The man in charge of the latter-est of those, Dan Szymborski, was kind enough to sit down (electronically, at least) and answer some tough questions.
Why do we find projections so compelling? Which projections have particularly surprised Szymborski? When is it okay to drink on a boat? All of those questions are answered in what follows.
Szymborski consented to be interviewed Sunday by means of EtherPad, a program that allows multiple users to create and edit a document. Hence, the “etherview” — the phenomenon that no one anywhere describes as “the single most important contribution to news media this year.”
Carson: Dan, I want to make some sweeping generalizations about you, but I’m having trouble. Like, you’re from Baltimore, I know. But the only thing I really know about Baltimore I learned from John Waters movies. Is that what your life is like?
Dan: Not the Wire?
Carson: No. But a smart friend recently chastised me for not having watched it. It seems, uh, gritty. Is your life gritty?
Dan: I’m not sure that I would describe it as gritty. In the context of the Wire, my life bears little resemblance to high-end drug dealing. The hardest thing about Baltimore is learning how to properly eat steamed crabs. You figure out how to eat a crab and which segregated mass transit portion to ride and you’re pretty much set as a Baltimoron.
Carson: Okay, I know two other things about you. One: you’re a Malamud on your mother’s side. Two: you play classical piano.
Two questions. One: Any relation to Bernard Malamud (author of The Natural, among other things)? Two: You ever bang out any Erik Satie?
Dan: Malamud was my mother’s father’s first cousin. They lived down the street from each other but didn’t really talk much after they went to their respective colleges. He was also my grandfather’s second cousin, due to a part of the family tree being unfortunately close to another branch.
As for Erik Satie, I’ve played a bit of Satie. He was kind of the Bill Simmons of Impressionism.
Carson: Okay, that’s a lot to work with. Let’s leave the, uh, “close relations” for your big Barbara Walters interview. But, wow, you really are a Malamud. I was thinking about The Natural and Roy Hobbs in the context of your work with ZiPS. I was wondering how you might go about formulating a projection for Hobbs. Great performances at a young age, long injury layoff: that seems tough to me.
Dan: Well, you would have to translate his minor league performances first.
Carson: He struck out that one guy (The Whammer?) on three pitches. Does that count?
Dan: Walter Whambold is such an unintimidating name for a professional baseball player. It’s like waking up one day and finding out that Callix Crabbe was voted MVP.
Carson: Do you think it’s the -b- sound? Like, Charlie Brown’s favorite ballplayer was/is Joe Shlabotnik.
Dan: Once we start talking about Charlie Brown, you’re dealing with a boy with a lot of serious psychological issues. Everyone knows that his baseball rash was self inflicted as a result of his overcontemplation of his bleak, Kafka-esque life.
Carson: Yeah, duh.
Wait, what are we talking about?
Dan: I’m the question answerer and that question is beyond my purview and responsibility!
Carson: Let’s move on then. Dan, you are the progenitor of one of the most respected and referenced baseball projection systems in use by baseball nerds today. When did you realize that you were becoming an authority on this sort of thing?
Dan: I think that moment came when I was lurking on a bulletin board a few years ago and someone mentioned ZiPS. One of the people was skeptical that I was actually Dan Szymborski after I made a response. That’s actually happened a few times since, but the first time that someone actually thought Dan Szymborski was a voice worth impersonating was quite an ego stroke.
Carson: So that’s one of the highs. What, if any, are the lows of being a capital-a Authority on baseball projections.
Dan: Trying to be polite and constructive as much as possible at all times when discussing projections. Like a lot of statheads (Law, Kahrl, Malcolm, Cameron, McCracken, etc), I cut my teeth on usenet in its prime. The sadly departed Doug Pappas actually kept an archive of ways he told this one person to off themselves. Sometimes, it’s fun to just get into a knockdown-dragout verbal brawl.
Carson: Are there anytimes when ZiPS spits out a projection and you’re just like, “Aw man, people are gonna jump all over that”? Can you think of any, in particular?
Dan: All the time! I’m always rooting for it not to underestimate Mark Buehrle. Kevin Gryboski used to screw me over every year. I always feel kind of guilty because there are a few players that I root for and root against every season to prevent me from looking dumb. Nelson Cruz is my best friend this year. I had also projected Tim Lincecum to be awesome as soon as he hit the majors and he never let me down. Rich Harden, on the other hand, has earned my ire for being both healthy (for him) and rather ordinary.
Carson: Do you ever looks at some other projections and think, Woah. Like, you mentioned Nelson Cruz. I know Bill James had Cruz’s teammate Chris Davis batting like .300 in 2009.
Dan: The Bill James-branded projections seem to have everybody hitting .300 with power. I actually tend to worry a bit when a projection of mine is as positive as the James projections.
Carson: How much time do you spend looking at the other ones out there? Either for reference, for comparison, whatever.
Dan: I try not to look at other projections until I’m finished with my projections. I don’t want to start second-guessing myself every time there’s a disagreement between projections. For example, I was really worried this year when there was such a discrepancy between my projection for Travis Snider and the James one, but after the CHONE came out, I did feel better about it. I never evaluate the results of ZiPS vs. others simply because I have an obvious conflict of interest. ZiPS, CHONE and PECOTA are close enough and there are many different approaches to determining the “winner” that I try to leave that analysis to 3rd parties like Tango.
Carson: Have any surprised/excited you so far this season? Like, you just released the Philly projections and minor leaguer Michael Taylor has a slash line of .284/.343/.456. By ZiPS’ standards, he’s major-league ready.
Dan: I was surprised that ZiPS likes Charlie Morton. I kind of poo-poohed his pickup months ago, but ZiPS likes him and his performance was hardly bad. I’m a little disappointed that ZiPS isn’t projecting a good Pat Burrell bounceback considering I liked the Burrell signing better than the Ibanez signing at the time.
Carson: When I asked you to participate in this absurd exercise, you said that you liked the timing. After a day of watching football (and presumably) drinking, you could become “Professor Dan.” What is the Professor Dan mode for you?
Dan: That is actually a reference to a Simpsons episode when Barney was drinking and Homer described a certain phase of inebriation as “Professor Barney.” My friend Phil started calling me that after we were combining boating and drinking with a group of friends and I started talking about the Battle of Loudoun Hill.
Carson: “Boating and drinking”: this doesn’t necessarily sounds like what I’d call a “safe” combo.
Dan: It’s kind of natural selection. For the future of humanity, we need to weed out those that don’t fall out of boats and drown.
Carson: For real.
Well, listen, I was hoping I could ask you to enter Professor Dan mode for a moment and answer a question. Here’s the question: What do you think it is about baseball proections that people find so alluring? Is it the idea of being able peek into the future? Is it something more banal — like just assembling a killer fantasy team?
Dan: Absolutely. Humanity has always had a fascination about the future, from TekWar to Magic 8-balls to Miss Cleo. The unknown has a certain attraction because the future is all about possibilities. There’s kind of a romantic notion that a year from now, the whole world could be drastically different from any reason. Or the world could end in 2012, like Darren Daulton warns us. In addition, some of the future, because of our inherent mortality, is always hidden behind a veil that we’ll never see.
Carson: Does ZiPS have any other projections for the future besides just the baseball ones?
Dan: Lottery numbers, which I will gladly share at $9.95 a minute. Projecting what happens in baseball is already hard enough. I’ve always toyed with trying to do something similar with football, but there are some greater challenges there given the interaction between players.
Carson: Finally, any messages you need to get out to your adoring public? Like Brian Bannister, for example? He’s a big reader, I’m led to believe.
Dan: My main concern with the use of stats is that sometimes people do take the stats too seriously and aren’t quite aware of the weaknesses in various stats. Too often, people write acronym soup and come to conclusions too strong for the data involved. For instance, too many people don’t seem to quite understand that 1-year UZR data shouldn’t be taken as gospel for what a player’s defensive abilities are. Some people do take projections too seriously – the computer projections are generally designed to be objective projections about statistics, but the computer simply doesn’t know things like who will get playing time and who is or was injured. Just today, someone reviewing the Phillies ZiPS projections was a bit upset that ZiPS projected Jayson Werth to get fewer at-bats than in 2009. ZiPS or PECOTA or CHONE can’t say why Jayson Werth didn’t play more in recent years. That’s not really the best use of a projection system.
Carson: Can I put that last paragraph on your gravestone?
Dan: I plan to have a pyramid erected in my honor, so you can certainly engrave that somewhere on the base, if you wish.