
    be completely uncorrelated with the PR score. I might feel confident
    that I've made all the right plays, only to be given a PR of 24.
    Or maybe all my decisions seem dubious and I turn out to have
    played near-perfectly.
    What do others experience? Are others able to guess correctly
    how well they played?

    Thank You,

    Paul

    Your failing is believing that what the bot tells you is gospel.
    This is irrelevant. Even if the bot has completely mistaken ideas of
    good backgammon play, it may still be possible to predict with high
    accuracy what PR it will give you.

    Actually, I do find the point relevant (which, of course, doesn't mean at all that I agree with it).
    If I didn't rate the bot's judgment highly, I wouldn't care much about the PR, and wouldn't experience "considerable tension."

    I agree that the point doesn't contribute to answering my question, but the point is very relevant to the importance of my question.
    For example, take the following exchange:
    "Do you think it's likely to rain tomorrow?"
    "It doesn't matter whether it rains or not since we'll be spending the day indoors."

    It seems to have the logical structure of this thread.
    Would you say that the response is irrelevant because it says nothing to address the likelihood of rain?
    I wouldn't.

    Paul

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Timothy Chow@21:1/5 to peps...@gmail.com on Wed Apr 13 11:40:47 2022
    On 4/13/2022 9:18 AM, peps...@gmail.com wrote:
    For example, take the following exchange:
    "Do you think it's likely to rain tomorrow?"
    "It doesn't matter whether it rains or not since we'll be spending the day indoors."

    It seems to have the logical structure of this thread.
    Would you say that the response is irrelevant because it says nothing to address the likelihood of rain?
    I wouldn't.

    It seems that a better parallel would be:

    "Are you good at predicting what weather.com will say about tomorrow's weather?"
    "Your mistake is taking weather.com as gospel."
    "That's irrelevant; one can be good at predicting what weather.com
    will say even if weather.com is terrible at predicting the weather."

    ---
    Tim Chow

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From pepstein5@gmail.com@21:1/5 to Tim Chow on Wed Apr 13 12:03:50 2022
    On Wednesday, April 13, 2022 at 4:40:51 PM UTC+1, Tim Chow wrote:
    On 4/13/2022 9:18 AM, peps...@gmail.com wrote:
    For example, take the following exchange:
    "Do you think it's likely to rain tomorrow?"
    "It doesn't matter whether it rains or not since we'll be spending the day indoors."

    It seems to have the logical structure of this thread.
    Would you say that the response is irrelevant because it says nothing to address the likelihood of rain?
    I wouldn't.
    It seems that a better parallel would be:

    "Are you good at predicting what weather.com will say about tomorrow's weather?"
    "Your mistake is taking weather.com as gospel."
    "That's irrelevant; one can be good at predicting what weather.com
    will say even if weather.com is terrible at predicting the weather."

    It's a fair parallel.
    Again, the person saying "Your mistake..." doesn't sound completely irrelevant to me.
    If weather.com is terrible, then the activity of predicting what it will say lacks motive.

    Again, the 2nd voice is relevant to the importance of the question, even though irrelevant to the
    answer to the question.

    In ordinary informal dialogue, it's fine (in my opinion) to address the importance/relevance of questions.

    Paul

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Stick Rice@21:1/5 to peps...@gmail.com on Wed Apr 20 13:11:42 2022
    On Saturday, April 9, 2022 at 3:30:12 PM UTC-4, peps...@gmail.com wrote:
    When I've finished a game with XG, I have a moment of considerable
    tension as I wait for its PR verdict.
    What I've found is that my guesses as to how I've played appear to
    be completely uncorrelated with the PR score. I might feel confident
    that I've made all the right plays, only to be given a PR of 24.
    Or maybe all my decisions seem dubious and I turn out to have
    played near-perfectly.
    What do others experience? Are others able to guess correctly
    how well they played?

    Thank You,

    Paul

    I know how well I played. There is margin for error of course as maybe on some of those decisions I was unsure of I got lucky and guessed correctly or maybe on some of those decisions I was unsure of I guessed and even though it was wrong it didn't
    really matter but overall I know if I played well, average, or blah before the bot tells me anything.

    Stick

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