BGonline.org Forums

Half as wild

Posted By: Daniel Murphy
Date: Thursday, 22 April 2010, at 2:37 a.m.

In Response To: Wild rollout results (bug?) (Nack Ballard)

Neil's long XG 3-ply rollout result for

24/20* 20/15 eq: +0.145

falls within the 95% confidence intervals reported in Chase's shorter XG 3-ply rollouts. They were

Conf: ± 0.009 (+0.127...+0.145)
Conf: ± 0.009 (+0.128...+0.146)

But Neil's long XG 3-ply rollout result for

24/20* 13/8 eq: +0.158

is an outlier to Chase's results:

Conf: ± 0.010 (+0.166...+0.186)
Conf: ± 0.010 (+0.165...+0.185)

Conversely, of course, Chase's results for that play are the outliers to Neil's 3-ply and 4-ply rollouts -- see below.

---

As a reader attempting to judge the validity of the various rollouts and their combination, I'm thinking it would be helpful if the summary data included both absolute equities and 95% confidence intervals.

For example, in the rollouts that Nack cites (other than Snowie's, since I don't know where that is) the absolute equities for the 24/20* 13/8 response were:

XG 3-ply	eq: +0.176 ± 0.010
XG 3-ply	eq: +0.175 ± 0.010
XG 3-ply	eq: +0.158 ± 0.003
XG 4-ply	eq: +0.152 ± 0.005
Gnubg 2-ply	eq: +0.137 ±0.005

Above, the two high results are Chase's, already discussed. But what about the Gnubg rollout? It's as unusually low as Chase's are high, compared with Neil's. Is it right to simply combine and average the relative equities of various rollouts by different bots/plies, without regard for absolute equities? One (or more) of these rollouts is "wrong." If, for instance, we want to combine the relative equities of the XG 4-ply and Gnubg 2-ply rollouts, even though their absolute results are out of each other's 95% confidence intervals, is there a rationale for doing that, other than the old standby, that "bots make mistakes playing both sides, and the errors even out (we think) so the relative equities are legitimate, despite difference in absolute equities"?

And another question: a rollout of two plays reports standard deviations or 95% confidence intervals for the absolute equity of each play. But what's the confidence interval of the relative equity? I'm thinking that we should have less confidence in that than the summary data might be seen to be suggesting.

Messages In This Thread

• 4 ply RO of 21$-54
  neilkaz -- Monday, 19 April 2010, at 5:18 p.m.
  • 4 ply RO of 21$-54
    Nack Ballard -- Monday, 19 April 2010, at 6:18 p.m.
  • 4 ply RO of 21$-54
    Chase -- Monday, 19 April 2010, at 6:21 p.m.
    • 4 ply RO of 21$-54
      neilkaz -- Monday, 19 April 2010, at 6:48 p.m.
  • 4 ply RO of 21$-54
    Nack Ballard -- Monday, 19 April 2010, at 6:25 p.m.
    • 4 ply RO of 21$-54
      eXtreme Gammon -- Monday, 19 April 2010, at 6:32 p.m.
      • 4 ply RO of 21$-54
        Nack Ballard -- Tuesday, 20 April 2010, at 7:19 p.m.
  • ... and XG 3-ply RO of 21$-54
    Chase -- Monday, 19 April 2010, at 7:39 p.m.
    • ... and XG 3-ply RO of 21$-54
      Neil Robins -- Monday, 19 April 2010, at 8:38 p.m.
    • Currently [S R5]16k *NM*
      neilkaz -- Tuesday, 20 April 2010, at 7:41 p.m.
    • ... and XG 3-ply RO of 21$-54
      Nack Ballard -- Tuesday, 20 April 2010, at 7:54 p.m.
      • ... and XG 3-ply RO of 21$-54
        Chase -- Tuesday, 20 April 2010, at 8:39 p.m.
        • ... and XG 3-ply RO of 21$-54
          Chase -- Tuesday, 20 April 2010, at 8:46 p.m.
          • 41$-63, 21$-63, 51$-63 and 62$-63
            Nack Ballard -- Wednesday, 21 April 2010, at 4:35 a.m.
            • 41$-63, 21$-63, 51$-63 and 62$-63
              Chase -- Wednesday, 21 April 2010, at 10:43 a.m.
              • 41$-63, 21$-63, 51$-63 and 62$-63
                Matt Cohn-Geier -- Wednesday, 21 April 2010, at 10:51 a.m.
                • 41$-63, 21$-63, 51$-63 and 62$-63
                  Chase -- Wednesday, 21 April 2010, at 11:03 a.m.
                • 41$-63, 21$-63, 51$-63 and 62$-63
                  Nack Ballard -- Wednesday, 21 April 2010, at 2:11 p.m.
                  • 41$-63, 21$-63, 51$-63 and 62$-63
                    Chase -- Wednesday, 21 April 2010, at 4:53 p.m.
                    • 41$-63, 21$-63, 51$-63 and 62$-63
                      Nack Ballard -- Wednesday, 21 April 2010, at 5:44 p.m.
              • Nacbrac information
                Nack Ballard -- Wednesday, 21 April 2010, at 2:43 p.m.
                • Nacbrac information
                  Chase -- Wednesday, 21 April 2010, at 5:04 p.m.
                  • Nacbrac information
                    Nack Ballard -- Wednesday, 21 April 2010, at 5:25 p.m.
        • ... and XG 3-ply RO of 21$-54
          Nack Ballard -- Wednesday, 21 April 2010, at 3:39 a.m.
      • ... and XG 3-ply RO of 21$-54 w/Jacoby
        neilkaz -- Wednesday, 21 April 2010, at 3:07 a.m.
        • Will extend this 3 ply to 100k
          neilkaz -- Wednesday, 21 April 2010, at 12:38 p.m.
    • XG 3-ply RO of 21$-54 - w/ Jacoby
      Chase -- Wednesday, 21 April 2010, at 11:16 a.m.
  • 4 ply RO of 21$-54
    Nack Ballard -- Tuesday, 20 April 2010, at 8:19 p.m.
    • 4 ply RO of 21$-54
      neilkaz -- Tuesday, 20 April 2010, at 10:11 p.m.
  • 3 ply RO of 21$-54..over 100k
    neilkaz -- Wednesday, 21 April 2010, at 10:11 p.m.
    • Wild rollout results (bug?)
      Nack Ballard -- Thursday, 22 April 2010, at 1:05 a.m.
      • Wild rollout results (bug?)
        neilkaz -- Thursday, 22 April 2010, at 1:53 a.m.
        • Re-run of RO w/ random seed and Jacoby in progress
          Chase -- Thursday, 22 April 2010, at 2:08 a.m.
          • Re-run of RO w/ random seed and Jacoby in progress
            Nack Ballard -- Thursday, 22 April 2010, at 2:27 a.m.
            • Re-run of RO w/ random seed and Jacoby in progress
              Chase -- Thursday, 22 April 2010, at 2:42 p.m.
          • Completed RO w/ Jacoby and random seed
            Chase -- Thursday, 22 April 2010, at 3:05 p.m.
        • My 10k with seed 2
          neilkaz -- Thursday, 22 April 2010, at 3:09 a.m.
      • Wild rollout results (bug?)
        Chase -- Thursday, 22 April 2010, at 2:02 a.m.
      • Half as wild
        Daniel Murphy -- Thursday, 22 April 2010, at 2:37 a.m.
        • Half as wild
          Neil Robins -- Thursday, 22 April 2010, at 11:34 a.m.
        • Half as wild
          Timothy Chow -- Thursday, 22 April 2010, at 5:01 p.m.
      • 21$-54 rollout data (equities, links)
        Nack Ballard -- Thursday, 22 April 2010, at 7:50 p.m.
        • 21$-54 rollout data (equities, links)
          Daniel Murphy -- Thursday, 22 April 2010, at 8:34 p.m.
          • 21$-54 rollout data (equities, links)
            Nack Ballard -- Friday, 23 April 2010, at 9:03 a.m.
  • 4 ply RO of 21$-54 19k and finished
    neilkaz -- Thursday, 22 April 2010, at 12:08 a.m.
  • Snowie RO of 21$-54
    Nack Ballard -- Thursday, 22 April 2010, at 7:16 p.m.
    • Snowie RO of 21$-54
      Daniel Murphy -- Thursday, 22 April 2010, at 7:37 p.m.