Could be luck.

Also, maybe if everyone else is using stats, then the one team that doesn’t use stats gets a competitive advantage. Even if its method is inferior overall, the team will be able to get the very best players ranked under that method since everyone else will be competing for a different pool of players.

Or, maybe the Phillies are using non-statistical heuristics that correlate with the results a statistical analysis would give them – so they’re doing the same thing, just unconsciously. (Is that the right word, “heuristic”? Maybe I mean “proxy”?)

It doesn’t really make sense as written. I wonder if they meanto to say that the Phillies do not technically employ a “statistical analyst”?

This actually makes a great deal of sense to me. I am not against advanced stats by any means, or statistical analysis in general. Also, I am not very familiar with it. But from what I read on blogs and such that like advanced stats, it appears that they are not really all that helpful in making relatively major personal decisions. As a brief (and probably terrible, I dont usually come up with good ones) analogy, imagine you are an investor with 100 million dollars to invest. Would you a) look at the past ups and downs of a stock and then invest money based on statistical probabilities of those re occurring or b) research into the companies, figure out what products they are releasing, who is in charge and there history and all the other stuff like that.

Method a) would be analogous to stats whereas method b) would be analogous to a scouting department, also including coaches and other stuff like that. Method a) is much cheaper, but if you are investing so much money it probably makes sense to spend some money on the better information. Probably it makes sense to do both though, haha.

Also, again not to dismiss stat studies, but something like WAR I imagine would be almost completely worthless to a Front Office. WAR basically collects a bunch of data, and then mashes it into one number. Why not, if you have the resources, just analyze all the raw data?

All that said, I imagine things like OBP, HR and other similar stuff that are essentially counting stats are used. Those would basically just be raw data. It wouldnt make much sense not to ask how many HRs a guy hit last season! Even then, though, you would probably want to look at the guys home runs and determine 1) if he was lucky/unlucky, because of the Park, an excess of meatballs etc. 2) if he has “power” in his swing, such that his HR total from the previous year makes sense with what informed people generally know about hitting mechanics. There is probably more qualifiers on the HR stat, but that is all I can think of right now.

Interestingly, though they are the least reliable, I imagine defensive stats are actually the most helpful to a Front Office, as comparing Defense among players is very difficult, even if you pay a ton of scouts to watch a ton of film/games etc. It is probably an element of the game that is most susceptible to bias of human perception, though I could easily be wrong in this.