Baseball Prospectus - Chat with Clay Davenport

Chat: Clay Davenport

Welcome to Baseball Prospectus' Monday September 13, 2004 8:00 PM ET chat session with Clay Davenport.

Clay Davenport is an author of Baseball Prospectus.

Clay Davenport: Greetings everybody. This is Clay, live from Maryland. What is going on tonight?

John M. Perkins (Macon, GA): When you expect the short season stats to reappear in the minor league EQAs?

Clay Davenport: Didn't realize they'd dropped off; something to do with the end of the season. They'll be back on tomorrow.

Marlon (Atlanta): I've got a trivia question that I need help with. What are the seven ways to reach base without getting a hit? I know the following; error walk hit by pitch catcher interference fielder's choice catcher misses third strike

Clay Davenport: If I recall the trick to that question correctly, being inserted as a pinch runner. It's a lousy trivia question, but it has been around forever.

Dan (Vermont): Clay, Did you come up with EqA? If so, how did you come up with such a useful metric? Did you just play around with certain numbers and see what came out until you found a formula that you liked? I'm interested in coming up with new metrics, but I don't know where or how to start.

Clay Davenport: Yes, I did come up with EqA. It originated by following up on idea Bill James had in the 1985 Abstract, regarding secondary average. In its original form, it was one part secondary average and two parts batting average. That was the base; it's been fleshed out quite a bit since then, with a lot of input from others.

I wasn't trying to create a new metric; I was just testing something out and found it was useful. As far as coming up with new metrics, my recommendations would be: know waht others have done so you don't reinvent the wheel, learn enough statistics to know whether yourwork is valid, and figure out what question you are trying to answer.

jschmeagol (new holland, pa): Hey, I was wondering why you guys haven't developed a play by play defensive system yet. Not that FRAA is a bad stat or anything but it seems that a system developed on PBP data is more methodolgically sound than one built on team performance like FRAA and defensive Win Shares. Do you guys see a problem with stats like UZR and ZR that I don't see? Thanks for answering.

Clay Davenport: Two parts to answer that: one is that my own interests lie with uniting baseball's present with its past; I don't like spending my time with a method that is only useful for the major leagues today, which are the only leagues that are going to have the requisite data for a PBP method. So that's one, admittedly personal, bias against working towards such a system.

The second part is that I don't have any interest in creating system that is essentially a copy of someone else's work. Unless/until I can find a truly different angle of attack on the fielding problem, I'm not going to make soemthing like that.

That's not to say I wouldn't use a good idea of someone else's as part of such a work, but there has to be some original Clay in it. I have *some* pride, after all.

Justin (Los Angeles): Who starts Game 1 in any series for the Red Sox?

Clay Davenport: Me, I'd send Schilling. I love Pedro, but I trust Curt more if I need someone to go on short rest.

Brian L Cartwright (Johnstown Pa): I really like the major league equivalent counting stats that are now included on the minor league batting and pitching DT's. It gives a way of reading the translations in numbers we are familiar with. Coudl the same kind of normalized stats be included on the major league VORP's? In the same way, we could see how park effects and the other factors skew the stats. It would also make it possible, whenever a player has performed in both the major and minors, to sum up all that player's stats for the year into a combined line - which would also then allow a depth chart, by position, for each team.

Clay Davenport: The numbers are available, sort of, on the Player cards, although they are listed for each individual or team, rather than by league leaders.

I can produce the summed line pretty readily, but don't you think I should save *something* for the book?

David Crowe (BC): Using translated Japanese EQAs what would Ichiro's career numbers look like now assuming he reached the majors as soon as he was above replacement level?

Clay Davenport: I only have his stats since 1994, which gives him 11 seasons. Over those translated numbers, we're looking at a career .335 hitter with over 2000 hits and 300 stolen bases and, say, five Gold Gloves, and a reputation that he took over from Henderson as the best lead-off man in baseball.

Justin (Los Angeles): Obviously, the Yankees have so many long-term contracts. For them to maintain their standard of excellence for the next 5 years, are they going to have to add 30 million to their payroll annually?

Clay Davenport: Not necessarily, but most likely. Their problem is going to be being locked in to contracts purchased at the top of the market, which combines with player's ages to make those players hard to trade. Age also means declining performance, and hopefully enough teams are catching on to that to keep from playing suckers to George. Getting the fresh talent is going to be a problem, and money is certainly the easiest route to getting it. It would be fascinating to see them figure out another way, though, much as I dislike them.

deadmonkeypaw (Chicago): I've read some criticisms of WARP3 that say that judging a player against both a defensive replacement level and an offensive replacement level is double counting. According to this claim, a player who is both a replacement level hitter and fielder would never get even close to the field. Do you have any sort of response to that?

Clay Davenport: It certainly is not double-counting; it just means that my overall replacement level is lower, on the order of a AA player rather than AAA.

And since there are numerous players who finish below even this extra-low replacement level, picking up a negative WARP-3 (the Orioles have 13 such players right now, although admittedly most have hardly played), the claim that they would never get close to the field is demonstrably false.

Adam J. Morris (Houston, Texas): How confident are you in the reliability of the defensive stats provided in the Davenport Translations? One of the things that I find a bit disconcerting is the relatively high level of volatility of the defensive RAAs from year to year for players. Mike Cameron, for example, is considered by scouts and statheads alike to be an elite defensive CF, but his DTCard reflects only one great defensive season, with his other years being just a little above average. Mike Young goes from being a well above average defensive 2B in 2001 and 2002 to a well below average 2B in 2003 -- much worse than Alfonso Soriano in 2003, in fact. So, what are your feelings on these variations? Do you think they accurately reflect the variations of a player's defensive abilities or contributions from season to season? Or do you believe it is just a side effect of the imperfect nature of defensive statistics?

Clay Davenport: I would only say 'moderately so', and even that depends on position. I'm not especially confident on ratings for first basemen; I'm really not confident on ratings for catchers when I don't have stolen base and caught stealing numbers. I have more confidence in the total outfield rating than I do in breaking it down to positions and individuals.

However, I do think there is more year-to-year variation than most people think. Players can learn to correct fielding mistakes (not just errors) the same way they can learn to lay off outside sliders. They get hurt and have their range go to pot. They have runs of bad luck. For a full-time player, I think the normal variability in the fielding stats is roughly 10 runs a year compared to an average fielder. But add in the uncertainties, and they are many, in the fielding statistics, and we see 15 or even 20 run differences more than we should.

Daniel Rea (Sherborn, MA): Why is it that John Schuerholz has been so succesful, while his proteges have done so poorly? Dick Balderson, Dean Taylor and Chuck Lamar have all failed miserably on their own with "Bravesball", so is Schuerholz a bad mentor, a poor picker of assistants, or someone whose style is virtually inimitable?

Clay Davenport: All of those are possible answers; my own gut-level answer is, 'His proteges didn't take Bobby Cox with them when they left'.

nuckleball (Va): I was checking some stats and saw that Randy Wolfe of the Phillies has some great batting numbers for a pitcher. .273, 12 hits, 3 HR, 8 RBI, 2 doubles in only 44 AB as of august 30th. Is this just a fluke of the stadium or does he just have good power? The next Highest batting average for a pitcher there is .182 for Paul Abbott.

Clay Davenport: Wolf has been an above-average hitter for a pitcher, although not this good. It is not the stadium; while he has what looks like a substantial home/road split (292/280/540 at home, 238/273/476 road), because we are only talking about ~22 pa apiece, the difference is basically one hit.

rmf101 (Colorado): Clay, one more time for us Rockies fans. How do you build a team with limited financial resources that plays at Altitude?

Clay Davenport: If I knew, I'd be working in Denver.

Mark (Portland, OR): Hello. I was wondering if, three years out now, the increased scrutiny of international identity papers has had any significant effect on the average age of international prospects signed by the majors. Have any significant or supposedly significant prospects had their expectations lowered by finding out that 18 year old phenom was really 20?

Clay Davenport: The Orioles were ridiculously high on Ed Rogers a few years ago, before they found three missing years (and I let them get away with it when I thought he was that young myself, not criticizing them nearly enough for their comparisons to Jeter, much to my chagrin). Pablo Ozuna was supposed to be a big deal after a monster Midwest league season, and was a key player in a trade; that didn't work so well. Those are the ones that stick out in my mind.

Suraj (New York): Say Barry Bonds eclipses Babe Ruth's HR total. With his six (soon seven) MVPs, nine Gold Gloves, records for most HRs and walks in a season, being the only player to reach 500-500, is he the greatest baseball player of all time. Is Ruth's stint as a pitcher really that important?

Clay Davenport: Whether or not he eclipses Ruth's total in HR isn't that important. No one had a single season like Bonds' last four; the way other teams avoid him dwarfs anything else in the history of the game as best I know. A couple of years as a pitcher is important if you are trying to answer a question about actually fielding a team of nine clones, one for every position, but then time spent as a catcher (Jimmie Foxx!) would work in your favor as well.

Still, by my way of ranking all-time players, he probably won't quite catch Ruth. He's probably passed Willie Mays this year; he has a good shot at Walter Johnson and Ted Williams, who I have at 3 and 2; but Ruth is just too far ahead.

Clay Davenport: Thanks for all the questions, and I'm sorry I couldn't answer every one, but I need to write up an article I was supposed to finish this weekend. Thanks to all of you for your time reading BP, Clay