Ben Bailey

Dillon Maples is not yet a great major league pitcher. On April 28th, he walked the bases loaded, forcing manager Joe Maddon to bring in Tyler Chatwood, who was apparently deemed more likely to throw strikes.

Like Chatwood, Maples is a Cubs pitcher with a high-spin repertoire and command issues. He consistently ranks near the top of the majors in both spin rate and walk rate. Watching this archetype of pitcher is a frustrating experience for fans, but for hitters it can be downright disconcerting.

When Maples is off, it often resembles this:

When he’s on, though, it can look like this:

Hitters have good reason to fear Maples, considering his frequent lack of command and his high velocity. Here are two at-bats from a game against the Brewers from last year.

These two pitches have the highest spin rates of any pitches that Maples has thrown, and one of them hit Aguilar in a dangerous area:

But the high-octane fastball and lack of command aren’t the only reason Maples scares hitters. The more entertaining and less dangerous reason is the incredible movement on his slider. Right-handed hitters in particular seem unable to pick it up out of his hand. This often causes a flinch-and-turn motion even when the pitch is a strike, sometimes even in the middle of the zone. These three at-bats nicely exhibit the effect Maples' slider has on hitters:

I’ve saved the best for last, though. On Wednesday’s game against the Mariners, Maples was asked to get the last three outs of a blowout win. He struck out the side, ending with this sequence to renowned slugger Edwin Encarnación:

Encarnación looks uncomfortable throughout the entire at-bat, and doesn’t come particularly close to taking a threatening pass at any pitch. As it has always been with Maples, the stuff is there. We get to watch him take hitters out of their comfort zones, even if he isn’t able to become a shutdown reliever.

Last season was a concerning one for Jose Quintana. The consistent lefty came with lofty expectations to the Cubs in a trade for top prospect Eloy Jimenez in 2017, and largely lived up to them. However, 2018 was a different story. Quintana struggled with fastball command for a large part of the season, resulting in a career-high 9.2% walk rate. In combination with his lowest strikeout rate since 2015, this led to his lowest strikeout-minus-walk rate since his 2012 rookie season. His 14.7% home run per fly ball rate was a career high, despite the league home run rate taking a slight downtick from record highs last year.

Quintana also greatly struggled the third time through the order in 2018, allowing a .934 OPS compared to his career mark of .770. This makes sense, as Quintana was mostly a two-pitch pitcher, throwing mostly his fastball and curveball. Because neither was an especially dominant pitch, he struggled to be effective when hitters saw him the third time. He was also slightly worse against righties compared to lefties than he has been over his career so far.

In Quintana’s last three starts, though, he’s pitched the best he has in a Cubs uniform. Outside of an early relief appearance and a bad start against Milwaukee, he’s been outstanding this season. So what changed?

The main difference appears to be Quintana’s increased confidence in his changeup. As Sahadev Sharma writes at The Athletic:

In his first two outings, a four-inning relief appearance and a three-inning disastrous start in Milwaukee, Quintana combined to use his changeup just nine times (5.7 percent of the time). In his three starts since, he’s gone to it 35 times (11.5 percent) and gotten nine swings and misses, including three on the 13 he threw Tuesday night.

Using the changeup more and establishing it as a potential weapon should help to alleviate some of Quintana’s issues facing hitters a third time. It should also help him quiet right-handed bats a bit–Quintana is throwing his changeup to left-handed hitters only 0.1% more than last season (3.8)%, but against righties his changeup usage has gone from 7.6% to 12.3%. That’s not the only change he’s made to his pitch mix, however. Quintana has also been relying on his sinker more this year, especially against lefties–he didn’t throw a single four-seam fastball to a lefty in his most recent two starts.

While Quintana’s altered pitch usage is likely the predominant factor in his recent success, there are others that stand out as well.

Quintana has been throwing his pitches in the zone one percentage point higher than in 2018. On the first pitch of each at bat, hitters are swinging two percentage points less (Quintana’s lowest rate since at least 2015), but first pitch strike rate is up by nearly the same amount. This pattern of hitters being less aggressive and Quintana being more aggressive is demonstrated elsewhere, too. Quintana has the tenth lowest in-zone swing rate among qualified starters so far this season, meaning he’s getting a lot of called strikes (despite the lack of assistance from Cubs catchers). Why aren’t hitters swinging?

It’s most likely a combination of two things: 1) Quintana is hitting the edges of the zone a little more, and 2) batters are expecting him to be more wild because of his lack of command last year, so they’re being more patient. Quintana is also tunneling his pitches better than he has since 2013, according to Baseball Prospectus' pitch tunneling data, meaning hitters could be having greater difficulty distinguishing pitches that tend to be in the zone from pitches that often fall below the zone.

Not only are hitters not swinging at hittable pitches from Quintana as much, they’re also swinging and missing much more (nine percentage points) than they were last year, which would be the highest whiff rate of Quintana’s career.

So, Quintana is throwing more strikes, and hitters are swinging less in the zone and more outside the zone. Not only that, they’re making way less contact when they do swing outside the zone and slightly less contact on pitches in the zone. When hitters have made contact off of Quintana this season, they’ve been hitting the ball into the ground. Quintana is known as more of a fly-ball pitcher, but so far this season his fly-ball rate is way down from his Statcast-era high of 23.7% to 16.4%, while his ground-ball rate is up six points to 50.7%.

All these changes have combined for a great stretch of pitching. The results? Quintana has produced half as much fWAR (0.8) as last year (1.6) in about 16% of the innings. He’s on pace for 224 strikeouts in 174 innings, compared to 158 last year.

There could be an adjustment coming if hitters start being more aggressive with Quintana, or if his command wavers. For now, though, Cubs fans should enjoy the ride and hope Quintana continues to push Eloy Jimenez farther from their minds.

As I write this, Tigers starter Matthew Boyd sits atop the Fangraphs pitching WAR leaderboard with 1.2 fWAR so far this season. Among starters, he ranks second in strikeout rate, fourth in swinging strike rate, and fifth in in-zone whiff rate. In short, he’s been really good!

Of course, I’m far from the first to notice Boyd’s breakout. In early 2017, Fangraphs' Jeff Sullivan suggested that Boyd’s slider could be the key to a breakout. While the slider did end up being the key to Boyd’s success, it was a new, slower version, not the faster version that Sullivan was excited about.

After the end of the 2018 season, Michael Ajeto wrote on Boyd at PitcherList, comparing him to fellow surprise Patrick Corbin, noting his increased slider usage.

Before spring training, Jason Beck wrote about Boyd using a data-driven routine to improve his health, positioning him as a prime “best shape of his life” candidate for the season.

After Boyd’s second start of 2019, Brandon Day wrote about Boyd’s surprising whiff rate at Bless You Boys, drawing attention to the new form Boyd’s slider has taken.

Most recently, Fangraphs' Sung Min Kim wrote a few things about Boyd, focusing on where the Tigers starter was locating his pitches and how well he tunnels his fastball and slider.

I think Boyd’s breakout can be summed up in five graphs and four gifs. Since 2017, Boyd’s slider and curveball have gained more (negative) horizontal break:

Since that same year, Boyd’s slider and curveball have both gained more vertical break as well:

He’s started using his slider and four-seam fastball a lot more, and made his curveball his primary tertiary offering:

But Boyd’s breakout isn’t due to increased velocity–both his four-seam fastball and slider have decreased from their 2017 peaks:

However, despite the decrease in velocity, the spin rate of his four-seamer has increased substantially since 2017, making it harder to square up:

Now that we’ve seen how Boyd’s repertoire has changed, we can look at some of his actual pitches from this year.

First, the high-spin four-seam fastball:

Next, two of Boyd’s best pitch, the slow slider:

Finally, two curveballs:

And an overlay of both breaking pitches:

Notice how similar the curveballs and sliders look. Not only do hitters have to deal with Boyd’s well-tunneled fastball and slider combination, but even if they identify that a breaking pitch is coming, the curve and the slider move similarly enough that it’s hard to differentiate before it’s too late. There is only a 4 MPH difference in velocity between the pitches in the gif, but the curve has more depth to it.

Matthew Boyd probably won’t keep dominating hitters at the rate he has been so far this season, but if he does keep it up he might find himself traded to a contender before long.

On Wednesday morning, Ken Rosenthal tweeted Josh Hader’s stats so far this season:

And…60 fastballs out of 62. https://t.co/gXiyGyCLJp

— Ken Rosenthal (@Ken_Rosenthal) April 3, 2019

The second tweet caught my attention. A 97% four-seam fastball rate is more or less unheard of, even through a small sample. In 2018, Hader threw his fastball 77% of the time, and went to his slider 21% of the time. This year, he’s thrown 60 fastballs and one slider. Even Kenley Jansen, one of the game’s best relievers who relies primarily on one pitch, threw his cutter just 80.4% of the time last season.

This lopsided pitch mix could be cause for concern in another reliever, but Hader his still been as dominant as ever. So far, he has a 52.9% K-BB% compared to 36.9% in 2018, and a 40.3% swinging-strike rate, more than double last year. Here he is throwing an immaculate inning with nine straight fastballs:

Hader seems to be doing this intentionally. So, what could be the reason for Hader’s fastball-only repertoire in his first several appearances? Let’s examine a few, from least to most likely.

1. Yasmani Grandal isn’t comfortable catching his slider.

   While Grandal has been behind the plate for all of Hader’s pitches, the idea that he is behind this change seems unlikely. The Brewers surely expect Grandal to be able to catch their best reliever’s full repertoire, and Grandal rates as one of the best defensive catchers (though he seems to be an average blocker).

2. Specific batter matchups dictated it.

   Hader faced Paul Goldschmidt, Paul DeJong, Marcell Ozuna, Yadier Molina, Dexter Fowler (twice), Jose Martinez, Tyler O’Neil, Yairo Munoz, Curt Casali, Scott Schebler, Joey Votto, and Yasiel Puig before throwing anything but his fastball. He then threw a changeup and slider (among fastballs) to Jose Iglesias. After that, Tucker Barnhart, Jose Peraza, and Kyle Farmer only saw his fastball. Hader has faced a wide array of hitters with varying approaches and skill levels, and it is unlikely that they all categorically struggle against fastballs.

3. The Brewers medical staff thinks that throwing the slider puts more stress on Hader’s arm, and relying on the fastball will keep him fresher throughout the season.

   While Hader’s average velocity stayed relatively consistent throughout the 2018 season, his spin rate peaked in April. It is possible that Hader feels the slider is harder on his arm and that he would rather save the pitch for later in the season. However, it seems more likely that Hader’s usage early in the season contributed to the observed declines in spin rate, which corrected themselves in the postseason regardless.

4. Shorter outings allow Hader’s fastball to play up a bit.

   Hader’s four seamer averaged 2043 RPM last year. This year (albeit in a limited sample with a fresh Hader), it’s all the way up to 2302 RPM, closer to his average slider from 2018 than his fastball from the same year. Hader has similarly experienced a slight uptick in velocity on his fastball of 0.7 MPH. So far this season, Hader has been used in a more limited role. Just like many starters' receive a velocity bump when they move to a one-inning bullpen role, it seems possible that Hader could receive the same bump.

   Both this and the previous explanation could explain why Hader feels comfortable throwing his fastball as often as he is, but they don’t explain why he has only thrown one slider. Which brings us to explanation five…

5. Hader hasn’t needed his slider yet.

   Hader was briefly asked about his early reliance on his fastball by the Milwaukee Journal Sentinel’s Tom Haudricourt:

   Josh Hader talks about relying on his fastball and Christian Yelich coming through again. pic.twitter.com/zRhldtp66L

   — Tom (@Haudricourt) April 2, 2019

   His reasoning is pretty simple: hitters can’t seem to deal with his fastball, so why throw anything else? When it stops working, he’ll bring out the slider, but he’s confident enough in the fastball to throw it almost exclusively until hitters give him a reason not to.

It is extraordinarily unlikely that Hader will keep up this rate of fastball usage for the entire season, but even this small stretch to start the season is impressive. Often, pitchers need to add another pitch to keep up with competition, but Hader, one of baseball’s best relievers, is showing that he can subtract a pitch and still get results.

Pitch data from Fangraphs. Video obtained through Baseball Savant’s new video search function.

Cubs manager Joe Maddon informed the media yesterday that Ian Happ would be starting the season in AAA. Spring-training results are not usually significant. In spring, teams tend to emphasize process over results. However, that doesn’t mean that results have no effect on opening-day roster decisions. Several years ago, FiveThirtyEight’s Neil Paine found that a raise in spring-training wOBA of 17 points compared to projected wOBA should raise a player’s projection by one point. Spring performances do have an effect on how we should expect players to perform–the small magnitude of the correlation just means it takes extreme results to shift expectations. With spring training nearing its end, we can see which players' projections should change the most based on their spring-training performance.

This is a table of the 15 greatest decreases in projected wOBA based on spring results, showing spring-training wOBA, Steamer projected wOBA, the difference between the two, and Steamer projected wOBA adjusted according to spring-training wOBA. The second column, “Opp,” is Baseball-Reference’s measure of opponent quality, where 10 is MLB-level pitching, 8 is AAA, and 7 is AA.

Player	Opp	PA	Spring	Proj.	Diff.	Adj.
Edwin Encarnacion	8.3	31	0.127	0.342	-0.215	0.329
Andrelton Simmons	8.6	37	0.116	0.316	-0.200	0.304
Marwin Gonzalez	8.3	28	0.147	0.330	-0.183	0.319
Matt Olson	7.7	35	0.173	0.347	-0.174	0.337
Brett Nicholas	7.4	25	0.153	0.311	-0.158	0.302
Kelby Tomlinson	7.5	27	0.132	0.283	-0.151	0.274
Ian Happ	7.7	56	0.181	0.325	-0.144	0.317
Sean Rodriguez	8.1	52	0.141	0.282	-0.141	0.274
Ichiro Suzuki	8.2	28	0.141	0.275	-0.134	0.267
Adolis Garcia	6.4	25	0.151	0.283	-0.132	0.275
Christian Vazquez	7.4	36	0.163	0.293	-0.130	0.285
Clint Frazier	7.0	53	0.195	0.322	-0.127	0.315
Socrates Brito	7.9	55	0.172	0.295	-0.123	0.288
Jacob Nottingham	7.1	32	0.156	0.277	-0.121	0.270
Jason Heyward	8.2	39	0.206	0.325	-0.119	0.318

In Happ’s case, his bad spring training took him from a 0.325 projected wOBA to 0.317, while Almora’s strong spring brought him from 0.309 to 0.316, nearly Happ’s equal. Another Cubs outfielder, Jason Heyward, also dropped his projected wOBA to just around Happ’s level, but it seems as if Heyward’s lack of options and superior defense won out. Other outliers of note are Marwin Gonzalez, who is off to a slow start after missing a the beginning of spring training as a free agent, the recently retired Ichiro Suzuki, and the aging Edwin Encarnacion.

This is the same table as above but for the biggest increases in projected wOBA based on spring-training results:

Player	Opp	PA	Spring	Proj.	Diff.	Adj.
Domingo Santana	8.1	28	0.570	0.320	0.250	0.335
Byron Buxton	7.8	41	0.547	0.306	0.241	0.320
Phil Gosselin	7.9	42	0.499	0.279	0.220	0.292
Rob Brantly	7.3	25	0.482	0.264	0.218	0.277
Lourdes Gurriel Jr.	6.7	44	0.530	0.313	0.217	0.326
Justin Turner	8.1	48	0.580	0.364	0.216	0.377
Brandon Belt	8.1	44	0.554	0.344	0.210	0.356
Jay Bruce	8.7	33	0.523	0.314	0.209	0.326
Corey Dickerson	8.1	34	0.541	0.332	0.209	0.344
Chance Sisco	6.8	42	0.500	0.296	0.204	0.308
Austin Hays	6.9	40	0.512	0.309	0.203	0.321
Christian Yelich	8.4	39	0.583	0.382	0.201	0.394
Isiah Kiner-Falefa	7.5	28	0.502	0.303	0.199	0.315
Cristhian Adames	7.8	50	0.472	0.273	0.199	0.285
Yan Gomes	8.1	37	0.499	0.303	0.196	0.315

At the top of the list, Domingo Santana appears to be somewhere in between his 2017 and 2018 selves after being traded to the Mariners due to a Milwaukee roster crunch. Paine also recently wrote about Buxton using the same methodology used in his earlier article. Meanwhile, Christian Yelich appears poised to beat the initial projections again and put up numbers closer to his MVP-level 2018 season.

We frequently put too much stock in spring results, but by dismissing all of them we miss out on the outliers that can be indicators of what’s to come. Not all of these outlier performances will prove meaningful, but it’s useful to look at players who have managed to move the needle a bit.

Spring-training stats from Baseball-Reference. Steamer projections from Fangraphs. The R code to produce the above tables can be found in a gist I published.

The pitch clock has been a significant point of public debate for the last several years. I was recently listening to a spring-training game on the radio in which one broadcaster hypothesized that younger pitchers pitch at a faster pace because they are more accustomed to the pitch clock. Intuitively, this seems plausible. Games at the Double-A and Triple-A levels have used the pitch clock since 2015, and it has had a noticeable (if not substantial) effect on the length of minor league games. One wouldn’t expect pitchers who are used to pitching more quickly in the minor leagues to suddenly slow down when they reach the majors.

What do the numbers say? There was a 0.21 correlation between service time and pace for relievers in 2018…

…but only 0.11 for starters…

…and the same 0.11 for all pitchers:

What can explain this discrepancy?

Role	Pace	Service
Starters	23.40	2.07
Relievers	24.95	2.08
All Pitchers	24.36	2.07

Relievers take an average of 1.5 seconds longer between pitches compared to starters. This is probably a result of two factors: first, relievers tend to pitch more with men on base, and pitchers slow down with men on base. Second, relievers throw harder than starters, and Rob Arthur found a correlation between slower pace and added velocity.

Of course, as Travis Sawchik noted on Frangraphs, pitchers are only half to blame for the pace problem. Pace data is also measured for batters, and some (such as new Twin Marwin Gonzalez) take significantly longer between each pitch than others. If the slight positive correlation between service time and pace that exists for pitchers is significant, we would expect it to exist for batters as well.

The correlation between batter service time and pace is almost exactly the same: 0.12.

From this, we can conclude that pitchers with less service time pitch at a slightly quicker pace. What does this mean about the pitch clock? It seems to work, at least to a small extent. Perhaps more interestingly, its effects seem to slightly linger even after pitchers move to a league without it.

Service time info from Cot’s contracts. All other data from Fangraphs.

The first full week of spring-training games just came to a close. Baseball is officially being played! A lot happens during spring training: pitchers hurt their arms, players are in the best shape of their lives, and top free agents sign. Most importantly, though, there is baseball. For some reason, this baseball will feature a number of intentional walks.

Jeff Sullivan wrote about this phenomena three years ago. In the years since, the slight downward trend he identified in spring-training intentional walks has continued. There were eight spring-training intentional walks in 2015, six in 2016, seven in 2017, and only four in 2018. This is not an egregious number of intentional walks, but it’s still more than one would expect. Why are there any intentional walks in games where wins and losses don’t matter? They only serve to take away opportunities for pitchers and batters to face one another.

Here are all the intentional walks that occurred over the last three spring trainings:

Date	Inning	Batter	Pitcher	Batter’s Team	Pitcher’s Team
3/4/16	9	Danny Ortiz	Mason Melotakis	Pirates	Twins
3/12/16	9	Daniel Robertson	Grant Dayton	Mariners	Dodgers
3/25/16	9	Carlos Peguero	Jeurys Familia	Cardinals	Mets
3/27/16	2	Greg Garcia	Jose Fernandez	Cardinals	Marlins
3/27/16	6	Giancarlo Stanton	Sam Tuivailala	Marlins	Cardinals
4/1/16	8	Jose Abreu	Jimmy Brasoban	White Sox	Padres
3/1/17	9	Yoan Moncada	Kaleb Fleck	White Sox	Diamondbacks
3/23/17	9	Matt Tuiasosopo	Daniel Stumpf	Braves	Tigers
3/28/17	6	Dustin Garneau	Mike Hauschild	Rockies	Rangers
3/31/17	8	Brandon Crawford	Ryan Dull	Giants	Athletics
4/1/17	9	Rymer Liriano	Corey Knebel	White Sox	Brewers
4/1/17	9	Kirk Nieuwenhuis	Michael Ynoa	Brewers	White Sox
4/1/17	9	Nate Orf	Michael Ynoa	Brewers	White Sox
3/12/18	9	Will Smith	Jacob Barnes	Dodgers	Brewers
3/15/18	9	Darwin Barney	Ernesto Fieri	Rangers	Brewers
3/16/18	6	Pedro Severino	Josh Lucas	Nationals	Cardinals
3/27/18	9	Ehire Adrianza	Trevor Gott	Twins	Nationals

It’s possible that, before 2017, when the rule change allowing managers to simply signal for intentional walks occurred, there was an idea that some pitchers needed practice throwing four intentional balls. Maybe in 2017 managers felt that this very small group of pitchers needed practice watching batters go to first base without any pitches being thrown at all. Perhaps that rule change was the beginning of the end for meaningless intentional walks, but I expect we’ll still see two or three this year, probably in the ninth inning and probably toward the end of spring.

We have to give managers the benefit of the doubt in most cases. They know what they’re doing. They know the games don’t matter, though. So why do they issue intentional walks in spring training?

Looking at the table above, you might notice that three of 2017’s intentional walks came in one game. That’s not a typo—on April 1st, 2017, three intentional walks were issued in the ninth inning of a spring-training game. Let’s see if we can figure out why.

This particular game was special: it was played in Miller Park, not a spring-training stadium, and it was each team’s last preseason game. The game has a normal amount of action, but the ninth inning is of particular interest.

The top of the ninth sees the White Sox pull within one run on a sacrifice fly. Unfortunately, catcher Rene Garcia is injured in a home plate collision on the play. After a fifteen minute injury delay, Brewers manager Craig Counsell decides to bring in a fresh arm in Corey Knebel, the not-yet-notable reliever. Knebel would end spring with a 17 strikeouts and three walks (one intentional), providing a glimpse at the breakout season to come, but he was not yet widely known. White Sox broadcasters Benetti and Stone go through several pronunciations of his last name (“Knee-bull” and “Knay-bull”) before settling on the correct “Kuh-naybull.”

Knebel’s first two pitches to Rymer Liriano are balls. Counsell then calls for the intentional walk.

The best part about the games from this particular year is that we get to see players get used to the new intentional walk rule. For each of these players, there’s a cycle of confusion, disbelief, and reluctant acceptance. Here is former White Sox outfielder Rymer Liriano learning he was being intentionally walked in the ninth inning of the last spring-training game of the season:

Knebel strikes out the next two batters without incident, sending the game to the bottom of the ninth. This is the sequence of the events in that half inning:¹

1. Jett Bandy is hit by a pitch
2. Bandy steals second
3. Bandy advances to third on a wild pitch
4. Eric Sogard strikes out
5. Kirk Nieuwenhuis is intentionally walked

6. Mound visit 7. Nieuwenhuis steals second 8. Nate Orf is intentionally walked

9. Mound visit and pitching change (before which Michael Ynoa receives nine pats from his teammates and manager)

10. Craig Counsell gets impatient 11. Kyle Wren wins the game with a single

The walk of Rymer Liriano was clearly strategic, since Counsell only called for it after Knebel threw two balls to start the plate appearance. The walk of Nate Orf was also strategic, as it only came after Nieuwenhuis stole second during the plate appearance.

As previously established, though, strategic validity is not in question here, but rather the reason for strategy at all. In this case, maybe the managers were trying to get their pitchers used to the no-pitch intentional walk rule that was instituted for the 2017 season. Or perhaps they wanted to help their young pitchers by putting them in better situations. Maybe they wanted to end spring on a high note. Most likely, though, they were just trying to win.

When the Brewers do win, their mascot still celebrates, even though it’s a mere spring-training victory. The fans still boo (like when Knebel struggles to find the strike zone) and cheer throughout the game, the first-base coaches still give fist bumps, the broadcasters still talk about the winning run being on deck, and the players still have an on-field celebration, although it is decidedly more subdued than that of a regular season game.

Spring training is baseball where the central conceit—that one team wins and the other loses and that those results have real ramifications—is suspended. Winning the World Series matters in real, concrete ways. It affects salaries and attendance. Winning regular season games matters for similar reasons, in addition to getting teams closer to the World Series. But even in spring training, when the results have absolutely no impact on whether a team makes it to the games that matter most, managers strategize and players try. Less established players are fighting for roster spots and playing time, but even more established players who have no concerns about job security want to play their best. Managers are not being judged by the moves they make in spring training, but sometimes they can’t help but call for an intentional walk that will tip the needle ever so slightly in their team’s favor. The games don’t matter, but sometimes teams can’t stop trying to win.

---

1. Not one of these players remains on the Brewers' roster besides Nate Orf, who is a non-roster invitee this year. ↩︎

The 2018 Los Angeles Dodgers began their season by being shutout 1-0 by the San Fransisco Giants at home. The next night began similarly, featuring a pitchers' duel between Johnny Cueto and Alex Wood. With the game still tied and each team held to a lone hit going into the top of the ninth, the Dodgers summoned their best bullpen weapon. Kenley Jansen has established a reputation as one of the most dominant closers in baseball over the last few years, largely due to his magnificent cutter. Although Jansen doesn’t project a typically aggressive closer mentality on the mound, he has been among the best in the game for a while now, and the Dodgers are now trusting him to keep the Giants at bay to avoid starting the season with consecutive losses to their rivals at home. The crowd has faith in Jansen; he is their closer, and he has proven himself many times over in similar situations.

The entrance of a closer into a close game is one of the more dramatic moments that baseball treats fans to on a regular basis. Despite evidence to the contrary, it’s hard not to get swept up in the idea that this is the climax of the game. Even in today’s era of shifting roles and anonymous relievers, fans still have favorite pitchers that they want to see coming out of the bullpen, and Jansen is toward the top of the list.

Giants' second baseman Joe Panik (who will go on to post a 75 wRC+ this season) steps into the batter’s box and receives the first pitch from Jansen: an 88 MPH cutter, nearly five MPH below his season average. This drop is cause for some concern, but it is possible that it was just a blip, or that the lower velocity won’t have an effect on the Giants' success against Jansen. The tension does not last for long, as Panik hits the second pitch (a 90 MPH cutter) for a home run.¹ Jansen would get out of the inning without further damage, but his average cutter velocity (89.83 MPH) was the lowest since 4/1/16. He just didn’t have it that day.

Sometimes, a reliever will come out of the pen throwing a bit harder than normal. Others, it will seem like he just doesn’t have the feel for the pitch he’s best known for. Some pitchers are more consistent in their velocity than others, and I wanted to come up with one number to compare pitchers to their peers.

To get this metric, I took a list of all pitchers who pitched in 2018. For each pitcher, I calculated the average velocity of each of his pitches in each of his outings. I then calculated the coefficient of variation² of that velocity for each of that pitcher’s pitches, weighted them by how much he threw each pitch, and summed them to get one composite number for each pitcher. I’ll call this metric the velocity coefficient of variance (VCV). VCV+ is just VCV normalized to have 100 be the league average. It’s a bit easier to read and contextualize, so I’ll be using it in the rest of this piece.²

The full table of VCV data for 2018 is available here, but I’ll attempt a brief summary of things I found interesting.

The pitcher with the highest VCV+ in 2018 was Brandon Maurer (who had a rough 2018), with 196. AL Wild Card opener Liam Hendriks (190) came in a close second, followed by Mike Fiers (187). The pitcher with the lowest VCV+ in 2018 was Carson Fulmer, who put up a VCV+ of 43 in just eight starts and 32.1 innings. In a larger sample of over fifteen starts, Alex Cobb and Dan Straily both had a VCV+ of 57. Meanwhile, Chris Sale had a VCV+ of 173, likely due to his wild fluctuations in velocity late in the year due to injury and illness.

For those curious, the Red Sox, Rockies, and Braves had the highest CVP, and the Astros, Diamondbacks, and Yankees had the lowest, though it is unlikely that this is something teams are selecting pitchers for.

Perhaps the most interesting finding is the difference between starters and relievers. The mean VCV+ of pitchers who made more than ten starts was 94, or 93 if you raise the bar to 15 starts, while it was 104 for pitchers who made no starts and pitched at least ten innings, and 106 for those who pitched at least 20 innings. In 2018, the average velocity of each pitch in an established reliever’s arsenal varied 13% more between outings than that of an established starter. In fact, VCV may have been most strongly correlated with innings pitched (cor = -0.179, p = 0.00006).

VCV in 2018 was significantly coordinated with BB/9 (cor = 0.160, p = 0.0003) and K/9 (cor = 0.140, p = 0.002). In 2017, VCV was similarly significantly correlated with BB/9, but not with K/9, so it’s safest to assume that BB/9 is the main significant correlation. The correlation of walks to VCV is small. However, the fact that they are correlated makes sense intuitively. Walks are often thought of as wildness and a lack of consistency, so it’s not surprising that those pitchers who show a slight lack of consistency in location would also show the same tendency in average velocity. VCV was not significantly correlated with FIP or xFIP.

Next time, I think it would be interesting to see if variation for each pitch correlates to pitch value data at all.

Pitch velocity and usage frequency data from Baseball Savant. BB/9, K/9, FIP, and xFIP data from Fangraphs. Data processed with the help of the baseballr and dplyr R packages. If you want to download the full table of VCV numbers, you can do so here.

---

1. Incidentally, the one run the Giants scored the previous night was also on a solo home run from Panik. After the Jansen game, he would hit two more over the rest of the season. ↩︎

2. The coefficient of variation is a way to standardize the standard deviation based on the mean. Because we’re using the coefficient of variation instead of the standard deviation, if Kyle Hendricks' mean velocities varied by the same amount as Jordan Hicks', Hendricks' VCV would be higher despite Hicks throwing much harder. (Note: this is not the case–in 2018, Hendricks had a VCV+ of 61, compared to Hicks' 134.) ↩︎