BGonline.org Forums

Morning update

Posted By: Nack Ballard
Date: Monday, 3 May 2010, at 8:26 p.m.

In Response To: Morning update (neilkaz)

Typical varience we can mathematically expect [P S2]11,000 games..

Right. Had the first and second halves of your rollout been separate, we would have seen

[S P4] 5740 trials
[P S9] 5260 trials

... averaging to the current [P S2] 11,000

... which reinforces the understanding that a mere 5k trials can be misleading for close decisions. By extension, four times as many trials only reduces the swing by half, and nine times as many trials by two-thirds. Thus, two independent 46k-trial rollout margins will differ from each other by .005 almost as often as two 5k-trial margins will differ by .013 (as we've seen here).

Nack

Messages In This Thread

• 21S-53 XG Rollout?
  Chase -- Sunday, 2 May 2010, at 8:58 a.m.
  • OK will do XG 4 ply now
    neilkaz -- Sunday, 2 May 2010, at 2:33 p.m.
    • OK will do XG 4 ply now
      Stick -- Sunday, 2 May 2010, at 8:31 p.m.
  • 21S-53 XG Rollout?
    Nack Ballard -- Sunday, 2 May 2010, at 6:50 p.m.
    • 21S-53 XG Rollout?
      Stick -- Sunday, 2 May 2010, at 8:31 p.m.
      • 21S-53 XG Rollout?
        Nack Ballard -- Monday, 3 May 2010, at 12:01 a.m.
        • 4 ply so far....
          neilkaz -- Monday, 3 May 2010, at 3:16 a.m.
          • 4 ply so far....
            Nack Ballard -- Monday, 3 May 2010, at 4:24 a.m.
            • Morning update
              neilkaz -- Monday, 3 May 2010, at 11:59 a.m.
              • Morning update
                Nack Ballard -- Monday, 3 May 2010, at 8:26 p.m.
                • Evening update
                  neilkaz -- Tuesday, 4 May 2010, at 4:33 a.m.