Wasn't finished....
So either Sean's game isn't as good as I though or his area is just out of whack. I believe he has run over 100 balls several times. I don't know any 100 ball runners that aren't 7's. So my hunch is these guys were on a bit of a rating island and what happened to them was due to this.
I like the APA, but it's not flawless. Over about 25 years, it's clear to me that the handicaps in a division can wander off in the high direction and once they get there, they stay there.
In the land of the blind, the one-eyed man is king. In a division with no real 7s, the best 6 will become a 7 eventually. That player will have a very high win percentage as a 6, and the system will lift them up to a 7 once they have a few games out of their last 30 that are at 7 speed (even if they have other games at 5 speed). At that point, there's a weak 7 in the division, but the 7 won't drop unless he goes below 50%, which is a terrible win % for a 7. This will lift other players up: 4s will beat that 7 in a 5-2 race (much bigger than the correct 5-3 spot . . . 250% vs. 67%) with fairly low innings. That pulls the 4 up to a 5, but it's really because the 7 is weak. Now the over-handicapped 5 causes 2s and 3s to win a bit more often and they go up.
I could go on, but the point is that if there aren't good 7s in a division then handicaps tend to drift up. Conversely, even just a few good 7s, even better a super 7, will be enough to keep the 5s and 6s where they belong, which keeps the 3s and 4s where they belong, and so on. Therefore, a team that comes out of a division with stronger players will tend to have lower handicaps.
I can't know, but it is entirely possible that the team in question never sandbagged yet was legitimately under-handicapped.
The APA relies on innings, which has been reasonably good for them over the years. But it really doesn't fully equalize across disparate divisions. That's one of the strengths of FargoRate: a fairly small number of connections (even second order connections) among two disparate pools of players is enough to make the ratings in the two pools comparable. To my knowledge, the APA doesn't have anything like that. But they should.