Pitchers Most Impacted By Defense: OAA Risers

February 12, 2020

Eric Samulski looks at the new Statcast OAA defensive metric and uses it to analyze which pitchers were hurt most by their team defenses during the 2019 season.

Early in January, Statcast released a full fielding leaderboard of their Outs Above Average (OAA) metric, which had previously been limited to outfielders only. The metric takes into account fielder positioning, reaction time, throw difficulty, and batter speed in order to calculate how likely a play is to become an out. The more unlucky, the more points a fielder will get for converting the out and the fewer he'll lose if the play winds up becoming a hit.

There's a more detailed breakdown of it here, but essentially it aims to tell us just how good a fielder is and just how many potential outs he recorded over the norm.

Does this new metric have any impact on fantasy baseball? Perhaps we can find a way to evaluate the stickiness of the metric or see just how much we should be using it when projecting pitching value for a given season.

The Unlucky Ones  (Pitchers with Low OAA)

Now that all infielders have been charted on this metric what can be extrapolated out of that is just how much a certain pitcher benefitted from elite defense behind him. In their own words, Statcast can track "the performance of the defense behind the pitcher while he was on the mound." The higher the recorded OAA when a pitcher is on the mound, the more outs his defense recorded that likely shouldn't have been outs based on the average outcome. The lower the OAA is for a pitcher, the more his defense gave away outs that should have been converted on average.

This alone isn't enough to determine if a pitcher is good or bad or if their 2019 fantasy season was real or fluky; however, placed in a larger context, it can tell us which pitcher's stats may have been the result of their own changes, defensive skills, luck, or perhaps a grouping of the three.

Since this metric is relatively new when it comes to its impact on pitching, I decided to dig into a few names from the top and bottom of the leaderboard just to see if there was anything to take away. I'm thinking aloud here or, more precisely, typing as I go, so feel free to take a look at the Statcast Leaderboard yourself and see what stands out to you.

Michael Pineda (SP, MIN)

According to OAA, Michael Pineda was the pitcher most hurt by his infield defense with a -6% success rate added and a -8 total OAA. What's interesting is that Pineda has a career 42.7% GB rate but only allowed 36.1% last year as his LD% and FB% jumped two and eight points respectively from his last full seasons. So, overall Pineda didn't require his infield defense to do a lot for him. Part of this could be attributed to an increase in fastball usage and a decrease in sliders in his first year back from injury.

As Jeff Zimmerman has pointed out in his recent article on the way injury affects performance, it would be smart to assume Pineda goes back to matching his career levels now that he is another year removed from injury. With that, plus the addition of Josh Donaldson at 3B (18th in OAA), Pineda should at least duplicate his low 4s ERA, so I would take advantage of projection systems that seem to project him around 4.60.

Matthew Boyd (SP, DET)

Boyd is generally more of a flyball pitcher, but he saw his GB% rise in 2019 as he began to use his slider more. The poor defensive numbers behind him were likely impacted by Jeimer Candelario's injury, since he had a 3.4 UZR at 3B, while his replacement, Dawel Lugo, had a -2.7 UZR. However, there might not be much improvement in 2020. If Willi Castro wins the starting shortstop job, he will bring league-average defense to the position. The only issue with that is that Niko Goodrum had a 1.9 UZR at SS (17th in OAA) but a -.6 at 3B, where he will likely play if Castro breaks camp with the team. That will then leave Jonathan Schoop and his -3.8 UZR at 2B and CJ Cron and his -.6 UZR and 74th-ranked OAA at 1B. Boyd's OAA could regress a bit to the mean, which would lead to a slight improvement in his .307 BABIP, but his new infield defense isn't going to be an asset behind him and shouldn't suggest actionable growth for 2020.

On a side note about Schoop and where this OAA will be intriguing to keep diving into: Schoop ranks 29th overall on the OAA leaderboard, but a closer look at his individual breakdown shows that Schoop's best plays come when he is essentially shifted to play the normal position of shortstop. A lot of the plays he has to make when he's positioned as normal second baseman are decidedly average. It's fascinating and overwhelming at the same time. 

Masahiro Tanaka (SP, NYY)

Tanaka is a splitter/slider pitcher who accumulates near 50% groundballs in a season, so he relies on defense more than a lot of other arms. While his slider improved last year to a 20.3 pVAL, the splitter plummeted, perhaps because of the new baseball, to a -5.5 pVAL, the worst of his career by a large margin. The ineffectiveness caused him to rely more on a fastball that has never been a good pitch, and he saw his O-Contact rise 10% and his SwStr% drop to a career-low 10.7%.

In order for Tanaka to have any success, he needs to find that splitter again. Without getting batters to miss, he is allowing more contact in front of a defense that is clearly not helping him out. Luke Voit was a -3.9 UZR at first base; Gleyber Torres was a -4.2 at 2B and -2.1 at shortstop, while Gio Urshela was a -2.5 UZR at 3B and still way better than Miguel Andujar. If we're using OAA, Voit ranked 38th out of 40 qualified first baseman, Gleyber Torres ranked 129th out of all infielders, and Urshela ranked 75th. Simply put, Tanaka isn't going to get help from his defense, and if the splitter isn't biting hard in the spring, it's best to stay away from him in most drafts.

Jon Lester (SP, CHC)

Lester showing up on this list is interesting because we generally think of the Cubs' infield of Javier Baez, Anthony Rizzo, and Kris Bryant to be a solid defensive one. In fact, Baez was the best infield defender according to OAA, while Addison Russell was 30th once he came back and slotted in at 2B. Rizzo has a 3.7 UZR, but he appears as the 102nd infielder based on OAA, mostly due to underperformance on plays close to the line. Since Rizzo performed much better in the same metric in 2017 and 2018, it's reasonable to expect improvement defensively from him. So, on the surface, there is no reason for the Cubs to have a poor defense, and very few Cubs appear negatively impacted by bad defense, according to OAA.

Perhaps it's just an issue of Lester being unlucky? Subpar defense performance might also help explain why Lester came into the season with a career BABIP under .300 but registered a .347 BABIP in 2019 despite the same Contact%. Now, Lester isn't totally free of blame. He gave up more hard contact than any other year in his career, while also seeing a 3% jump in O-Swing% and 2.6% jump in O-Contact%, suggesting that he was unable to effectively put hitters away or get them to chase. While much of his poor season had to do with regression in skills, the OAA numbers paired with a rise in K% and a decline in BB% suggest that Lester may pitch closer to his 2019 xFIP of 4.35 than current projections, which have him over a 4.70 ERA.

Steven Matz (SP, NYM)

A step lower on the OAA leaderboard for pitchers with the least amount of defense help is Steven Matz. It'snot entirely surprising considering that the Mets had one of the worst infield defenses last year and figure to this year as well. Amed Rosario was the 122nd infielder based on OAA, while Peter Alonso was 132nd, and Robinson Cano was 91st. Jeff McNeil replacing Todd Frazier at 3B should help a little since McNeil had a 2.4 UZR in his limited reps at 3B last year, but the overall unit won't be a benefit to Matz. Although Matz saw a .034 jump in BABIP last year despite allowing fewer groundballs, he also saw both his O-Swing% and O-Contact% rise, like Lester, and his Z-Swing% and Z-Contact% as well.

In all, Matz gave up more contact, produced a below-average 9.6 SwStr%, and generated a lower K% than in 2018. His defense may not have helped him, but I wouldn't be banking on a major difference in 2020 as Matz still has the same defense and a discouraging set of underlying metrics.

Adrian Houser (SP, MIL)

Houser's presence on this list intrigues me because of how much more reliant he was on his defense in 2019 than years past. Although Houser's Major League K% jumped from 13.6% in an admittedly small 2018 sample size to 25.3% last year, he also saw his GB% jump from 39.5% to 53.4%, which is more in line with his minor league numbers. Historically, Houser has allowed a fair amount of contact, particularly medium contact, and that was impacted last year by the added emphasis on a sinker, which he threw 36% of the time. The pitch registered a 17.1 whiff%, which was by far the lowest of any of his pitches but had a .242 BAA and a .221 xBA. Houser is likely to see more of a correction to the xBA given a stronger defense behind him. Eric Sogard, currently penciled in at third base, was the 27th best infielder in OAA, which is an improvement over Travis Shaw's 70th ranking, and new first baseman Justin Smoak has a career 2.7 UZR at first base. Additionally, as Keston Hiura gets more comfortable at the Major League level, he's bound to improve on his 130th ranking in OAA.

While we shouldn't expect a massive jump from Houser, a better defense behind him and more regression to the league average with OAA should lead to fewer hits and a potentially lower ERA, likely signaling that last year's 3.60 xFIP and 3.72 ERA was no fluke. If he's given a full season in the Brewers' rotation, he could certainly prove valuable.

More 2020 Fantasy Baseball Advice