People were arguing over whether Alex Rodriguez deserved the 2005 MVP award over David Ortiz, because David Ortiz had so many late-inning home runs. But the fact is that A-Rod hit more home runs than Ortiz did if the Yankees were tied or behind. And the Yankees had a better record of late-inning comebacks than the Red Sox. So, yeah, David Ortiz was on SportsCenter or Baseball Tonight hitting all these clutch home runs. But there are other things you can do to have an effect on a game. They’re just less likely to show up as a celebratory clutch homer. Clutch can come anytime in a game.
MM: One of your essays in the book is called “Can a Team Have Too Much Pitching?” It’s a baseball maxim, of course, that you can’t. It’s a situation in which the Red Sox currently find themselves [or did, at least, until they traded Bronson Arroyo].
SG: It’s all relative. In a literal sense, in the old days if you had spots for 10 pitchers on your team and you had 20 players for those spots, you controlled the players in every aspect of their lives and they weren’t paid that much, so you could just stash them in the closet somewhere in Poughkeepsie or something, and call them as you needed them. If a team today has excess pitching, they can get rid of them, or they can trade them. It’s not that easy to take excess pitchers and put them somewhere.
Baseball is a very simple game. You score more runs than the other guy, you win. So one aspect is offense, and one aspect is run prevention, or defense. The fact is that if you overbalance your team in one direction or another, it becomes very difficult to win. There’s a predictable relationship; Bill James called this the Pythagorean theorem, because the formula he found for calculating this looks a little bit like the old A-squared plus B-squared equals C-squared from grade school.
ADVERTISEMENT
 |
I.E.,
RS2
Winning Pct = WPct = -------------------
RS2 + RA2
And that is very good at showing the relationship between runs scored and runs allowed. Essentially, every 10 offensive runs you add to a team with, say, league-average pitching, is another win. Every 10 runs you take away, from the defensive side, is another win. So what I’m trying to get at is that you can have too much pitching if you don’t have everything else. And the reverse is true, also. If you overspend on the pitching side, as teams like the Twins have done — there’s a table in that chapter that shows the teams who have — they don’t go that far.
MM: Much has been made of this supposed conflict between statheads and scouts. Can’t we all just get along?
SG: It is a false dichotomy. You can’t do what we do without some recourse to scouting. And you can’t do what they do without some recourse to statistical analysis. When you get down to the very basics of everything, statistics are an objective record of something that happened. And baseball is filled with so many small moments. Over the course of a season, a regular player comes to the plate 600 or maybe 700 times. I don’t care if you’re the world’s best scout, no scout is assigned to one player to memorize his 700 plate appearances. There’s no time to watch that much video. A scout, if he’s talking about a minor-league player, pops out of his car and might watch the guy [hit] two or three times. And then he pops back into his car and has to render a judgment about that guy. He doesn’t have the ability to see everything, and even if he did, no one has a memory that good. So you have the statistical record. The guy batted 700 times, he has this many hits and this many home runs, and he struck out this many times.
What analysts do is, they further take that record, and they can look at it on a historical basis. This is what we do very successfully at BP, with our PECOTA [Player Empirical Comparison and Optimization Test Algorithm] prediction system. But you don’t even really need PECOTA if you have common sense. You can look at it and say, the scouts say that this player has great athletic ability. He’s 21. He played at A-ball last year. He’s very fast, he has a great body, incredible reflexes. So that’s the scouting end of it. Then you look at, well, he had a .280 on-base percentage and he walked only 30 times in 650 plate appearances. Looking at the way that players with that profile developed over the last X years, we can tell you that’s just not gonna work out most of the time, that this is somebody you should be very skeptical about, this is not somebody that the team should trade for, or at least not trade premium talent for, unless they have some special insight into making this guy a more rounded player. And that’s where the conflicts come up sometimes. The scout says, “This guy has talent.” And we say, “Well, the execution’s not there.”