BGonline.org Forums

Nine 5k results (report)

Posted By: Daniel Murphy
Date: Monday, 17 January 2011, at 9:43 p.m.

In Response To: Nine 5k results (report) (Nack Ballard)

I guess my reaction is similar to Xavier's. It's very good to remind people of the variance in rollouts, but "that's why people need to check CI," and people could check the confidence intervals if they were always included in reports of individual and combined rollouts.

If these 5k increments had been done individually and the result in one had been difference = 0.000 (as it was in one of Stick's 5k segments), the 95% confidence intervals for each play would have been about +/- 0.015. So in response to "Imagine if [a] 5k rollout yielded [that] result ... and you figured that was good enough ... you'd have quite a different perception of the closeness of the plays," I'd see right away that the CI's would be telling us that it isn't "good enough." It'd be wrong to conclude that the plays are equal; it'd be right to see that it wouldn't be surprising if either play was as much as +/- 0.030 different from the other.

By the logic that I presented here, the fairest estimate is that the difference is .01875 ... The order in which the 5k rollout segments occurred doesn't matter.

Yes, and so this (below) is somewhat interesting -- just confirming that the order doesn't matter, following the procedure you presented in post 85716:

When you took Stick's 5k increments in the original order and figured out what the differences must have been of each segment, you got:

Total	.029	.025	.023	.023	.019	.021	.018	.017	.019
Latest 5k	.029	.021	.019	.023	.003	.031	.000	.010	.033

(You changed the rounded 0.19 to 0.1875.)

If we reorder the segments by difference high/low/high/low, etc. and figure out what the difference would have been after the rollout had been extended by each increment, we get:

Latest 5k	.033	.000	.031	.003	.029	.010	.023	.019	.021
Total	.033	.017	.021	.017	.019	.0176	.0184	.0185	.01877

And if we reorder the nine segments by difference from high to low, we get:

Latest 5k	.033	.031	.029	.023	.021	.019	.010	.003	.000
Total	.033	.032	.031	.029	.027	.026	.023	.021	.01877

Messages In This Thread

• Opening 52 at 4 away 4 away
  Stick -- Saturday, 15 January 2011, at 9:00 p.m.
  • Opening 52 at 4 away 4 away
    neilkaz -- Saturday, 15 January 2011, at 9:37 p.m.
  • Stat sig
    Timothy Chow -- Sunday, 16 January 2011, at 2:35 p.m.
  • Opening 52 at 4 away 4 away - extended
    Stick -- Monday, 17 January 2011, at 9:55 a.m.
    • Nine 5k results (report)
      Nack Ballard -- Monday, 17 January 2011, at 3:20 p.m.
      • Nine 5k results (report) - that why people need to check SI
        eXtreme Gammon -- Monday, 17 January 2011, at 6:00 p.m.
        • Nine 5k results (report) - that why people need to check SI
          Stick -- Monday, 17 January 2011, at 6:15 p.m.
          • Nine 5k results (report) - that why people need to check SI
            eXtreme Gammon -- Monday, 17 January 2011, at 6:56 p.m.
            • Nine 5k results (report) - that why people need to check SI
              Nack Ballard -- Monday, 17 January 2011, at 10:59 p.m.
            • Nine 5k results (report) - that why people need to check SI
              Neil Robins -- Tuesday, 18 January 2011, at 9:34 p.m.
              • Nine 5k results (report) - that why people need to check SI
                eXtreme Gammon -- Wednesday, 19 January 2011, at 4:47 a.m.
                • Nine 5k results (report) - that why people need to check SI
                  Timothy Chow -- Wednesday, 19 January 2011, at 2:33 p.m.
              • Nine 5k results (report) - that why people need to check SI
                Daniel Murphy -- Wednesday, 19 January 2011, at 9:10 p.m.
        • You mean CI *NM*
          neilkaz -- Monday, 17 January 2011, at 6:41 p.m.
          • You mean CI - I did indeed *NM*
            eXtreme Gammon -- Monday, 17 January 2011, at 6:56 p.m.
        • Nine 5k results (report) - that why people need to check SI
          Timothy Chow -- Tuesday, 18 January 2011, at 5:30 p.m.
      • Nine 5k results (report)
        Daniel Murphy -- Monday, 17 January 2011, at 9:43 p.m.