Here let me explain it again. It has nothiing to do with the fact you're mostly playing lower skilled players, it has to do with the formatting when you're playing against lower skilled players. You're almost exclusively playing handicapped tournaments at the lower levels, and they are designed to even the playing field. So if you're opponent only has to go to say 3 games to your 6 or 7, they will often get to that 3 or 4 games before the opponent gets above 2,3 or 4 games. This artificially creates an input into fargo that makes the higher rated player appear lower than they actually are. You may 99 out of 100 times if playing the full race to 6 or 7 end up 7-2 or 6-3. If those scores are inputted iinto fargo, the fargo will remain accurate.
But because of the nature of handicapped tournaments, those other inputs will keep the better player artificially low and the lower player artificially high.
I think the issue is being misunderstood a bit.
The mere fact that one side is going to 3 and the other is going to 6 does not, by itself, make the stronger player look artificially weak in the game data.
In a handicapped race,
both players have a stopping rule. In a 3–6 match, it stops either when the weaker player gets to 3 or the stronger player gets to 6. So it’s not really true that the weaker player is uniquely getting “artificially early” scores entered while the stronger player is not.
A good test is this: imagine a biased coin that comes up heads 2/3 of the time and tails 1/3 of the time, with heads racing to 6 and tails racing to 3. If you simulate a huge number of those matches and then total up all the flips, the overall heads-to-tails ratio still comes out 2 to 1 when the numbers get big. In other words, the stopping rules do
not by themselves distort the underlying game-win ratio.
What the handicap changes is the
match win probability, not the underlying game ratio.
