| |
BGonline.org Forums
Critical match score recube decision vs MCG
Posted By: John O'Hagan In Response To: Critical match score recube decision vs MCG (neilkaz)
Date: Tuesday, 2 June 2009, at 12:01 p.m.
Does Matt have a take? ATS, his naive 4-cube takepoint (if he just takes and never redoubles) is around 24% plus 38% of Neil's gammons minus 66% of his own. There's also recube vig to consider: Neil's 8-cube takepoint is around 31% ATS. We'll also have to consider the possibility that it's both too good and not good enough to redouble.
How many g's for each side here? I think it's about 40%-10% in favor of Neil. This makes MCG's gammon-adjusted 4-cube takepoint ~33.4%. Now we need to adjust for cube ownership. Neil needs 31% to take an 8-cube so, assuming 70% efficiency, cube vig figures to be worth around 7% (70% of 33.4%*31%). This means Matt should take if his cubeless winning chances are at least 26.4%. It looks to me like they are. He's got a few ways to win this game: 1)If Neil hits loose on the ace, Matt could win the blot-hitting contest (although he's an underdog to do so). 2) If a 2nd man gets hit, Matt figures to have at least a couple chances to roll D5 from the roof which should make him at least even money in the game. 3) He could make the ace point and either win going forward or hit a shot as Neil is bearing off. It looks to me that these possibilities are enough to give Matt around 30% cubeless chances in the game which is enough to take.
Should Neil redouble? The agruments for redoubling are that he's within 3-4% of Matt's takepoint and there are lots of mkt losing sequences on tap. These conditions normally mean that Neil's side should be redoubling. Could this score change things? Well...maybe. Let's look at how much match equity(ME) Neil gains or loses when he wins or loses a singe gamme or gammon. When he wins a single game, redoubling raises Neil's lead from 4-0 to 6-0. My OTB estimates for these scores are 72% and 83% so redoubling gains 11% when Neil wins a single. This happens 30% of the time so the gain from redoubling and winning a single game is 3.3% ME. Winning a g without redoubling gives Neil a 6-0 lead with 83% ME vs. the 100% ME if he does redouble. He wins 40% g's so redoubling gains 7.2% ME when he wins a g. The gains from recubing are therefore 10.5% ME.
Now we need to look at the other side of the coin. How much does Neil lose from redoubling and then losing a single or double game? Using the same method described in the above paragraph, I get 2.6% ME lost from single game losses and 3.7% ME lost from his double game losses for a total of 6.3% ME.
So the gains from redoubling outweigh the losses, right? It's actually not that simple. There is a flaw to the above analysis in that it assumes each side will win the same percentage regardless if the cube is turned now or not. This is obviously not the case since not redoubling this turn keeps open the possibility of doing so at some future point in the game. The net effect of this is that the above estimate of +4.2% ME gained by redoubling overstates the benefit of redoubling now. If he doesn't redouble now, Neil's CUBEFUL chances are certainly greater than the 70/30 CUBELESS chances that I he has if he redoubles now. So the question is whether or not keeping the cube will allow him to win enough extra games to overcome this 4.2% ME deficit. Hard to say for sure, but I don't think it does.
Another reason to recube is that this position is pretty hard to evaluate so there's a good chance of getting an incorrect pass. OTOH, maybe my estimates are off a bit and the opponent should pass.
R/T looks right to me.
| |
BGonline.org Forums is maintained by Stick with WebBBS 5.12.