clock menu more-arrow no yes

Filed under:

Breaking down Luis Arraez’s sophomore slump

New, 10 comments

The surface stats don’t show it, but he’s improving in year 2

Milwaukee Brewers v Minnesota Twins Photo by Brace Hemmelgarn/Minnesota Twins/Getty Images

After a breakout rookie season (.334 / .399 / .439) that put Luis Arraez’s offensive production about 25% better than league average (125 wRC+), expectations were inflated for the Twins’ diminutive second basemen coming into 2020. The major projection systems were bought in on him because of his elite contact skills (93.3%) and plate discipline (36 BB to 29 K). Both ZiPS and Steamer, the publicly available projection systems hosted at Fangraphs, projected Arraez to lead the major leagues in batting average – projecting .310 and .312 respectively.

Projecting the future with any degree of consistent certainty is impossible. After a little more than half a season we’re reminded of that, and it’s safe to say that many in Twins Territory are disappointed by Arraez’s 2020 campaign. Battling a nagging knee injury, his traditional numbers have declined from his rookie levels across the board, coming in at .264 / .323 / .299 through 25 games and 96 plate appearances (through Sunday, August 30). Altogether, that line works out to an 73 wRC+ that ranks 23rd among MLB second basemen with at least that many plate appearances. Considering the MLB average wRC+ for second basemen this year is 90, Arraez’s disappointing start to 2020 is not that far off from MLB average for his position. It’s important to keep this in mind. Despite our high expectations not being met and maybe even frustration at his start, Arraez is still providing value. Nonetheless, he has not performed the way we expected, and his lower production is a factor in the team’s declined overall offensive production.

Are there legitimate reasons to be concerned? Is this a sophomore season learning curve or was his rookie year a flash in the pan? Let’s dig into his data and see if we can figure out what’s going on.

Strikeouts and Walks

The first things I check when a hitter’s top-level numbers are less than expected is strikeouts and walks. These are good proxies for understanding if a hitter is experiencing a change in their ability to control the strike zone. In Arraez’s case, his walk and strikeout rates have both changed ever so slightly for the worse in 2020. It’s important to point out that 96 plate appearances are a very small sample size. In a normal 162 game season, a qualified player will take around 600 plate appearances. So Arraez’s 96 PAs are only about 15 percent of a typical full season amount. Regardless, he’s struck out in 11.5% of his plate appearances after doing so in just 7.9% last season. This could just be noise, though. He’s struck out 11 times. Using last year’s rate, we would have expected about 7 strikeouts at this point. It’s hard to say with confidence that something is very different here.

It’s also worth remembering that 7.9% was easily the lowest strikeout rate in the majors among players with at least 350 PAs last season. This season’s 11.5% still ranks in the top 5% across the game. On the walk side of the equation, his 8.9% rate is slightly down from last year’s 9.8% - but across 96 PAs that difference equates to just one fewer walk than we would expect from last year’s rate. That’s clearly just noise.

Given this data, I’m concluding there isn’t much change in Arraez’s ability to control the strike zone. He remains elite here.

More granular swing data

Strikeouts and walks are outcome stats, so I next like to dig into more granular process-oriented swing numbers. Frequent causes of hitters underperforming include chasing pitches out of the strike zone, being too passive in the zone (i.e., letting hittable strikes go by without swinging), or just plain old swinging and missing.

In Arraez’s case, he’s had no change in his overall swing rate (42.0% both seasons). To his credit, he’s also slightly improved his chase rate, swinging at pitches out of the strike zone only 21.6% (down from 24.2% in 2019). It’s clear chasing out of the zone is not an issue with Arraez.

How about in the strike zone? No doubt having learned from last year, pitchers are approaching Arraez a bit more cautiously in 2020, throwing him fewer pitches in the strike zone. Arraez has responded to this by maintaining his eye as I described and being more aggressive on pitches in the zone. This season he’s swung against 66.7% of pitches in the zone (last year, 60.8%), including 74.1% against pitches classified as “meatballs.” These are the kind of improvements you want to see from a young player – getting himself into favorable counts and then being aggressive to try to take advantage.

Where Arraez has experienced some decline is in making contact, seeing his swing and miss rate increase (11.3%, up from 7.9%) and his zone contact % decrease (90.4%, down from 94.0%). While he’s performed worse in this area, he’s still elite in terms of putting the ball in play. The 11.3% whiff % is good for 4th-lowest in MLB (99th percentile) among hitters with at least 25 PAs.

It doesn’t look like there’s much in his swing data that explains his lower production. The swing and miss increase may be a little factor, but it’s small, and he remains among the games absolute best in this area. I’d actually argue he’s made some positive improvements to his already strong approach.

Batted ball profile

Another common cause of disappointing production is a shift in a player’s batted ball profile, such as hitting more groundballs than the past, or an abnormal number of popups that turn into easy outs. How does Arraez look in this area?

With Statcast data you could argue he’s improved slightly in this area too. So far, he’s hit a few less grounders (39.7%, down from 42.0%) and popups (1.4%, down from 3.3%) in exchange for slightly more line drives (35.6%, up from 32.3%). For a contact-oriented spray hitter like Arraez, these are good trades to make, even if they might be noisy because of the sample size. In total, his average launch angle has increased to 12.5 degrees from last year’s 11.5 degrees. That is a positive change, too.

Last season, batted balls with a launch angle of 12 degrees yielded a batting average of .786. Batted balls at 13 degrees yielded a batting average of .816, which are the two highest points of the gold line in the chart above. Arraez was already in a good position here as a hitter for average and he’s made some improvements to be even better.

In terms of his spray chart, Arraez is pulling the ball slightly less (21.9%, down from 27.0%) and going up the middle more (45.2%, up from 36.3%), but that change shouldn’t be that impactful for a hitter who doesn’t rely on pull power to be productive.

How about the quality of his contact? Are there any significant changes in how hard he’s hitting the ball? By Statcast data, Arraez had an average exit velocity of 87.1 miles per hour in 2019. So far in 2020, that’s 86.8 miles per hour – essentially the same. Any batted ball with an exit velocity of 95 mph or higher get classified as “hard hit,” and Arraez’s hard hit percentage is basically the same in 2020 (23.3%) as 2019 (22.8%). Pretty much no matter how you want to slice his contact quality stats, Arraez is as good or better than he was last year – less weak hits, fewer topped grounders, fewer fly balls that he got under, more solid hits, and more barrels. There doesn’t appear to be an issue here either. If anything, this data suggests he’s the same or better than he was last year.

Expected stats and optimism

So, what gives? There doesn’t appear to be anything we can pinpoint in Arraez’s data that would explain his depressed production. By and large, he’s doing just about everything the same, or in some cases, better, than he did in 2019. It seems this is mostly just a case of bad luck in small sample sizes. His batting average on balls in play (BABIP), is just .299 – a figure that is lower than any of his seven professional seasons, and significantly lower than last year’s .355.

Statcast calculates “expected” statistics based on a hitter’s contact quality and launch angle. Arraez’s combination of batted ball data points should have a batting average of .298 and a slugging percentage of .445. Those are figures that would be much more in line with our expectations for him. Altogether, his data works out to an expected weighted on base average (xwOBA) of .361 that is exactly in line with last season’s actual .360. Perhaps even more encouraging is the fact that his expected stats in 2020 are better across the board than his 2019 expected stats were.

Ultimately, the difference between Arraez’s actual slugging percentage (.299) and expected slugging percentage is 148 points – a gap that is 6th largest in baseball. The 84-point gap between his .277 wOBA and his expected wOBA is tied for 5th largest. At the league level, those gaps are 22 points and 18 points, respectively. While it might seem simple to dismiss the advanced stats as being noisy because of small samples, Arraez’s differences are clearly much larger than should be expected.

The data indicates that Luis Arraez has been just what we expected and is continuing to develop in positive ways in his second season. We should expect to see some positive regression for Arraez the rest of the way, something that will be a welcome addition to a team that has struggled offensively and is relying on its current roster to return to form for the playoff push.


John is a contributor to Twinkie Town with an emphasis on analytics. He is a lifelong Twins fan and former college pitcher.