Should We Adjust How We Evaluate Pitching Prospects?

Evaluating pitchers is a real challenge. A combination of experience and knowledge can help one to better understand how variables like velocity, spin, and pitch mix translate to the majors. Even with that information, though, the influence of other factors — like injury risk, like a pitcher’s likelihood of responding to mechanical or mental adjustments — creates a great deal of uncertainty.

Nor is this a challenge that faces only prospect analysts like myself and Eric Longenhagen: even front-office execs who have the benefit of substantial resources — in the form both of data and personnel — have trouble reliably projecting outcomes for otherwise similarly talented young arms.

In my role as a talent evaluator both with FanGraphs and with a few major-league clubs, the question of how best to assess pitchers is obviously one to which I’ve returned with some frequency. In my recent efforts to get some final looks at certain top pitching prospects, however, I began to rethink how Eric Longenhagen and I should approach rankings this offseason. Three prospects, in particular, help to illustrate my concerns.

Tigers righty Matt Manning was the ninth overall pick in 2016, is an athletic 20-year-old who stands 6-foot-6, and was promoted to Double-A last week. In addition to that, he sat 94-96 and hit 98 mph in my look, mixing in a spike curveball that flashed 65 on the 20-80 scale. The positives here are numerous, and very few other minor leaguers could match even a few of these qualities.

The issue with Manning is the same as it was in high school and earlier this season: the consistency of his stuff, the consistency of his command, and the quality of his changeup. At 86-88 mph, Manning’s changeup was firm and lacked life, grading fringe-average just once or twice, often out of the zone. Manning works up in the zone with his four-seam fastball, as the pitch and its plane dictate he should, but the pitch is straight and his command of it is below average. When Manning is having trouble throwing an offspeed pitch for a strike, hitters can sit on the fastball and will have a good chance of getting a straight, elevated strike. This very weakness led to two homers the night I saw him.

His curveball got more consistent as the night wore on, a somewhat common development for young power pitchers who gain more feel when the nerves wear off and fatigue prevents them from overthrowing, but his changeup didn’t improve. Manning is a great athlete for his size. Because of that, his shortest path to a big-league starting role is probably to address his command issues by adopting a delivery that allows his athleticism to shine through, aids the repeatability of his mechanics, and allows him to throw more strikes. This could help everything play up a bit and possibly even add movement to his changeup.

As it is now, it appears as though Manning is over-striding. In concert with a slight cross-body delivery, the result is a complication of what could otherwise be a simple motion. In its current form, Manning’s delivery creates some head action at release, among other things. It also makes the act of releasing the ball more physically stressful and thus harder to replicate, which leads to the command issues.

While his numbers have been pretty good this year, Manning belongs to that class of pitcher who will likely have issues with Triple-A and MLB hitters — hitters, in other words, who can catch up to mid-90s velocity and will wait for a strike on an offspeed pitch. In the lower minors, only some hitters have the skill and discipline to do that.

There’s a lot of similarities with Rays righty Tyler Glasnow at the same stage, including the size, the stuff, the extremely long stride, and a profile that may not lead to a challenge until the upper minors. Glasnow took a little longer than expected to work these things out, and I fear Manning may have the same issues until he improves his consistency.

White Sox righty Dylan Cease has a profile similar to Manning’s, with comparable raw stuff. Cease held 95-97, touching 98 mph, into the fifth inning of the start I attended. His mid-70s curve was above average to plus all night, but his control/command made it difficult to project him comfortably as a starter.

As with Manning, Cease’s heater has hop up in the zone. Cease’s plays even better in games, though, because of the slow-to-fast tempo in his delivery that makes the velocity that much more jarring to the hitter. As was true with Manning, Cease’s curve was also inconsistent, but his changeup flashed average a few times and had a good velo differential at 80-82 mph. When considered all together, his current arsenal is more typical of a big-league starter.

The key point with Cease, for me, is that he lacks the feel to start. In Manning’s case, he’s still young and tall (big frames take the longest to develop command) and he has something clear to improve in his delivery, so it isn’t hard to imagine improvement. In Cease’s case, though, he’s two years older, lacks real height, has already endured Tommy John surgery, has trouble holding runners (or did in my look), and has failed to develop much since his underclassman days in high school. Also, there isn’t a clear flaw in his delivery that, if addressed, could unlock more feel.

Cease’s delivery is fine, but very few big-league starters have this kind of size, health history, and delivery tempo. He’s been showing us what he’s good at for a while (swing-and-miss fastball, chase-pitch curve, usable changeup) and that’s useful in the current game more than ever; it probably just won’t be as a 200-inning starting pitcher. Cease actually fits well in the Rays’ mold of the second pitcher after the opener — and, to be fair, he’s made strides this year in terms of durability and figuring out how to succeed with his skillset. Sometimes these types of prospects discover feel later in their careers, but there are enough data points here to point to a hybrid role as a best fit.

Luckily, the reports we’ve been getting this year generally cite the same strengths and weakness I saw during my looks at both pitchers — and further support their rankings at the site. Some scouts prefer Cease and think he will be a starter. The two are pretty similar in broad terms, but Manning has the advantage in age, size, and health history — while also having a slightly better chance to start, in my opinion — but they’re still very close.

You can see why I group Cease and Manning together. You can probably see, just from the video, that Brendan McKay is almost the opposite of these two:

McKay was a two-way player in college at Louisville before going fourth overall last summer. He has been gaining momentum this year as a pitcher who acts as a DH in the big leagues, similar to the way Shohei Ohtani has been used for the Angels. McKay has slimmed up his frame and found a little more velocity this year — sitting 93-95 and hitting 96 mph in my look — but doesn’t rival either Cease or Manning in terms of arm speed. While he didn’t throw any curves to batters during my look, he still went to the pitch during his warm-up period between innings. It’s been his best offspeed pitch (flashing 60), so it appears he was just working on other pitches in the outing. McKay threw an 86-91 mph cutter and an 84-87 mph changeup, both of which project above average — as does his command.

Before the draft last summer, more than a few scouts thought McKay was a primary hitter, projecting 60 hit and game-power tools. This hitting ability gives McKay both a unique backup plan (if he fails at pitching or gets hurt) and a concurrent plan (DH when he isn’t pitching). That seems less important now than it was last summer, as it’s appearing more likely that McKay’s mix of above-average stuff and command will be his primary meal ticket at the big-league level.

***

My interest in these three specific pitchers relates to my work on last winter’s top-100 list with Eric. He and I purposely moved pitchers, as a population, down our rankings a bit to better reflect how many wins pitchers create relative to position players. Traditionally, there appears to be too little consideration in these lists for pitcher attrition. (This is the case for our own old lists, as well.) It’s easy to downgrade the Cease/Manning types due to questions about their future roles, but it’s also just as easy to boost them on the grounds that they could be the power-over-feel types who turn into Max Scherzer.

With the increasing emphasis on velocity in today’s game, it seems counterintuitive to start celebrating the merits of command-oriented pitchers — nor am I proposing something as simple as that. I remain reluctant to project much value for legitimate soft-tossers — and there are plenty of data points to support that reluctance. That said, this season has witnessed the rise of some young pitchers who, like Indians RHP Shane Bieber and Braves RHP Bryse Wilson, feature above-average stuff and command — pitchers, that is, more in the mold of Brendan McKay than Cease and Manning (the latter two of whom possess plus stuff but only fringe-average command).

You could argue that the only other pitchers on our lists who might belong to the McKay/Bieber/Wilson class are Brewers RHP Corbin Burnes, Rays RHP Brent Honeywell (if he returns from Tommy John surgery as the same pitcher), Athletics LHP Jesus Luzardo, Indians RHP Triston McKenzie, Tigers RHP Casey Mize, Padres RHP Chris Paddack, and Braves RHP Mike Soroka. Reasonable people could suggest a list even shorter than that.

These pitchers are also mostly ranked higher than the Cease and Manning types anyway, so you could ask if I’m just saying I believe more in highly ranked prospects than lower-ranked ones. My instinct says that the stuff-over-command types make up a disproportionate amount of the washouts on these lists, but we continue to rank them highly so we don’t miss out on hailing the next ace as early as other lists. You could argue, even if we knew that was true, that it was worth an adjustment to our process if we’re less likely to correctly project the next ace that emerges.

Eric and I have said we think the big lists (our own past efforts included) have underrated the MLB-ready, two- to three-win type of player in favor of the upside A-ball prospect. We entered last offseason with this thought — and pitcher-attrition metrics — in mind. It’s hard to say how this hunch will affect any specific pitching prospects in terms of ranking, but it’s definitely something at the front of my mind as we get closer to list season.





Kiley McDaniel has worked as an executive and scout, most recently for the Atlanta Braves, also for the New York Yankees, Baltimore Orioles and Pittsburgh Pirates. He's written for ESPN, Fox Sports and Baseball Prospectus. Follow him on twitter.

58 Comments
Oldest
Newest Most Voted
Inline Feedbacks
View all comments
LHPSU
5 years ago

“My instinct says that the stuff-over-command types make up a disproportionate amount of the washouts on these lists”

I would think that’s baked in by design, no?

ThomServo
5 years ago
Reply to  LHPSU

“My instinct says that the stuff-over-command types make up a disproportionate amount of the washouts on these lists”

A huge shortcoming with this eye-test, list process is the lack of accountability and review.

It doesn’t appear that Kiley/Eric are tracking their projected FVs to test for accuracy. Therefore, as with the whole process, they are taking guesses about how accurate or inaccurate their system is.

A competent system will be transparent, future-proofed, and will provide comprehensive coverage and analysis of the talent-ID method. Reading Eric/Kiley work, it becomes obvious that there isn’t a serious methodology, with back-testing and proofs-of-works to show predictive value and insight. Instead, it’s just guess work.

That’s why the author has to rely on instinct to guess that the guys with high-stuff-low-command grades are failing to live up to FV grades. If the work was being done in even a remotely competent way, the author would be able to make much more certain observations, i.e. “players with a ‘Stuff-Command’ discrepancy of 20 or greater, underachieved FV projections by 38%.”

In reality, the methodology being used is nowhere near serious.