I finally see something useful for WPA, as it explains the discrepancy between the Red Sox record and their Pythagorean projection:
Again, WPA is a story stat. And a story is ALL about timing. It’s like comedy. I’m presuming the Redsox scored alot of runs when they didn’t need them, and the Giants scored their runs when they really needed them. Giants are funny, and Redsox are boring, is the story here.
Redsox have as many wins as losses, and yet they scored 47 more runs than they allowed. If anything, you should be disbelieving that they actually scored 47 more runs than they allowed! That their WPA on offense is just below average is consistent with their W/L record. Indeed, WPA is EXACTLY equal to their W/L record.
Story stats tell you why things happened, they don’t tell much about ability. The Red Sox are a good offensive team. I like to look at the distribution of runs in games, which is flat instead of bell shaped for the Red Sox. They win big but lose small, another way of saying their timing is bad.
Okay, it’s a “story stat”. Not only does it say nothing about ability; it doesn’t even really give an explanation. The Red Sox have a total WPA of 0. But that doesn’t *explain* the fact that they are a .500 team. It’s just the same fact! (Any .500 team has to have a WPA of 0, and any team with a WPA of 0 has to be a .500 team — they’re just two ways of saying the same thing.)
They actually hit well with runners on base and RISP. Here are their BA-OBP-SLG-OPS with
RISP) 0.283 0.359 0.458 0.817
None on) 0.260 0.313 0.425 0.738
ROB) 0.279 0.349 0.463 0.813
Now their overall BA-OBP-SLG-OPS and in close and late situations
Overall) 0.268 0.329 0.441 0.770
CL) 0.241 0.310 0.383 0.693
How can a team which hits so well with runners on base do so poorly when it is close and late? Are they at a platoon disadvantage? Just bad luck?
Now the same thing for their pitchers
RISP) 0.256 0.343 0.432 0.775
None on) 0.265 0.341 0.431 0.772
ROB) 0.251 0.306 0.395 0.701
So it looks like their pitchers do much worse with runners on base.
Now their overall BA-OBP-SLG-OPS and in close and late situations
Overall) 0.257 0.321 0.410 0.731
CL) 0.271 0.335 0.403 0.738
So their pitchers do worse when it is close and late.
Overall, they have an OPS Differential of .039 (.770- .731). But when it is close and late the differential is negative. It is -.045 (.693 – .738)
James: WPA is DESCRIBING a team that scores 47 more runs than they allow, winning as many games as they lose.
It doesn’t have to do more, just like OBP doesn’t have to do more than it does (treating BB=HR), and just like Kate Upton doesn’t have to do more than she does.
If that’s not good enough for you, then fine, you have your own tastes in parties.