I guess the simplest method....assuming you had all the data....would
be to look at a player's historical win percentage in singles
by post position.
How often does Goiko win when he is in post 1...post 2...all the
way thru 8. Then use the average (mean) payouts when he wins in
post 1, etc. So that could be your win probability times your
expected payoff. Assume that generally his opponents are the same,
but adjust for when the other stars like Aritz, Lopez, Irastorza
are out. I don't know how you would quantify that adjustment....unless
you went so far as to look at how often does Goiko win in post
1 when Lopez is off for the day. Or the others.
In theory I guess you could run that for every player who plays
a singles game and may find a scenario or few where there is a
positive EV. Certainly not for Goiko...but maybe a lesser known
player gives you a positive EV and you try to outfox the public
that way.
