[almostbusto]

no adjustments made:

answer is super easy if you have taken a 100 level stats class.


the result of a heads up match is a binary variable (1 = win, 0 = loss). this means the results of a heads up match follows the binomial distribution. moreover, standard deviation of a binomial variable is sqrt(n*P*(1-P)) where n is your sample size. this means that variance is maximized when your winrate is 50%. so people who think they can lower variance by playing less optimal are mistaken. play as optimal as you can and you will maximize the probability that you win that leg of the match and the match itself.

so in short: maximize your edge over the other guy and you will maximize your chances of winning AND minimize your standard deviation (that is to say its the min standard deviation you can achieve while still being a winning player, i add this caveat because everybody could have a standard deviation of 0 if they forfeited every match).

so if you were maximizing your edge the first match, and he hasn't changed, then you shouldn't change.