I am not seeing at all how the efficiency will be extremely high. The cash point going to 8 is 58% (ignoring gammons). It will often be correct though to double on the very next roll with only about 20% win chance.
For example the best play with 32 according to XG is 18/15 13/11 and then it is a small cube. I was guessing it would be a close call on that cube and even if it were no double but close it also would mean cube efficiency is low since a big market loss is being risked. Also note that the hop and miss market loser is of a repeating nature. Thus it will be unlikely to "time" the recube efficiently. There are a lot of immediate positions giving this close choice. Thus I think the TP is not far from the TP assuming an autorecube of 17.4% lowered by gammon vig on an 8 cube where the gammon value is only about 1/3.
After 32 it is a small cube to 8 despite only 19% win chance and 5% gammon chance:
                      |
| ||||
| XGID=--aBBBB-B--ABA-A------cdg-:2:-1:-1:00:6:1:0:11:10 | |||||
| Blue on roll, cube action? | |||||
| Analyzed in Rollout | No redouble | Redouble/Take |
| Player Winning Chances: | 19.251% (G:4.980% B:0.003%) | 19.247% (G:4.974% B:0.052%) |
| Opponent Winning Chances: | 80.749% (G:3.874% B:0.086%) | 80.753% (G:3.593% B:0.066%) |
| Cubeless Equities | -0.5443 | -0.4136 |
| Cubeful Equities | ||
| No redouble: | -0.4433 (-0.0297) | ±0.0051 (-0.4484..-0.4382) |
| Redouble/Take: | -0.4136 | ±0.0056 (-0.4192..-0.4080) |
| Redouble/Pass: | +1.0000 (+1.4136) | |
| Best Cube action: Redouble / Take | ||
| Rollout details | ||
| 1296 Games rolled with Variance Reduction. Dice Seed: 16132000 Moves: 3-ply, cube decisions: XG Roller | ||
| Double Decision confidence: | 100.0% | |
| Take Decision confidence: | 100.0% | |
| Duration: 4 minutes 14 seconds | ||
eXtreme Gammon Version: 2.10, MET: Rockwell-Kazaross