BGonline.org Forums

Stopping a rollout early based on JSD

Posted By: Ian Shaw
Date: Wednesday, 11 May 2011, at 1:27 p.m.

In Response To: Stopping a rollout early based on JSD (Tom Keith)

Thanks for the warning about rollouts of more than 2 plays.

Can I check my understanding with you, please, using data from Simulated Rollout: Maximum trials: 1296 Minimum trials: 324 as an example. I understand your results to mean:

If the difference between the plays would be 1.65 jsd on completion of 1296 trails, the chance that the setting [minimum 324m cutoff 2.5 jsd] gives the wrong result is 0.05014 (or 5.015%). "Wrong" means ranking the plays in the reverse order.

When setting up a cutoff rollout, there still seems to be the problem of knowing what the jsd would be on completion.

Messages In This Thread

• Stopping a rollout early based on JSD
  Tom Keith -- Sunday, 8 May 2011, at 2:57 p.m.
  • Stopping a rollout early based on JSD
    Ian Shaw -- Tuesday, 10 May 2011, at 4:11 p.m.
    • Stopping a rollout early based on JSD
      Tom Keith -- Wednesday, 11 May 2011, at 11:43 a.m.
      • Stopping a rollout early based on JSD
        Ian Shaw -- Wednesday, 11 May 2011, at 1:27 p.m.
        • Stopping a rollout early based on JSD
          Tom Keith -- Thursday, 12 May 2011, at 12:40 p.m.
  • Stopping a rollout early based on JSD
    Chuck Bower -- Thursday, 12 May 2011, at 4:25 p.m.
    • Stopping a rollout early based on JSD
      Tom Keith -- Thursday, 12 May 2011, at 9:27 p.m.