I clearly need to look at these more. I guess what isn't clear to me is the real explanation - is splitting simply bad when significantly behind in the race, or is something else about the position (stripped midpoint, 5's that play badly that now can play as 6/1*, etc.) that makes splitting wrong. I've been looking at all sorts of variants to try to get a handle on this. For example, here are two where the 3 point is missing and the race lead instead comes from a third man back:
| gnubg | 167 |
|---|---|
| |
| Susan | 182 |
Position ID: 4HPwATDQOfIAOA Match ID: cIkSAAAAAAAA | |
| # | Ply | Move | Equity | |
|---|---|---|---|---|
| 1 | 4 | 24/20 13/8 | -0.075 | |
| 0.491 0.136 0.006 - 0.509 0.165 0.012 | ||||
| 2 | 4 | 24/15 | -0.103 ( -0.028) | |
| 0.487 0.124 0.005 - 0.513 0.171 0.013 | ||||
| 3 | 4 | 13/9 13/8 | -0.143 ( -0.068) | |
| gnubg | 167 |
|---|---|
| |
| Susan | 181 |
Position ID: 4HPwATDQOfIANA Match ID: cIkSAAAAAAAA | |
| # | Ply | Move | Equity | |
|---|---|---|---|---|
| 1 | 4 | 24/20 13/8 | -0.061 | |
| 0.494 0.139 0.007 - 0.506 0.167 0.011 | ||||
| 2 | 4 | 13/9 13/8 | -0.091 ( -0.030) | |
| 0.488 0.141 0.006 - 0.512 0.178 0.015 | ||||
These two both indicate splitting even though way behind in the race. The two variants are simply to try to control for how many blots are left to attack after splitting and not, and it doesn't seem to make much difference. Here's a similar position with the 3-point made:
| gnubg | 159 |
|---|---|
| |
| Susan | 181 |
Position ID: jGfwATDQOfIANA Match ID: cIkSAAAAAAAA | |
| # | Ply | Move | Equity | |
|---|---|---|---|---|
| 1 | 4 | 13/9 13/8 | -0.303 | |
| 0.450 0.123 0.005 - 0.550 0.228 0.028 | ||||
| 2 | 4 | 24/20 13/8 | -0.303 ( -0.000) | |
That makes the plays close, but splitting is certainly not clearly wrong. I will spare you the position with the 3-point made and all three back men on the 24 point, it comes out much the same. I'm not an idiot, I know a third man back changes a lot. However, it does seem to indicate that it's not just about keeping the anchor/not leaving shots when way behind in the race. the other positions here and in my previous post indicate it's also not just about the relative strength of the boards.
I guess I'm just having some trouble fitting this opening knowledge into any larger understanding. The difference between what happens after 55P and 53P is clear enough, but I don't know how to generalize that knowledge to use it to decide whether to split in other positions that are at least superficially similar. If you can answer that, I also have a question about how to probe a moyo built around an ogeima shimari ... probably OT.