Few teams draft as well as the Dodgers, and even fewer do so with a data-driven approach that melds risk aversion with a gunslinger’s bravado. Logan White and company aren’t shy about taking high school pitchers in the first round, nor are they unwilling to take a calculated gamble that a highly-prized sleeper will fall into their laps in a later round. The results have been impressive, with the likes of Clayton Kershaw, Matt Kemp, James Loney, and Russell Martin already in Dodger blue, and Dee Gordon, Chris Withrow, and Ethan Martin soon to follow. White, the club’s assistant general manager in charge of scouting, sat down with Baseball Prospectus to talk about his approach to the amateur draft and why "The Dodgers Way" has been so successful.
David Laurila: Scouting directors have reputations. How do you think people around the game view you?
Logan White: That’s a great question, because I’ve never been asked about it and have never really thought about it. My concern is that my staff respects me, so I would hope that if I have a reputation it is that people respect the job I do and that I have the knowledge base and a plan.
DL: Taking the best player available is a mantra among scouting directors, but it really isn’t that simple, is it?
LW: You’re right; it’s not that simple. We always say we want the best available talent when we’re picking, but I like to think that I’m a little bit more of a strategist. Sometimes you might take a player who is a little lesser than another player if you think you can flip them and get both, whereas if you don’t do it that way, you won’t. There is strategy involved and we’re not afraid to use it. When I first started, a lot of the major-league teams weren’t drafting as many high school players and were strictly college-oriented. That was in 2002 and 2003, particularly teams like Toronto, and knowing that we were certainly going to be more high school oriented. So, I like to think that we’re strategy-based and that we pay attention to the trends, but at the end of the day, you still want to draft the guy who is going to have the best major-league career.
DL: To what extent do you analyze draft data and trends?
LW: I believe that we were ahead of the curve back when I started, because everybody was focused in on the individual statistics of the players. At the major-league level that is certainly very valuable, and even in the minor leagues it’s going to show you trends, but I think it is less applicable in the amateur draft. What we focused on was the use of statistical data and analysis of what the draft has produced. One of the first things I did when I came here was look at the history of the Dodgers drafts. I looked at how many players they produced and I broke it down by the decade. They produced roughly 30 per decade up until the 1990s and then it was about 12. I found out by looking at that that the key to success—at least getting to the World Series and having championship teams—is that you need to produce about 30 major-league players. And when I say 30, what I mean is that they need to have five years of major-league service time or better. Guys who just a got a cup of coffee don’t count.
We also look at where we’re picking in the draft. Every year we’re picking late. A lot of teams have a pretty good idea of what they’re doing out there, so what I found by researching past data is that if you’re picking 26th, a lot of time the quality college players have been taken. I’m not saying there isn’t an occasional guy that is there, or an exception to the rule, but as a history you had Jimmy Rollins, Scott Rolen—all of these guys were second-round picks, whereas if you say, "I’m going to get Barry Zito" or "I’m going to get Mark Mulder," well, those guys were taken by the time we picked. Prince Fielder, Zach Greinke—you can go on and on. These guys were taken fifth, eighth, ninth, and in slots like that, so what I did was pay attention specifically to what the draft has produced where we’re usually picking, and I let that trend lead us.
Another thing I did was take trends, from the entire history of the draft, breaking it down as to how many players made it to the big leagues versus how many signed out of each area. So, I knew how many players came from which state, and which region of the country, and I also knew the success rate that each position had. Not surprisingly, left-handed college pitching had the highest success rate—by a little bit of a margin, not a lot. That’s the type of stuff that we utilize. I like to think that we use statistics and probabilities an awful lot.
DL: Is there anything else that maybe differentiates you?
LW: A lot of teams get caught up in making sure that they turn in every player that is going to get drafted. If you do that, you have 1,500 players. What I want our scouts to focus on is, “Let’s really know the 600 that we’re going to turn in.” I don’t want every player; I want them to be more selective and more discerning and be able to eliminate guys who aren’t going to be major-league players. If we do a better job of scouting we should be able to pick the roughly 150 that the draft will usually produce—hat’s from cup-of-coffee guys to all-stars to whatever. I figure that if your scouts are good enough to analyze and break it down to 600—and maybe they miss two dozen cup-of-coffee guys—it is easier to figure 150 out of 600 than 150 out of 1,500. Those are things we look at.
Another thing we do—and we’re probably one of the few teams that do it—is that we work the draft board backwards. In all my years of being in draft rooms, we’d go in and we’re already tired because we’ve been traveling all spring, and we’d rank the top players right away. We’d rank our top 60, we’d rank our top 100, and whatever. But then—my experience has been—later on, guys would kind of get put on the board randomly at times. I felt that if we go in and rank the board backwards, starting with the senior-sign guys and down-the-line guys, and try to pick guys out of that group and work our board backwards, we’d have more success in the later rounds. Closer to draft day is when we actually put our rank board together. We already have a pretty good idea of who our first and second round guys are, or who we are focused in on. We’ve taken a number of kids down the line who have played in the major leagues. We took Russell Martin in the 17th round, Eric Stults 15th, James McDonald 11th—you can see our track record on that. We got Travis Denker around the 20th round.
DL: When do names first begin going up on your board?
LW: Our guys turn in reports, and we have all the lists in our computer, but we don’t begin putting them on the board until around the end of March. Now, I don’t pay any attention to it until we get in there for the draft and really start working it. One of the things that we’re trying to do this year, and I don’t know how it will work because we’re doing it differently than we ever have in the past, is that we had a meeting here in February with our cross checkers and supervisors, and we went over all of our scouts top-12 players—who they had going into the spring. We watched video film of them, we talked about them—what we need to know more about them, what we need to see more of, if there is someone we need to eliminate already. Things like that. I think it may actually give us a better direction to start the season.
The fact that video is better than it ever used to be is a help, although that’s just a tool, like anything. You certainly can’t draft on it. I’ve tried that and made my mistakes in the past, drafting off of video. It’s a tool, just like the radar gun. You can’t draft off of it, but you certainly want to utilize it.
Another thing I do is keep a private log of certain types of arm actions – -the success rates of them. Certain types of deliveries—their success and failure rates. The same with hitters. There are certain things that we will either like or stay away from based on our own statistics of how those have been working over the past 10 or 15 years. I’ve kept these since I was an area scout. Let’s say for example that a guy is a slinger or he has a bad wrist wrap. How many guys have that who have been drafted and signed, that I’ve seen, and have actually made it? And how far? Things like that. I’ve kept pretty good records and I haven’t publicized them, not even to my own staff, but I do utilize that kind of stuff.
DL: Did you use that type of data when assessing Chris Withrow?
LW: Absolutely. And for us, the thing with Chris is that in terms of athleticism he was off the charts. In terms of arm action and delivery, he was at the top of the chart with everything we see. Obviously, the physical stuff is there. He was 90-94 in high school and he’s throwing harder than that now. He’s got real good rotation on the breaking ball. He has all of those things and he has a father who played at the University of Texas, and minor-league baseball, so he has some history with professional baseball in his background. He scored really well on all of the things that we look at, even the psychological matrix that we use. And I want to make sure that this is understood: You can have all of those factors working, saying that someone is really good, or even off the chart, and they still fail. There are two things that happen to pitchers, and I talk to our pitchers in camp here—there are two things that keep pitchers with major-league stuff from making it. That’s injury and choices. More specifically, bad choices, whether it’s off-field, work habits, whatever. If they have the stuff, it’s usually injury or choices, so that’s one of our outlooks.
Pitching is the one position that you draft where guys are going to get taken out through no control of your own. By that I mean the injury factor. With position players you don’t really have that come up. It comes up occasionally on a knee, or something like that, but it’s pretty rare that a position player gets taken out by an injury.
DL: To what extent can you quantify injury risk in young pitchers?
LW: I’ve done studies on it and you have to look at what I call the two-four rule. I’ve told people this in the past, and basically what I’m saying is that 50 percent of the time you’re going to have an injury of some form or another where they have to have a surgery. Then you have the risk of how many of those 50 percent that break down are going to come back. About 20 or 30 percent aren’t going to come back to the same level, so you have to factor in that maybe 30 percent of your pitchers are going to get wiped out by an injury that you have no control over. That is one reason that we draft a lot of pitching and why you have to draft a lot of pitching.
Maybe it’s a bad analogy, but say you’re looking at automobiles and you’re looking at a top-of-the-line Mercedes Benz. It has the best engineering and everybody says that the aerodynamics are great and it has the smoothest ride. It’s the top car on the market. Even that car might sometimes be a lemon. You’re going to end up with some lemons, but if you end up buying that car over a Kia, or something like that, you’re probably going to have more longer-term success than you would with the Kia. It’s kind of like that with pitching. You know you’re going to have some factors, but you still have to work off the ideal.
DL: What constitutes an ideal?
LW: We have what we consider the ideal pitcher and the ideal hitter, and we work off of ideals. We haven’t found a perfect pitcher yet, nor a perfect hitter, but that’s what we work off of. We try to get that person. We’re looking for Roger Clemens; we’re looking for Pedro Martinez; we’re looking for Curt Schilling—guys who had those types of deliveries or arm actions that allowed them to be successful for a long time. Nolan Ryan. And then we have a similar thing we look for with hitters. We try to teach our scouts what to look for—here is the ideal; here are some of the key points we’re looking for.
I think, too, that one of our biggest successes is that while we haven’t given psychological tests, we do have a psychological matrix that I’ve come up with over my years of scouting. I got a minor in psychology, in college, and I utilize that. There are certain things we ask the scouts to find out about the players that we have to know. Examples are GPA and SAT scores. We find out if their parents are married or not; we find out their birth order. We find out all of that stuff and put it together in a matrix, and it gives me a composite score that tells me an order of makeup. When we go into a draft, we know where guys rank from a makeup standpoint.
DL: Do you use standardized tests or only your own?
LW: We don’t do the standardized testing at all. I don’t want to disparage all psychological tests, but I’ve had too many in the past, with teams that I’ve worked for, that have been totally wrong. To me, it was a lot like reading a horoscope—it was hit and miss and too generic. I never had one rank the players, whereas this one is different. It’s based on a lot of different factors that I don’t think they look at. It has a lot to do with past history and not just the player—it could be his family and things like that. Believe it or not, it has been pretty key in helping us make decisions on some players, and I think that’s part of why we’ve had some success. Plus, it helps our scouts realize: "Do you know what? Our butts are on the line." It tells them: "I’ve got to know. I need to know what kind of student this guy is. I need to know if he has a job after school. Has he ever worked in his life? Are his parents together or are they divorced? Maybe one of them has passed away." We want them to find out all of those things.
Going back to what I was saying about 1,500 players, we have 17 area scouts and there are maybe 100 players on their lists. No way can these area scouts get into 100 homes and talk to 100 families, and do a thorough job of it. That’s one of my reasons for getting my guys to really focus in on the talent—who they think are the real players—and let’s really know them. I’d rather have my area scout have 40 guys and really know them, rather than have 100 guys. We’ll take a chance that we might miss a guy, but I’ll promise you that of the 40, he’s going to know them better than the other scouts in his area because of how much data they have to gather and know. It’s a process that I think is important, and it also gives us the ability to sign players because most of the time the parents know us from our talking to them. That’s unless it’s a case where we’re trying to lay low in the weeds and not tip our hand or something.