Jose Quintana Is Finding His Level

Jose Quintana’s season is a little difficult to figure out. In three of his first four starts, he allowed 17 runs combined; in that fourth appearance, he pitched six shutout innings. Over four of his past five starts, meanwhile, he’s conceded one run total; he allowed six runs in the other.

Is he improving as the season goes on? He struck out 16 batters against 11 walks in 19.2 innings in those first four starts and struck out 29 hitters and walked 14 in 28.2 innings over the past five outings. His ERA is much lower in the latter of those two periods, but his FIP hasn’t moved a great deal, going from 5.15 to 4.37. These samples are small enough that it would be fair to conclude little to nothing had changed at all, but given his excellent track record, there has to be something to Quintana’s struggles.

A year ago at this time, there were some questions about Quintana’s trade value for the White Sox after the left-hander started slowly. Over his first 11 outings last season, Quintana had authored a 5.60 ERA and 4.40 FIP. Most of Quintana’s issues in terms of ERA stemmed from an increase in homers and a poor left-on-base percentage. After investigating Quintana’s numbers, I felt that most of Quintana’s issues were probably luck-based and unlikely to continue. Brushing off those concerns looks pretty good in hindsight, as Quintana put up a 3.30 FIP and 3.40 ERA and allowed just under one homer per nine innings the rest of the way, eventually helping a Cubs teams desperate for quality starter innings. This season, Quintana’s issues aren’t as easy to brush off.

While Quintana’s current FIP might resemble last year’s figure at roughly the same point in the season, his 4.68 FIP is nearly 20% worse than league average after accounting for the change in league and park. Additionally, there isn’t a gap in contact quality that suggests perhaps Quintana is just getting unlucky. So far, Quintana is giving up home runs because he has deserved to give up home runs. He’s striking out fewer batters than he did a year ago while his walks have gone way up. These are all bad things.

If there’s any cause for hope, it is twofold. One, Quintana has shown some flashes of being the very good pitcher he was before the season started, putting up four starts of at least six innings and zero or one run with at least five strikeouts. The second reason for optimism is that Quintana is still tinkering.

Before examining Quintana’s good performances, let’s look at how he got here. The graph below depicts Quintana’s pitch-mix changes over time. Notice that 2016 is fairly steady, while 2017 features big changes, seeming to correspond with a change in teams.

Heading into the 2016 season, Quintana used his four-seam fastball about 40% of the time, his sinker and curve about one quarter of the time each. His change accounted for the remaining 10% of the pitches. Near the beginning of last year, he started using his curve a lot more, and then after the trade to the Cubs, his sinker and four-seam flipped before ending the season pretty close together.

At the start of this season, meanwhile, Quintana went to extremes with his four-seam fastball while using his sinker sparingly. The move made some sense: last season, he gave up 12 homers on the sinker, the most of any pitch. The results haven’t been there for Quintana in the early going, however, as hitters have swung at the four-seam fastball out of the zone just 18% of the time. Getting batters to chase a four-seam fastball out of the zone is a good strategy — if the hitters actually swing. Without those swings, though, we just see a lot of taken balls and a resulting spike in walks. The higher swinging-strike rate with the four-seam fastball isn’t providing enough of a trade-off to make the increased usage worth it.

Over the past few games, we’ve seen the decreased four-seam usage with the number of sinkers rising. The curve is down a little bit, but we can see those numbers at the end all approaching his averages over the past two seasons. It looks like Quintana is still searching for the right mix for him this season, but his willingness to change last year might have helped his numbers progress as the season wore on.

When examining Quintana’s performance before his most recent start, I was unable to identify anything that might have led his present issues. His last start, however, helped provide a potential theory. It was one of his best, resulting in seven strikeouts, four walks, and just one hit in seven shutout innings. It was also the start in which he recorded his highest average four-seam velocity on the season. Like most pitchers, velocity for the Cubs’ lefty fluctuates by start. Here’s what it looks like since the beginning of the 2017 campaign:

Quintana’s most recent start was one of the best of the season; it also, perhaps not coincidentally, featured his top velocity mark. Quintana’s first and third starts with low velocity featured just three strikeouts, eight walks, and 13 of the 24 earned runs Quintana has allowed on the season. Last year, Rob Arthur and Greg Matthews published a piece at FiveThirtyEight which found a connection between individual pitcher performance and velocity. While perhaps somewhat intuitive given the general importance of pitch velocity in the game, it does make some sense that the faster a pitcher throws in a given start, the more effective he might be if everything else is the same — i.e. the pitcher isn’t overthrowing and sacrificing location or allowing some other variable to affect his pitches.

The graph below shows xFIP and average fastball velocity by start since the beginning of 2017 for Quintana. I used xFIP — which normalizes home-run rate per fly ball — instead of FIP, for example, because homers can skew results such as these somewhat.

The relationship isn’t incredibly strong, but given the only information we know above is fastball velocity, it seems to be a pretty strong correlation. We can see a whole bunch of 92.4-mph starts. The table below shows Quintana’s performance when he’s throwing at least 92.4 mph compared to when he’s below that number.

Jose Quintana doesn’t need velocity to succeed. The study by Arthur and Matthews showed Quintana’s numbers didn’t move downward a ton without velocity, but they did move up a decent amount when the velocity increased. This conclusion might not be particularly satisfying — we don’t know at what velocity Quintana is going to pitch going forward — but it is encouraging given velocity tends to increase as the season wears on. While the start for Quintana has certainly been discouraging, as long as Quintana isn’t hurting — and his velocity in the last start is a positive indicator he’s throwing well — he seems likely to find the right formula for success.