| |
BGonline.org Forums
OLM 20160130A The Prime Factors Consultation
Posted By: Casper Van der Tak In Response To: OLM 20160130A The Prime Factors Consultation (Jason Lee)
Date: Sunday, 31 January 2016, at 8:59 a.m.
18/10 and it does not seem close at all.
Consider: of the 17 hitting rolls we leave, the Dillies should not hit with 51, 42 and 22. So we are looking at 12 rolls that will actually hit. Of these, 62 and 64 hardly gain vs 62 and 64 when we break the board. So we are looking at 8 rolls where breaking the board clearly loses relative to the alternative. The degree of loss is limited because of the returns we get after 61, 63 and 66 (5 rolls). 65 and 33 are super jokers (3 rolls). say we lose 0.2 on the first 5, and 0.4 on the last 3; total loss 0.2*5 + 0.4*3 = 2.2
On 24 rolls, we gain. We gain by not having wasted pips for the race, and we gain by keeping a board which makes our hits semi-decisive (we can cash after a dance, which we cannot if we are reduced to a 4-point board). Not having wasted the pips alone should be worth a lot more than 0.1 ppg which would about equalize the gains and losses from running now. 0.1 ppg would be a 5% swing in gwc; say of the 8 pips we play, we avoid wasting 5, then it would be 1% gwc per pip, which is very low, it is probably more like 2.5%. So run and break the anchor now.
I did not try to estimate by how much 18/10 is correct. My guess would be that we gain over 0.3 ppg on the 24 non-hits; then we end up with a net gain of over 5 points over 36 games, which would mean that anything but 18/10 is in blunder territory.
| |
BGonline.org Forums is maintained by Stick with WebBBS 5.12.