I must not understand what the numbers are measuring. How are there non whole numbers for the cube on 4 in the first table?
It does seem to show that the cube is being taken after a dance and that looks to be an error. The rollout plays are according to XGroller which gives .9819 take equity. However XG++ gives 1.0186 and a rollout indicates that taking is a significant error at 1.04. That happens 1/4 of the time so if the rollout is good the equity for 18/14* is being overstated by about .01. An overall lesson from this is that if the next play is a cube decision or a key play, use the settings option allowing the next 2 plays to be at a higher play setting.
It also shows a large number of cubes occurring later in the game. When most of losses will be gammons which will lose the match anyways, do not need much possible market loss to make cubing correct.
For those who are better at me with tech stuff, why are the images showing up on my computer but not my iphone 15?
Question: Why did I use "much possible market loss" instead of "many market losers"?
                        |
| ||||
| XGID=a-ABBABAA--b-BA--bAcbcA-b-:1:1:1:00:2:6:0:9:10 | |||||
| Blue on roll, cube action? | |||||
| Analyzed in Rollout | No redouble | Redouble/Take |
| Player Winning Chances: | 51.711% (G:16.378% B:0.655%) | 52.443% (G:17.885% B:1.004%) |
| Opponent Winning Chances: | 48.289% (G:28.055% B:2.093%) | 47.557% (G:30.764% B:2.462%) |
| Cubeless Equities | +0.0445 | +1.0443 |
| Cubeful Equities | ||
| No redouble: | +0.5522 (-0.4478) | ±0.0031 (+0.5490..+0.5553) |
| Redouble/Take: | +1.0443 (+0.0443) | ±0.0049 (+1.0394..+1.0491) |
| Redouble/Pass: | +1.0000 | |
| Best Cube action: Redouble / Pass | ||
| Rollout details | ||
| 10368 Games rolled with Variance Reduction. Moves: 3-ply, cube decisions: XG Roller | ||
| Double Decision confidence: | 100.0% | |
| Take Decision confidence: | 100.0% | |
| Duration: 29 minutes 20 seconds | ||
eXtreme Gammon Version: 2.10, MET: Jacobs & Trice