| |
|---|---|
| 52S-63R-42 | |
The Snowie truncated rollout I have for 52S-63R-42 (diagrammed above) is [@ P21], while the full rollout (though no live cube) gives [@ P6] 5k. It's too little data to make strong conclusions, but if we average the two low-quality results we get [@ P13], meaning that @ is best and P is -.013, and we might say that there is an anti-P trunc bias of .008.
Here are Snowie trunc results for all 52S-xxx-42 positions I have where @ (anchoring on the 20pt) and P (making the 4pt) are the two top plays:
52S-21S-42 [@ P66] tr
52S-41T-42 [@ P25] tr
52S-41U-42 [@ P9] tr
52S-42P-42 [@ P87] tr
52S-43S-42 [@ P66] tr
52S-51R-42 [@ P34] tr
52S-51S-42 [@ P35] tr
52S-52D-42 [@ P73] tr
52S-52S-42 [@ P35] tr
52S-54S-42 [@ P56] tr
52S-61P-42 [@ P64] tr
52S-62S-42 [@ P49] tr (hitting is a whopper)
52S-63R-42 [@ P21] tr, and [@ P6] 5k no live cube
52S-63S-42 [@ P31] tr
52S-64P-42 [P @16] tr
52S-65R-42 [@ P24] tr
If the trunc bias I inferred earlier is is our best guess, then decrease the above error sizes by 8 (thousandths) except increase that of 52S-64P-42 (from 16 to 24).
Paul and I had thought (after proper rollout verification) to make 52S-42P-42 and 52S-64P-42 a contrasting spread (as shown below) in one of our books. But really what it comes down to is that @ (anchoring on Opp's 5pt) beats P (making own 4pt) for all 52S-xxx-42 positions except when xxx is 64P (or 41U is likely a tie).
|
| ||||||||||||||||
I tell my students that a good rule of thumb is:
Either 5pt beats either 4pt.
Nack