[ View Thread ] [ Post Response ] [ Return to Index ] [ Read Prev Msg ] [ Read Next Msg ]

BGonline.org Forums

How did you analyze it during play?

Posted By: John O'Hagan
Date: Tuesday, 15 March 2011, at 11:10 p.m.

In Response To: How did you analyze it during play? (Casper van der Tak)

When the leader is 5a and redoubles to 4, my OTB rule of thumb is that the trailer's basic takepoint (if he takes and never redoubles) is around 25% - D/2 (D is the difference in scores). So it starts at 22.5% since D = 5.

Then estimate the gammon values. I've got 3 off so I can't get g'd, but I could win one on a real good day. When the leader is 5a, my RoT for the trailer's 4-cube gammon value when there's no overage is 80% + D, so it's 85% ATS.

Then estimate gammon wins. Pretty unlikely, probably about 1% - 1.5%.

Then subtract gammon chance x gammon value from basic 4-cube takepoint. This lowers the takepoint to around 21.5%.

Then figure out leader's 8-cube takepoint (using Neil's numbers except for the 2a score). Drop makes the score 5a-6a which is about 57% MWC, take and lose makes it 5a-2a with about 26% MWC, take and win wins the match with 100% MWC, take and lose a g loses the match with 0% MWC. So the takepoint is around 42%.

Then deduct cube vig from the gammon-adjusted 4-cube takepoint (which was 21.5%). Cube ownership is usually worth about 70% of the product of the gammon-adjusted takepoint and the oppt's cubeless takepoint at the next level. You then deduct this amount from the gammon-adjusted takepoint. So 21.5% - .7 (.215 x .42) = 21.5% - ~6% = 15.5%.

Then ask urself if you think the cube will be worth this much in this position. If the cube is worth less or more than normal, adjust the cube vig factor accordingly. Here I thought it would be in the normal range since I could probably give a pretty efficient recube if XG fails to cover, I come in and out, and it fails to hit back. Also, occasionally later after he closes me out but I enter right away and then roll a good set.

Then try to determine if your cubeless winning chances are at least 15.5%. 20 numbers cover and I win around 11% if I was completely flush and he has perfect spares on the 6-5-4. I won't be flush but he'll probably end up with something close to perfect so 11% is about right. That's 2.2 wins. 16 don't cover and I enter on 4.8 of those. Around 2.5 of these also come out, .5 keep me on the 22, and the other 1.8 enter and expose another blot or two. I gave myself 1.6 wins in these 4.8 games for 3.8 wins in 24.8 games so far. The remaining 11.2 games are from the no cover/dance sequences. These must be a little better for XG since he will have even more covering numbers on his next try. So maybe I win about 1.45 of these remaining 11.2 games.

That's 5.25 wins in 36 games whereas 15.5% of 36 is 5.58. A little short of takeland but I've noticed that I usually underestimate the dogs wins (and maybe his g's as well) in this kind of position so I figured this plus the fact that I was up against a better player in a simple position would be enough to make it a take.

Messages In This Thread

 

Post Response
Your Name:
Your E-Mail Address:
Subject:
Message:

If necessary, enter your password below:

Password:
 

 

[ View Thread ] [ Post Response ] [ Return to Index ] [ Read Prev Msg ] [ Read Next Msg ]

BGonline.org Forums is maintained by Stick with WebBBS 5.12.