Tuesday, December 18, 2012

An Interview with Astros Analyst Mike Fast

To call this an interview might be a bit misleading. My past interviews mostly consist of my asking questions and then shutting up and listening to the interviewee. When I talked to Mike Fast, former writer for Baseball Prospectus and Hardball Times, and current member of the Astros Decision Sciences brain trust, I'll admit it was more of a discussion than an interview as I interjected my thoughts and ideas into the conversation ... often.

I was granted this interview on the condition that I wouldn't try to ferret out any Super Secret Decision Science Proprietary Juju™, but frankly I think it would take the team from Leverage to penetrate Sig's inner sanctum.

But that wasn't my goal, anyway. What I wanted to hear about was using stats to judge minor league talent and the possibilities and pitfalls of so doing. I wanted to talk about exceptions to the "rules" that guide us in looking at minor league players.

What emerged was a very random discussion as we bounced from one topic to the next, but nevertheless a constant theme emerged. The following excerpts from our discussion were edited for brevity and clarity.

First we talked about relying too heavily on stats in general in judging minor league prospects without taking context into consideration. According to Fast, "Definitely there are pitfalls at all levels but more so in the minor leagues. Even in the major leagues, you have to look at parks. For example, Coors Field is sort of the one that everyone knows about. And that can be [a factor] in the minor leagues [as well], like our Lancaster park. I think most of our Astros prospect [followers] are aware that Lancaster’s a big hitter's park and so you have to give the pitchers the benefit of a doubt there and sort of take the hitter’s stats with a grain of salt.  The Tri-City park can be a little bit that way. And then even at the league level – the Pacific Coast league is more of a hitter’s league than the International league at the AAA level so there are all sorts of league and park contexts that you have to get an idea for."

Fast went on to other considerations that have to be factored in when looking at stats, "And the difficulty of the league itself. The age or experience of a player relative to the league. So if you’ve got a college player coming in to the Gulf Coast League or the Appy League, he’s more advanced than most of the players that he’s going to be playing against so he ought to do really, really well. He ought to be a man amongst boys really if he’s going to be at those levels. The New York-Penn league is more typically where college players come in after the draft. Nolan Fontana … he went to (LoA) Lexington. The fact that he was sort of able to hold his own there is a good sign. He’s playing against more advanced hitters who are in their second or third season of pro ball. That’s a level higher than what players typically go to out of the draft. And then the same thing, of course, applies generally in terms of looking at the age of a player relative to the level they’re in or if they’re repeating a level. If they’re old for a level or if they’re repeating a level, you sort of expect them to do well. Foltynewicz repeated this year at Lexington and he did really, really well and that’s a good sign. Same thing with DeShields. He came out and was sort of in a league where he was competing against tougher competition than he was used to and then when he repeated the level, he really stepped it up and that’s what you want to see. But if you have a player who is repeating and he did OK, he did better than last year but nothing special, that’s not a very good sign."

Always looking for the exceptions to the rule, I asked Mike about a particular starting pitcher who actually fit that description, one who had repeated a level and had a fairly pedestrian season, but was still young enough to merit more time to improve, at least in my opinion. He agreed that this pitcher was still a somewhat interesting prospect and went on to talk about starters vs. relievers as prospects, "One other thing that’s important to take into context with pitchers is that relievers will usually have better numbers than starters. If you take a guy from a starting role and put him into relief, he’s only going through the lineup once, he can throw maybe a mile or two per hour harder and you sort of expect their strikeout numbers to go up when they go into relief. I think it’s easy to look at [certain relievers] in our system, guys that we think are good prospects, but as a group, when you think of future major league value, the guys on the starting side who may be in that same class of prospects just in terms of perception are going to end up providing more major league value. The group of starters is much more likely to create major league value even though they’re not putting up the sexy numbers."

I specifically wanted to know whether or not there were any advanced metrics that he would recommend as something the average minor league follower could use to look at pitching talent. "One thing that I would be careful with, with those advanced pitching metrics … just looking at batted ball outcomes and whether it fell in for a hit or not is sort of a poor way to get an idea over the course of a season of a pitcher’s talent. That’s pretty true at the major league level. It’s less true in the minor leagues. The ability to prevent hits is a pretty good indicator of talent in the minor leagues. The farther you get away from the major leagues the more important that is. One way to think of it is that, at the major league level, pitchers who get there have sort of been screened for their ability to prevent good contact so if you’ve got a guy who gets hit hard, he’s never going to make the majors.  The guys who come to the majors are all at least pretty good at that. And there may be some differences between them but they’re all pretty good at preventing good contact because their stuff moves or they throw hard or they mix their pitches well, know how to locate, whatever. They’ve got some skills in that. And that’s less true particularly in the lower minors. If you have a pitcher in the low minors that just gave up a ton of hits, I don’t know that I’d be so quick to ascribe that 'Oh, he’s got bad luck or he had bad fielders behind him.' That might be true, but it’s also quite likely that the batters are just saying this guy’s stuff is not that hard to deal with."

What statistics would Fast encourage a casual fan to zero in on when looking at a minor league pitcher then? "I think strikeout rate and WHIP would probably be two sort of quick easy ones to look at. Strikeout rate – that’s showing that you’re fooling the batters and that’s hopefully something that would translate to a higher level. There might be reasons why it wouldn’t but, all things being equal, that’s a very good thing. And then  WHIP would be [the pitcher is] keeping base runners off the base paths, whether it’s by not permitting hits or by not walking guys. Those are pretty important."

Again, looking for the exception to the rule, I wondered if a high strikeout rate is crucial in a prospect's ability to succeed. According to Fast, "It’s really tough to make it in the big leagues without striking guys out. If he has other ways to succeed, if he’s not walking guys, he’s keeping the ball on the ground, I think there’s definitely a path to the majors for guys like that. They’re not going to end up being Justin Verlander, but there are definitely guys with that profile that end up with major league value. Jake Buchanan is kind of a guy along those lines. He doesn’t strike out a lot of guys but he doesn’t walk very many and he gets a ton of ground balls."

Does he think that Game Score can tell us anything beyond a quick look at the pitcher's effectiveness in a specific game? "The reason that Game Score is meaningful is because it’s incorporating hits and walks and strikeouts. It’s incorporating those other things that are important so if you add it up over a season, yeah, it’s going to reflect if you have a good pitcher or a bad pitcher, but I don’t know that it’s adding anything on top of that. It’s subject to the same league, level, park context. There’s nothing magic about Game Score. It’s just reflecting those other statistics. It’s a nice number for one game where you can easily see if the guy pitched a good game or not."

After Fast told me that a good strikeout to walk rate for a hitter is something that a casual fan can look at to gauge a hitter's talents, we got into a discussion about high strikeout rates for power hitters. Fast told me, "The higher the level you get, the more you feel they’re demonstrating power and the strikeouts are sort of the price of that. At the lower levels, it’s kind of a concern if pitchers are finding a hole in their swing. I think that’s true at the higher levels too, but if they’re still demonstrating the power you feel like, the pitcher may have found some holes but they can’t get them out all the time. But at the lower level, you worry that they’re sort of jumping on mistakes that they’re not going to see in the big leagues and if they’re striking out a ton, that’s worrisome."

The conventional wisdom is to ignore a pitcher's win-loss record when judging his abilities. Again, I wanted to know if there was anything that we could take away from a good win record. According to Fast, "As an analyst, I don’t really care about it. When we’re trying to pick one pitcher over another, that’s not something that we’re really paying attention to. You’d love to have the bulldog, the competitor, the guy who sort of figures out how to out think the opposition or pace himself or whatever to get the win. I think that that aspect is important but I think that’s tough to measure by looking at a win-loss record because there’s so many other things that go into that, [but] we’re definitely getting input from the coaches and the player development staff about their personal aspects – their work ethic, their gamesmanship and all that."

One basic theme had emerged by this point. It is something that I have come to understand clearly over time as I have studied the Astros minor league system and I articulated that idea at this point, "To me, you have to understand the whole player." Fast responded, "Yeah, definitely. This is true in the minor leagues but it’s true in the major leagues too. I keep coming back to the term work ethic but it comprises something bigger than that. Players, when players first get drafted or they first get signed in the Dominican, they are nowhere close to good enough to make it in the major leagues. And so that whole process of improvement – their personal character and their effort and all of that plays a big part in getting to the majors and also the teamwork aspect of that, coachability or however you want to say that. Being able to interact well with others and gain insight from coaches and from other teammates, being able to share what you learned with other teammates – that [plays] a big part in getting guys from where they’re entering the system to where they’re ready to be in the majors. And I think even once they’re in the majors, getting from the rookies … to really being an accomplished player, there are aspects of that which don’t show up in the stats but they’re still incredibly important."

You heard him right. For those of you who were worried that stats would become the sole guiding factor in looking at prospects under Sig Mejdal's esoterically named Decision Sciences group, Fast's statement should  serve to reassure you. He went on to tell me, "It’s been eye-opening for me being on the inside – the human piece of the business. Like when we’re making decisions on trades or promotions or releasing guys, whatever. This is somebody’s life and somebody’s career that we’re dealing with. And you get around some of these guys and they’re neat people and you cheer for them, you want them to succeed."

I asked Mike if he had any final words for my readers about what the Astros analytics group is all about. He answered, "I think most of your readers are probably aware, obviously we’re pulling scouting reports and player development reports together with the stats when we’re making decisions. Hopefully they recognize that. That’s something I think the broader baseball community isn't always aware [of]. They sort of think of the guys in their mother’s basement with their spreadsheets, picking out some guy with good numbers and sort of ignoring all the rest of it. It’s definitely an integrated operation where we’re constantly talking to our player development guys and our scouts and they’re talking to us. We’re sharing information and working together."

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

There are many "rules" in evaluating prospects, but there are always exceptions to those rules. I suppose it's partly my tendency to root for the underdog that keeps me looking for those exceptions, but it's also based on looking at the total player – his stats, his trends, his personality, his work ethic and his interaction with other players – that keep me interested in the potential of players who may not be considered top prospects.

Interestingly enough, I often come to the same conclusions about a player as those who are privy to the Super Secret Decision Science Proprietary Juju™. Perhaps that is because, thankfully, they are also interested in looking at the entirety of a player. That is a good thing and very, very nice to know.

Thank you, Mike, for your time and thank you for always being happy to answer even the silliest of my questions along the way.

1 comment:

  1. Nice interview, Jayne. It provides some good insight to what the team is attempting. It is interesting to read that Fast has had his eye opened that they are dealing with people's lives. I heard Goldstein say something along the same lines. Players are people, imagine that!

    I wonder if this same dynamic has taken place with the marketing side of the organization? Often their approach makes it look like they think they can just plug numbers(or words) into formulas and the fans will show up like robots and be invested in the venture.

    ReplyDelete