Mid-Tier Hitters I Like
I’ve always struggled to understand the “ceiling” projections that accompany prospects. It’s not that I don’t get the concept — 95th- or 99th-percentile outcomes aren’t the same for everyone, and that’s interesting. And it’s not that I don’t think some prospects have higher ceilings than others. It’s merely that I have a hard time discerning which types of prospects have the greatest chance of being superstars. For every Bryce Harper where the talent smacks you in the face, there’s a Mookie Betts. Fine, it’s not a one-to-one ratio, but plenty of prospects exceed their supposed ceilings, and I’m no good at figuring out which ones are the most likely to do so.
Because of this deficiency, I’ve always looked at prospects slightly differently. I tend to look for players who have a good chance at becoming average regulars, assuming that the best way to find prospects who’ll go completely ham and turn into MVP candidates is to find as many minor leaguers who have the skills to make the big leagues as possible.
This process has led to some successes — like Eric Longenhagen, I was optimistic about Randy Arozarena when the Rays acquired him in 2019. It has also led to some failures — like former FanGraphs editor Carson Cistulli, I’m still waiting for Max Schrock to ascend to batting-title contention. This year, I’ve been working on putting a little more intellectual rigor behind my process, and Eric and Meg Rowley were kind enough to let me share the Prospects Week stage to yammer on about it.
As befits any Ben Clemens project, I used a combination of statistical modeling, careful observation, and semi-rigorous gibberish to synthesize a group of hitters I’m interested in. If you want to ignore the methodology and just get to the sweet, sweet list, I totally understand, but for everyone else, let’s talk about that “statistical modeling” part.
The natural impulse when looking at minor league statistics is to apply some kind of translation. Strike out 15% of the time in Double A? You’ll strike out 23% of the time in the majors. Nine percent walks and a .330 BABIP? That’ll be 6% and .295, thank you very much. Then you add a little age-related improvement dust, and bam, major league prediction. One problem with that method: it’s terrible and doesn’t work.
After a variety of translation-related failures, I settled on a different method. I threw a gigantic pile of minor league statlines into a database and linked them to each player’s major league career (or lack thereof). From there, I asked the same question in a variety of ways: given a player’s statistics, age, position, and level, how likely are they to have, at minimum, a meaningful major league career?
I won’t walk you through the individual niceties of what I did, largely because I used four models without great preference given to any of them. After all, I’m using many models because I’m not overly confident in any of them, and I’m also doing plenty of non-modeled filtering of the lists as well. It’s hardly a scientific process, so pretending it is would both overstate my confidence in it and unfairly impugn the creators of actual statistical minor league models like KATOH.
That said, a rough outline is in order. I looked at everything I could get stats on: batting lines, but also age, level, position, playing time, whatever I could dig up. I stayed away from weight, because I don’t have a lot of confidence that those numbers approximate reality, but everything else was fair game. From there, I used some buzzword-y techniques to make predictions. I tried everything my limited programming skills would let me, with the most time spent on k-nearest-neighbors and multiple binary logistic regressions. I modeled each level separately, because skills that are predictive in Low-A won’t necessarily be as useful at Triple-A, and vice versa. A quick word of acknowledgement here: non-baseball friends Ryan McManus and Curt Harting provided valuable input on this part of the exercise.
There are obvious limitations to these methods. There was no minor league season in 2020, for one. I used 2018 and 2019 statistics (with more weight given to 2019), but that’s an imperfect solution. Still, the output looks at least somewhat believable. My chosen variable is the chance that a player will end up as at least an everyday player during their team control years. The top of the list looks pretty good:
Player | 2019 Level | Major League Likelihood |
---|---|---|
Yordan Alvarez | AAA/MLB | 76.8% |
Wander Franco | A/A+ | 75.4% |
Yes, I know, breaking news, Wander Franco and Yordan Alvarez might just make it. I was happy to see their names at the top, but that’s not what we’re here for. This is just the statistics portion of what I did, but trust me when I say that my manual adjustments didn’t do anything to make Wander Freaking Franco move down the list.
In any case, the same methodology that told me Alvarez and Franco will likely stick in the majors (Isaac Paredes was third and Dylan Carlson fourth, for what it’s worth) also ordered every minor leaguer who played in 2019. Let’s head deeper into the list, past the bright lights at the top, and look for some players with a better chance to stick in the big leagues than might first meet the eye.
First, some honorable mentions.
Gabriel Moreno, a catcher in the Blue Jays system, is our 93rd-ranked prospect. Eric wrote him up recently. Discount his numbers slightly, because he played in one of the most offense-friendly parks in the minors, but his age, position, and ability to avoid strikeouts jump off the page. He also had a great 2020, which this analysis doesn’t touch on.
Tyler Freeman also cracked the Top 100, at 88. Eric also recently covered Freeman. Like Moreno, he’s an up-the-middle defender who makes good contact and is young for his level, though he’s less of an impact defender than scouts first hoped. By the time a player reaches Hi-A, several of my statistical models start to penalize a lack of power more heavily, which is the main worry in Freeman’s game. A discount Nick Madrigal is still a good player, and Freeman can really hit, but the lack of power means there’s less room for the hit tool to decline without impacting his odds of sticking in the majors, a concern that scouts share.
Finally, Ryan Vilade is a different kind of Hi-A shortstop. Like Freeman, he’s sliding down the defensive spectrum — he played some third in 2019 and some outfield in 2020, as Eric covered as part of the 2020 preseason Rockies list. Vilade’s power means that if he can keep the strikeouts reasonable, his bat will play, and he’s showing signs of doing just that.
With that group of touted prospects out of the way, let’s get to the meat of who I want to talk about today: some 40 and 45 FV prospects who I’ve taken an irrational liking to.
Miguel Vargas, 3B, Los Angeles Dodgers
Relevant Stats: Vargas turned 21 last fall, and demolished Midwest League pitching in 2019 (.325/.399/.464) to earn a call-up to Hi-A. He hit a respectable .284/.353/.408 there and only struck out 16.9% of the time in a challenging assignment
What I Like: Vargas might just hit too much to miss the majors. Indeed, he’s done nothing but hit at every minor league stop, even without the home run power you’d like to see out of the 6-foot-3 frame that likely makes him an infield-corner defender. The Dodgers experimented with him at second, but even without that, his 38 doubles in 559 PA in 2019 offer an enticing glimpse at what he could grow into.
Warning Signs: He’s a 6-foot-3 corner infielder who hit seven homers in 559 plate appearances in the low minors. If he ends up as first base only, his power will need to develop significantly. And his Trackman numbers don’t jump off the page, something my models don’t know. Double-A should reveal whether he can maintain his gap-to-gap production and reasonable strikeout rates against advanced pitching.
Eguy Rosario, INF, San Diego Padres
Relevant Stats: Rosario turned 21 last fall. He plays all over the infield, though he fits at second long-term. He showed surprising power for his 5-foot-9 frame, hitting 25 doubles, eight triples, and seven dingers. Rosario also swiped 21 bags — hello, fantasy players.
What I like: Rosario combines a lot of factors I find intriguing without excelling at any. He’s repeatedly played at advanced levels relative to his age, and gotten by with a well-rounded skillset. No single tool jumps off the page, but the total package just works. He makes enough contact to avoid being sunk by his strikeout rate, hits for just enough power to make me hopeful that he has some Jose Altuve in him, and plays enough defense that he doesn’t need his bat to be transcendent to find a major-league role.
Warning Signs: He’s 5-foot-9, doesn’t usually play shortstop, and struck out 20% of the time in Hi-A. I hardly think I need to explain to you why that profile is risky. The raw power here is really a sticking point — my various predictive models don’t know what The Board does: Rosario’s 86 mph average exit velo and 103 mph max don’t portend any hidden potential. The low ceiling on his power means he’ll probably need to develop better than expected in at least one other area to stick in the majors.
José Devers, SS, Miami Marlins
Relevant Stats: Devers turned 21 in December, and he’s a plus defender with plus speed. He makes plenty of contact and has limited strikeouts despite an aggressive approach at levels advanced for his age. He also has one home run in 784 professional plate appearances.
What I like: Devers is an exaggerated version of Eguy Rosario. Better defense, better speed, probably a better hit tool…and even less power. I’m more pessimistic than any of the predictive models I used — they’d rank Devers ahead of Vidal Bruján, just to give one example, and I think that’s nonsense. At the lower levels, power isn’t a prerequisite because players grow into it as they age, but this little power is definitely a red flag — one borne out by Trackman data. I’m intrigued by Devers, but more as a name to remember in case he grows into at least gap-to-gap power.
Warning Signs: Yeah, we’ve covered that part.
Nick Allen, SS, Oakland A’s
Relevant Stats: Allen turned 22 in October. He’s an elite defensive shortstop who can play second and third. Allen hit .292/.363/.434 in Hi-A Stockton in 2019. His slight (5-foot-9) build limits his power, but his bat-to-ball skills look like a plus tool; he struck out only 15.9% of the time in 2019 and 16.6% in 2018.
What I Like:
I know, I know, another slight middle infielder who makes a lot of contact but doesn’t hit for power. This is the last one, I promise! Allen is the best defender of the bunch, by potentially a lot. Allen has such good bat control that I really want to buy into an offensive transformation that could make him a 15-homer type, and it’s certainly a good sign that his 2019 ISO was higher than his ’17 and ’18 ISO’s combined. Even if he can’t take that step, though, he has a rare combination of defense and contact. This is, after all, a list of players I think are likely to have a meaningful major league career, and Allen does so many things well that I can live with a few holes in his game.
Warning Signs:
As with Devers, power could be Allen’s downfall. He doesn’t strike me as quite so worrisome a case, but how he fares in Double-A this year will tell us a lot about him. I’m more worried about any slip in bat control; when you’re built like Allen, pitchers are going to come after you until you burn them for it. Slapped doubles are a fine way to do that, but if some swing-and-miss creeps into his game, the whole thing could fall apart quickly.
Hudson Potts, 3B/1B, Boston Red Sox
Relevant Stats: Potts turned 22 in October. He hit 16 homers as part of a .227/.290/.406 line in Double-A Amarillo in 2019. He played mainly third, with a dash of second thrown in for taste. He maxed out at a 112 mph exit velocity in that campaign, a frankly absurd number for a 20-year-old.
What I Like: It’s so hard to play in Double-A at 20 years old. Potts wasn’t good, but he was able to tread water despite being three to four years young for the level, which is often a better sign than hitting well at an age-appropriate level. That said, don’t sleep on his 2018, when he was also quite young for Hi-A and put together a fearsome power season. Every single model I used doesn’t hold strikeouts against players as much once they reach Double-A, and that mostly tracks with intuition; a Low-A hitter striking out 30% of the time is doomed in a way that a Triple-A slugger who does the same simply isn’t. Potts straddles that line; he’s always had contact issues, even at lower levels. The combination of his power and age are simply more enticing than the whiffs are worrisome.
Warning Signs: The big one is contact — that’s not the kind of thing you can paper over with other skills. Eric has a 40 FV on Potts’ hit tool, and if it turns out lower than that, it might make his bat unplayable. He’ll also need to find a defensive home; he looks like a corner guy, though San Diego experimented with a Mike Moustakas-esque second base assignment before trading him. Corner-only sluggers with contact issues aren’t exactly in short supply, so that’s the worry here.
Omar Estévez, 2B, Los Angeles Dodgers
Relevant Stats: Estévez turns 23 this month, and hit .291/.352/.431 as a 21-year-old in Double-A Tulsa. He split time between shortstop and second base, but is likely a second baseman in the long run. He hits for power (107 max EV) despite his 5-foot-10 frame.
What I Like: This one doesn’t count! Estévez is short, but he packs a punch. He made the transition to Double-A without falling flat on his face, always a worry for someone with intermittent strikeout issues in A-ball. He already elevates and pulls the ball with the best of them, which gives me hope that he’ll continue to hit for power even at higher levels. Subjectively, Estévez feels to me like one of those guys who will just out-hit his physical tools every year; maybe it’s the Cardinals fan in me, but I see some Paul DeJong in his bat, albeit without the plus shortstop defense. Alternatively, think a bargain bin Brandon Lowe from the right side of the plate. Credible middle infielders with pop don’t grow on trees, and Estévez has a chance to be just that.
Warning Signs: One downside of being a sum-is-more-than-the-parts player: if one of the parts breaks, the whole thing does, too. If Estévez made less contact or had slightly worse plate discipline, his power probably wouldn’t be enough to rescue him. If his power doesn’t translate to the bigs, he’s just a slap hitter without a premium glove. If his defense forces him to a corner, the offensive bar goes up. There’s a Goldilocks feeling to Estévez right now, and that will always be a concern.
Kevin Padlo, 3B, Tampa Bay Rays
Relevant Stats: Padlo launched 21 homers in 432 PA between Double- and Triple-A. He’s the three true outcomes personified — he struck out more than 25% of the time and walked more than 15% of the time. Padlo might be able to stick at third base, but might end up in left or at first. He does boast an impressive 92 mph average exit velo and 109 mph max.
What I Like: Padlo is mainly here as a reminder that the skills I’m interested in at the Triple-A level are quite different from what I’m hunting for in the low minors. The closer to the majors (and older) you go, the more you can make a straight port to the major league game today. Padlo strikes out too much, but he absolutely mashes the ball, enough that it’s easy to picture him as a capable big leaguer tomorrow. Doing enough to ensure a major league career doesn’t only mean playing a slick second base and dumping singles into the opposite field power alley. I think that Padlo could probably credibly play in the majors today as a second-division platoon bat, a pretty incredible feat for the 25th-ranked prospect in the Rays’ deep farm system.
Warning Signs: Going peak José Bautista to get your power bump in the minors means you can’t do it again in the majors. Padlo pulls the ball so much already that he doesn’t have much power projection remaining; what you see is mostly what you get. If he doesn’t make enough contact, or if he can’t stick in the field, I’m not sure he has another adjustment to make. Additionally, it’s fair to wonder about the sustainability of his current approach; he looked like a non-prospect after blah showings in 2017 and ’18, so there’s no track record of this approach being sustainable.
***
That’s it for the prospects I want to highlight, though I’ll throw a few more Padlo-esque names at you, Triple-A performers who can put together a major league career without too much need for projection. Ty France would rank highly if he hadn’t graduated, as would Josh VanMeter — middle infield power bats tend to get major league reps. Justin Williams tapped into his power in an injury-marred 2019, but probably needs a swing change to really shine. Yusniel Diaz hasn’t reached Triple-A yet, but he looked like a rising star before injuries slowed his progress; he’d be a marquee prospect if not for his injury history, so I left him off the main list.
Those seven names (plus some honorable mentions) now described, let me tell you why I think this kind of prospect hunting is valuable. It’s certainly not because it’s a superior way of finding stars, or because these types of players are ignored by traditional prospect analysis. It’s also certainly not because teams don’t know what they have on their hands; clubs are doing this kind of data-driven analysis with better data and smarter analysts.
Still, there’s value in looking for these types of players, because it provides an extra perspective. I don’t think this should come anywhere near replacing traditional public-side analysis, but I do think it’s an additional tool to consider — particularly since it’s a given that teams are using more advanced versions of this approach.
It also beats simply lumping every contact-heavy middle infielder into the same boat, or hunting a given profile across every level of the minors. Looking back at what types of profiles actually translated into major league careers is a useful way of distinguishing between a group of similar hitters.
With that said, my expectations for this project aren’t tremendously high. Public-side prospect analysis is far more data-aware than it was five years ago, and many of these players are no doubt being downgraded on our lists with full knowledge of their solid statistical profiles, for reasons that the statistics I have simply don’t pick up on. There’s a reason this list looks a bit like the island of misfit toys; these days, an enticing statistical profile can make you a top 100 prospect if there are no meaningful red flags. Someone like Gabriel Moreno might have flown under the radar back in the day, but analysts and teams have become excellent at spotting and elevating these types of talents.
I plan on refining the tools I used here further, but even before I do that, I’m glad to have a chance to highlight a few interesting names. All of these players have weaknesses, which is why none of them are 50 FV prospects, but at the very least, I think that they’re all interesting in a way you might not see if you merely glanced at a list of names and grades. Even in an age where scouts look at statistical markers in addition to qualitative skills, looking at the problem from an extra angle can’t hurt.
Ben is a writer at FanGraphs. He can be found on Twitter @_Ben_Clemens.
You had me at “semi-rigorous gibberish”, that’s my jam, and encapsulates why I enjoy reading your stuff.
Ben’s “semi-rigorous” stuff is way more rigorous than some peer-reviewed papers.
SEMI RIGOROUS GIBBEISH!!!