I had a moment over my lunch break and thought I'd kick off an interesting discussion we began while recording the TCB Top 30 podcast but were not able to fully explore. During a debate over--I believe--Astros' starting pitcher Akeem Bostick, you pointed out that our methods of evaluating prospects are quite different.
Your background as a professional physical therapist and inclination to translate what you know into a study of pitching mechanics seems naturally have put you on the path of evaluating based on motions, physical tools, and work ethic. In short, scouting. Not to put words into your mouth, I hope you'll explain your perspective in more detail. Whereas I, an engineer and project manager by training and profession, am highly interested in the risk/reward equation, as well as putting professional playing performance, as well as perceived Major League "floor" at the forefront of my evaluations.
A simplistic and aggravating description often heard about people such as myself--and let's be honest, the Astros front office--is that we are spreadsheet junkies that toss aside scouting completely in favor of numbers. That couldn't be further from the truth, but it is fair to say that I weigh the values of scouting reports and athleticism differently depending on how much experience a player has.
For example, in my prospect rankings, if a player has several years of professional experience, I am going to weigh performance and statistical analysis of their peripherals far more heavily than any scouting report, though I will still consider scouts' opinions. For me, the reason is simple: the objective is to play baseball, not to just be a great athlete. Former Astros shortstop Jonathan Villar is a great athlete. But he has seldom been even average for his league level at baseball performance. Former Astros pitchers Mike Foltynewicz and Jarred Cosart light up radar guns. But their results indicate that so far they have not developed as successful pitchers.
So for non-traditional prospects like 1B/DH Tyler White, 3B Colin Moran, SP Joe Musgrove, and others, I value performance way more heavily than scouting because they have had consistent success playing the game of baseball at every level. Meanwhile, for very young players like OFs Daz Cameron and Kyle Tucker, despite that they were fairly disappointing in 2015, I am going to lean heavily on the scouting reports because their professional sample is too small to evaluate any other way.
I have more to say about my method, but first please chime in with your basic approach to ranking prospects.
It's the classic floor vs. ceiling debate. I'm glad our community at TCB has grown over the last several years that it's not a simple subject. This isn't a black and white subject but indeed one of every shade in between. The great part about it is that I feel we are able to create a very reasonable balance. The media has colored the scouting vs. sabr debate as a "for me or against me" type. In fact, if you're fortunate enough to work in baseball or logical enough to follow deeply enough as this community, you know it's quite the opposite. It's a spectrum.
I do tend to lean toward the scouting side some but I'm far from anti-statistics. Like you said, this is a game about performance, this ain't crossfit where you're trying to be good at exercising (sorry I had to throw that in there). You have to produce to move on. However, I think that stats are an outcome of skills and tools. I think that is a very important statement.
Tools are extremely hard to quantify. Skills are even difficult to quantify with a lack of tools. And therein lies the problem with going all in with either side.
Scouting tends to lean towards guys with higher upside because they're required to be good. A guy with average tools is rarely going to be more than average because tools have limited ability to be developed. Skills can sometimes be developed with good coaching. Not always. But sometimes. There's plenty of examples of being un-coachable or an inability to develop them, but there's a bigger chance of that to improve than tools. So, scouting gravitates to high end tools with hopes of developing skills to create a high end player. Doesn't always work. But, you don't find elite players with poor tools and elite skills.
I tend to lean a lot on stats with AA and AAA, so that we will usually agree on.
You said a couple of things right off the bat (yay baseball clichés!) that I like. First, the oblique shot at crossfit junkie Brady Aiken. Second, I love that you called prospect evaluation a spectrum of tools and performance. I was already thinking of making that same point when you mentioned the media's "us or them" narrative.
The spectrum--and I am going to keep using that term, because I like it--for me is even more highly gradated than just a static blend of tools and performance. I throw in the added wrinkle of sample size, which changes whether I weigh stats or scouting more heavily. A better explanation of my approach in prospect evaluation is one you labeled me with during the podcast: probabilities. Obviously, nobody can calculate the exact percentage chance that a minor league prospect reaches stardom in the major leagues, though many have tried to quantify this. But without actual numbers, my evaluation boils down to the questions: 1) "Is there a high probability that this player becomes a major league regular?" and 2) "Is there a comparative high probability that he will become a star if he makes it?" Both of those questions matter greatly to me, so it's more complicated than just saying that I focus more on floor than ceiling. If that were the case, James Hoyt would be among my Top 5 prospects because it's hard to imagine him not being one of the safest bets in the Astros' system to make the majors at some point and stick for a few years.
Back to probabilities and those two questions. For a guy like Colin Moran, who I graded highly: Is there a high probability that he can become a major league regular? In my opinion, yes, that probability is extremely high--perhaps one of the highest batters in the minor leagues. Probability of being a star? That's murkier. Certainly, it's not impossible, and I can squint and see an above-average player. But the balance of those two questions puts Moran solidly in my Top 5 even without the superstar ceiling. Comparable in the opposite sense is Francis Martes. Love the ceiling, but he is far enough away from the majors that his floor is still a fairly wide range. I'd grade him as about equal in value as Moran, with the caveat that by the end of 2016 I may have Martes way ahead because of a larger body of work.
This is important to remember when a perceived "slide-rule mafia" person such as myself is evaluating prospects like Daz Cameron or Kyle Tucker. I mean, wow, their ceilings, right? But that floor is so far away. The ceilings and tools and draft pedigrees are enough to have them easily in my Top 10. But an uncertain path to even major league playing time is questionable enough at this point to keep them out of my Top 3. In fact, if one were grading primarily by floor and performance, neither of those guys would have a prayer of making a Top 30, much less be mentioned among baseball's Top 100 prospects. That spectrum has to be balanced depending on a player's own situation.
Would you please close out this conversation by filling me in on how your knowledge of biomechanics and pitcher motions affect your evaluations of pitchers?
I think our process on evaluations are much closer than they appear on the surface. The word 'probabilities' is extremely important in this entire process and you outline it perfectly. The probability of being a major leaguer versus the probability of being a star is the difference. I think that's where each end of the spectrum has it's flaws.
Too much weight on probability to reach majors=miss on raw stars and hit on lots of role players
Too much weight on probability to be stars=ignores role players and finds more stars
Which is why you correctly have to apply both.
So, you have to take the stats and see if they apply to the tools and vice versa.
For example, a pitcher with a great fastball and nothing else can potentially rack up strikeouts in short-season ball but will likely struggle in AA. So, you have to be skeptical of the stats. But, the opposite is true. There have been examples listed where the stuff is there but the stats don't follow. So, there are skills that are missing.
I'm a pitching guy like you mentioned so I'll put some detail in how I use my background.
With you being an engineer, I think this will make a lot of sense for you since you understand physics.
The body is a kinetic chain that transfers energy up the chain from joint to joint. However, it's a little more complicated than just creating energy as it goes up. You have to look how efficient the body is at transferring the energy from each segment to the next on top of creating new energy at the next segment in order to build off the previous.
That's why you expect a young pitcher to get more velocity as he "fills out" because he'll have more muscle to create energy with. Not always true. What really is more reliable is determining how efficient the delivery is. Someone who hits 90 with a bad delivery can throw harder, if they can create a more efficient arm path that transfers the energy better. If the pitcher is athletic enough you can project for the efficiency to improve and improve velocity. But, that requires understanding of how the body moves.
Obviously success requires more than just velocity. You have have off-speed pitches and of course you have control/command them. Repeatability is somewhat of a lie with mechanics. No two pitches are delivered the same. You won't see the difference but they are. It requires repetition of the delivery to solidify the neurological firing patterns. The more complicated the mechanics, the more difficult it is to solidify that pattern. You'll always have examples of control with poor mechanics and lack of control with great mechanics. That's something that you have to really watch and see how well they control their body during the mechanics. How consistent are they? That's what tells you if the control can potentially improve. But, you have to take into account how well they actually control it, not just the mechanics.
Thanks for the great chat. I always enjoy hearing different viewpoints from my own because it leads to me refining my own ideas and processes. In this case, through these emails and our conversations on the podcast, you've caused me to re-think a couple of my less charitable grades.
I wonder if TCB's readers lean to one viewpoint or another, or if any of them have a wildly different method of evaluation?