[Anyong Bluth]

Originally Posted by duncan_idaho

I like WAR a lot. It just doesn't explain 100 percent of what happens for a team. I'll buy 80-90 percent as a baseline.

Even at that missing 10 percent, though, that leaves a lot of room for variance. Would an 8-win swing in either direction for a team whose WAR says it should win 85 games be shocking to you? That would make the WAR prediction 90 percent accurate.

80 percent, and all of a sudden you're talking about a 17-win swing.

I think it's a good measure but you can't rely on it without context. Same with base runs or run differential.

For example, you can have a bad team that has a lot of power and bashes the crap out of back-end starters that looks better in run differential than its record because it is blowing teams out once every few weeks.

Or you could have a good team that doesn't have an explosive offense that looks worse in run differential because they just don't blow teams out very often.

Pitching, a dominant bullpen, speed, as defense are things that can lead to a team being really great in tight games and winning them at a high percentage but not having impressive run differentials.




Sent from my iPhone using Tapatalk

Rep