| |
BGonline.org Forums
arguing against the definition
Posted By: rambiz In Response To: What I did over the board (Timothy Chow)
Date: Tuesday, 21 June 2011, at 8:58 a.m.
Granted! "Arguing against the definition" would have surely been the more appropriate expression. EPC is the definition for the length of the race, whereas Thorp's count, Keith's count etc are definitely no definitions. I assume you don't want me to define the term "definition", but I am going to elaborate on my understanding of it....(feedbacks highly appreciated)
OK! If we want to handle the race, the first thing we want to know for sure is the "length" of it or a "measure" for its length. EPC devided by 8.167 gives the average number of rolls to bear off all checkers. This is the "definition" of EPC! So what is actually going on here, is that we assume we knew the average number of rolls and then we go on and calculate EPC. Now, if we are smart enough, we can get out, what we have already put in. The apparent shortcoming of EPC to fully qualify for the prestigious title "Holy Grail" is that it doesn't tell us the exact distribution (that is percentages of successful bear offs in 1-roll, 2-rolls,3-rolls, etc) . It is only a measure for the average. However, I think, EPC has a tighter grip on races than any other effort to linearize a highly non-linear function. That function being the percentage of wins as a function of the distribution of checkers.
| |
BGonline.org Forums is maintained by Stick with WebBBS 5.12.