Draft Q&A with Keith Law
April 30, 2010
For those of you who don't know him, Keith Law joined ESPN.com in 2006 as the lead baseball analyst for Scou... Wait, if you don't already know who Keith Law is, we're not really sure what to say. How did you even get here?
Keith is a very good guy, a card-carrying Friend of College Splits, and always an interesting read. We thank him for taking some time to chat.
College Splits: When will a team's initial draft planning meetings generally start? How many players are in originally in the mix? How many scouting reports get turned in, how quickly does that number start getting whittled down? How many will make it onto a team's draft board?
Keith Law: I assume you mean the predraft meetings where they set up the board. Some start in late May, then break and regroup right before the draft, but I think the most common setup is to start ten to twelve days before the draft. In Toronto, we started with the director, the cross-checkers, Lacava, and myself, for maybe 5-6 days, covering all the top guys who'd been seen by at least one person in the room, then the area guys came in and we'd do the rest of the class. I'm sure it's different there now, because they'd have to hold the meetings out on the field to fit all those area scouts in there.
CS: How ironclad are teams' draft budgets? Is it something generally set in the fall before as part of the regular budgeting process? At what point does ownership get involved if a player falls who proves to be just too good to pass up, but perhaps the budget won't allow it?
KL: Depends on ownership. Corporate parents may want that set 6-9 months in advance. Otherwise it's usually set before the spring. And ultimately a good owner will let a GM shift money during the season - save a million bucks at the big league level through a trade, plow it back into the draft by signing a "tough sign" kid in August.
CS: The MLB Network's coverage has changed this to some extent, but the draft moves along at an incredibly fast pace. How realistic is it for teams to just "let the draft come to them" and move around their draft boards accordingly? What does it take for a Scouting Director to change their approach from round-to-round?
KL: You've spent 10-12 days in a room setting up the board. Making those battlefield decisions is pretty dangerous. If there's a run on catchers or left-handed pitchers, do you really want to overdraft some catcher or LHP you had ranked as an 8th rounder, just because it's the 4th round and those guys are flying off the board? Seems like a good way to foul up your draft.
CS: Are there any teams who seek to formalize the way they diversify the risk amongst their pool of picks? Is there anyone whose job it is to sit back and say..."hold up guys, we've taken a ton of toolsy high school guys, isn't it about time we look for a few "safer/low ceiling-high floor" college guys?"
KL: That's the director's job - but not every team cares. That was part of why I was generally a fan of McLeod's drafts in Boston - there was a clear strategy of portfolio management within each draft, and from draft to draft. That doesn't mean that, say, Philly's way - taking lots of toolsy, high-risk, high-upside high school kids - is wrong, but it's not how I'd do it.
CS: We've heard from a number of agents that they get their best leads from scouts. How does that relationship work?
KL: Many agents were once scouts, many are just friends with scouts, and sometimes a scout might steer a player to a specific agent in the hope that down the road the player/agent will be more inclined to do a deal with that scout's team. All of these amateur kids need agents, no matter what the NCAA says. It's the biggest financial or legal transaction of their lives. They should have counsel.
CS: How do you balance your own eye vs. your statistical background vs. what you hear from those in the industry -- not just in your recent Top 100, but also in your overall approach?
KL: At the end of the day, I try to rely on my own eye as much as possible, and use the other two factors as reality-checks or to slide players' grades up or down. If I see a player, especially if I get multiple looks, and I really like what I see, I stick with that. It becomes a learning experience - I'm not going to learn anything from rolling with other people's hunches.
I will say that that formula varies in pro ball. The closer you get to the big leagues, the more relevance the statistics have. That's why you're seeing teams like Toronto add staff on the amateur side, where scouting is most critical, and you want as many looks as you can get.
Of course, I do use the data you guys share with me; I've found L/R splits extremely useful, as well as home/road splits for players in noticeably skewed parks, and occasionally ground/air ratios if I see a swing issue or a fastball that appears to have good sink or tail. If I was a scouting director, I'd want that data. Nothing more embarrassing than taking a LH hitter only to find out he couldn't hit a left-hander if the kid was throwing 75 mph and announced his pitches before throwing them.
CS: You've identified the Red Sox' recent drafts as some of the best in all of baseball. How might that change with [former Red Sox Scouting Director] Jason McLeod's departure?
KL: They lost a great evaluator when Jason left, but they promoted his assistant, Amiel Sawdaye, to keep the overall system going, with the staff largely intact from last year. I don't expect to see significant changes in how they draft.
CS: More generally, how common is their for any front office to have a "Red Sox [or any team] Way" of approaching the draft that goes beyond the Scouting Director? How much control and/or direction comes from above their pay grade?
KL: That's up to the GM - some are very involved, some set a philosophy and tell the director to implement it, some hire the director and tell him "go get me some players." I tend to favor the middle approach; if I was a GM, I'd probably want to see our candidates for the first pick, but ultimately a GM won't know the country well enough to override the staff's pref list and choice for the top pick. I've told the story many times of how that could go wrong.
CS: How difficult is it to scout for potential pitcher injury risk? Is it ultimately a fool's errand, or are there definitive things in a pitcher's mechanics where the inherent risk impacts your overall talent evaluation?
KL: It's incredibly difficult, and there are very few people in the industry - that is, people who do this full-time, are paid to do it, are even trained to do it - who do it well; the ability to animate a GIF is not exactly correlated to the ability to identify a flawed delivery. There are certain really obvious red flags, but I always come back to B.J. Ryan: His arm action seemed destined - even designed - to blow out, and he did blow out, but by the time he blew out he'd already provided six years of service, several of them quite valuable. So if you'd said at the time he was drafted that Ryan was going to blow out because of his arm action, you would have been right, and yet passing on him for that reason would not, in the end, have been justified.
And just to be clear, I don't think I'm any kind of expert on mechanics or injuries, and while I often discuss mechanics I seldom tie that to predictions of doom.
CS: You were once a baseball interweb Young Turk, then had a stint in the game. Now you're literally and figuratively an Insider. What's your perspective on the extent to which front offices seek to integrate knowledge from outside of the industry?
KL: That varies widely from front office to front office, but the most progressive ones are trying to keep up with everything, much as, say, a good econ professor is reading industry journals and research papers to stay up on the latest thinking. That's a funny offshoot of this absurd RBI-centric argument we saw all this week regarding Ryan Howard: So, you really think that the way baseball players were evaluated statistically 50 or 60 or 80 years ago was the the be-all and end-all, and there's been no progress since then? Really?
CS: A final line of questions on your recent Top 100. Who has the greatest chance of rocketing up your next rankings, and who has the greatest chance to fall? Who's that one guy when after you gave him a number you thought to yourself..."Oh boy, people on twitter are going to go crazy about this one?"
KL: I may have had Rojas 10-15 spots too high, although with guys just now racing in to see him, it may be a question of insufficient data (scouting data, that is) and my own lack of experience with the player. Dave Filak from Oneonta could make a major move up if he throws well in his last 2-3 outings; with their season winding down earlier than those of D1 schools, they'll see a ton of cross-checkers every time he throws because it could be their last look at him. And Paxton ... who the heck knows. He hasn't pitched this year, and he'll pitch in the controlled environment of independent ball, where Luke Hochevar hit 97! and everyone got all excited. I do like Paxton, but this is not the ideal way for a team to evaluate him.