As I was lurking on a popular baseball podcast’s Facebook page, as one does, I happened upon a hypothetical three-way trade proposal concocted by one of the groups’ countless members:
Just below was the correct answer to their question:
This clearly would be a deeply ill-advised trade for the Cleveland Indians. It’s uncertain, or perhaps unlikely, whether the original poster considered this proposal for more than 30 seconds before hitting enter. Certainly, it’s not worth my time or energy. And yet, I couldn’t stop wondering: On what grounds was this proposal considered fair or good by its conjecturer?
My guess is that this hypothetical was created not to say anything about Mike Clevinger, but instead to highlight the heaping pile of prospect capital required to move Albert Pujols and his eyesore of a contract.1 Our original poster suggests that Uber-Prospect Jo Adell and ultra-athletic Jordyn Adams would be required in order to counterbalance Pujols and receive useful players in return. As the purpose of the post, this is likely where much of his mental energy was expended. He knew of Clevinger’s rumored availability and the Angels’ organizational pitching starvation. This, I believe, was the purpose of the theorized trade, and however well-trodden that topic is, the original poster did make the point successfully.
What made me so curious about this trade offer is that, from a value perspective, it’s horribly unbalanced. Common thought dictates that arbitration is when players begin “getting paid”, so it’s worth a reminder just how team-friendly arbitration remains. For a moment, let’s assume* that the Steamer projections on FanGraphs are in fact predictive in each of the following cases. Clevinger, projected 4.3 fWAR for the upcoming season, signed a $4.1M contract for 2020, his A1 year. To project his future earnings, Noah Syndergaard and Trevor Bauer, each projected within .1 WAR of Clevinger, signed for $9.7M and $17.5M for their A2 and A3 years respectively. If Clevinger is truly a 4-4.5 win pitcher, let’s assume his salaries will be somewhat similar to Thor’s and Bauer’s in ’21 and ’22 (inflation is negligible). In this hypothetical, the Indians have a 13 WAR signed to what amounts to a three-year, $31.3M contract. For comparison, Rick Porcello, projected 1.9 WAR in 2020 and two years older than Clevinger, signed a three-year, $36M contract this past offseason. Meanwhile, Stephen Strasburg, and his seven-year, 234-million-dollar contract, is the free-agent pitcher with the closest 2020 projection to Clevinger. The point is, cultivating pitchers as good as Clevinger is difficult (for most teams) and extremely costly to acquire. Unless someone drops a tractor load of MLB-ready prospects on Chris Antonetti’s doorstep, Clevinger isn’t going anywhere.
*If there’s one thing my political science training taught me, it’s that assumptions matter. In the paragraph above, which laid out, imprecisely, how valuable Clevinger’s employment is projected to be, I assumed the accuracy of the projections. In general, it’s a good idea to trust the projections. For better or worse, they’re unbiased, using only historical context and certain stats to predict future production, which is usually regression to the mean. They hit the mark much more often than they miss, which makes sense: the best individual predictor of the future is the past. However, sometimes there’s reason to believe the status quo of the past has shifted, thereby making the mean to which a player performance was to regress obsolete. And, with player self-improvement becoming more and more common, one must always be on the lookout for those who change something about their game. Reader, you know where I’m headed. Clevinger got better at pitching in a way the projections do not account for.
Here are Clevinger’s career stats:
|2016||0.2||53||5.26||4.86||21.5 %||103||12.5 %||157||29.8 %||78.6 %||9.3 %|
|2017||2.1||121.2||3.11||3.85||27.3 %||126||12.0 %||143||27.2 %||70.5 %||12.5 %|
|2018||4.2||200||3.02||3.52||25.6 %||116||8.3 %||100||30.4 %||74.1 %||12.0 %|
|2019||4.5||126||2.71||2.49||33.9 %||149||7.4 %||87||31.5 %||66.7 %||15.2 %|
A quick scan across the table portrays a huge leap forward in 2019. The strikeouts shot up, the walks plunged, hitters chased more frequently while making contact less often. Now, projections assume that 2018 and 2019 Clevinger were the same pitchers, and therefore his “true skill” level lies somewhere in between the two years. Like our trade proposer, the projections have underestimated the Cleveland ace.
The offseason between 2018 and 2019, Clevinger had prioritized strengthening his fastball, as batters had always seen the pitch well. The 131 wRC+ batters managed off his heater in 2018 was actually a career-low to that point. Thankfully, his winter toils proved successful. Now, a list of the fastest pitches from Clevinger’s career starts with his first career 99 MPH fastball, which he delivered in June 2019, and includes 98 pitches from 2019 in its top 100. But, it wasn’t just his maximum speed that skyrocketed. In 2018, his average fastball left his hand at 93.6 MPH; in 2019, that velocity flared up to 95.4. Put another way, from his rookie year through 2018, just 12.1% of Clevinger’s fastballs clocked in at 95+ MPH; in 2019, 63.4% of his fastballs traveled faster than 95 MPH.
But, Clevinger’s fastball changed in more than simple velocity—he also drastically altered his fastball location. Below, you’ll find two strike-zone diagrams with percentages that represent the number of fastballs in each location as a fraction of the total number of fastballs. On the left is 2017-2018; on the right is 2019.
As you can see, last season, Clevinger focused his fastballs higher in the strike zone or above it all together, compared to the two years prior, during which his fastballs were much more evenly distributed. This adjustment mirrors a league-wide trend. In early 2017, FanGraphs writer emeritus Jeff Sullivan wrote a post entitled “In Search of the High Fastballs” in which he reasoned that hitters had increased their aptitudes for hitting the low fastball, and was therefore perplexed to find that pitchers hadn’t more widely adopted high fastballs (Sullivan defined a high fastball as greater than 2.5 feet above the ground; I’ll do the same). As was so often the case, Sullivan was onto something here. Below is a modernization of a graph from that article, with an extra 2.5-ish years of data:
The league did exactly what Sullivan said it should have done; last season saw the highest percentage of high fastballs in the pitch-tracking era. Clevinger simply did what many pitchers are doing. Before 2019, Clevinger threw 42.6% of his fastballs at least 2.5 feet off the ground; in 2019, that number shot up to 56%. Between its increased velocity and refined location, 2019 was by far the most productive year for Clevinger’s fastball. The stats in the table below include only the fastball.
|Year||Average Velo||>95 MPH %||Average Height||>2.5 ft. %||Batting Average||wRC+||SwStr%|
|2016*||94 MPH||20.8 %||2.47 ft||43.5 %||0.279||159||6.3 %|
|2017||92.5 MPH||1.4 %||2.33 ft||37.2 %||0.256||138||4.8 %|
|2018||93.6 MPH||16.4 %||2.52 ft||46.2 %||0.263||131||7.7 %|
|2019||95.4 MPH||63.4 %||2.73 ft||56.0 %||0.192||75||12.7 %|
*His rookie year, during which he threw 53 innings. Let’s just say the season was a learning experience for Clevinger.
Again, until last season, hitters used to prey on Clevinger’s fastball. A reference: The 131 wRC+ from 2018 indicates a batting line 31% above league-average, or approximately at Mookie Betts’ 2019 rates. A 75 wRC+, by contrast, is 25% below league-average, or slightly worse than Jake Bauers’ 2019 line. Still, the biggest evidence that the right-hander’s fastball has become a weapon is its ability to generate swings and misses. I compiled a list of four-seam-fastball reliant pitchers from 2017-2018; Clevinger’s swinging strike rate (in which I did not include foul tips) stood in the 22nd percentile at 6.7%. Last season, that rate nearly doubled; at 12.7%, Clevinger fell in the 83rd percentile, just below Max Scherzer.
In past years, despite not being able to touch his secondary pitches, smart hitters were able to wait on Clevinger’s fastball, which he threw (and still throws) over 50% of the time. That’s no longer the case—by upping both his fastball velocity and location, Clevinger turned his one weak pitch into yet another strength. Unless that pitch reverts to its old ways, I expect that Clevinger’s 2019 performance is more indicative of his present-day “true skill” than the projections.
Now, there is room for some regression in other areas. Despite being the Year of the Juiced Ball, Clevinger allowed homers on just 9.3% of fly balls, the third-lowest amongst any pitcher who threw 120+ innings.2 Perhaps this partially a skill (on flyballs off fastballs >2.5 feet above the ground, hitters went 1 for 25, the one hit being a Mitch Garver solo shot), but even so, regression is likely. He also cut down on his walks last season, and who knows whether or not that’s sustainable.
Still, I don’t think anyone would be underestimating Clevinger had he not missed 3/8ths of the season. If you pro-rate his 2019 over 200 innings, here are some of his potential traditional stats:
All the usual caveats to pro-rating apply here—this isn’t actually what would have happened, it’s a magical fantasy in which we can stretch a performance over an arbitrary duration. Still, even if he would have fallen short by a full win, at 6.1 WAR, he’d have tied Charlie Morton for fourth in the AL in fWAR and firmly entrenched in the Cy Young conversation. Entering 2019, Clevinger was Cleveland’s fourth-best pitcher; just one year later, he’s the cream of the crop, and his improvement is largely thanks to his refined fastball. Only one of the Indians divisional rivals plays in the Windy City, but Clevinger will be producing plenty of bat-induced gusts across the Midwest in the coming years.