More Than a Putout Puzzle: Revisiting the Problem of Outfield Alignment

During the playoffs, Nick Castellanos made a few noteworthy catches, darting and diving to his left. I wondered whether the Phillies’ right fielder, with his mitt on his left hand, was alone in relishing plays to his gloveside. After all, Castellanos didn’t have to reach across his body to make the catch when moving in this direction. I also wondered what, if any, implications this would have on outfield positioning and alignment.

In terms of directional Outs Above Average (OAA), my findings demonstrated that right-handed outfielders actually performed marginally better moving to their armside than their gloveside. But they were considerably stronger to their left than left-handers, who in turn performed much better to their own gloveside. If that word salad confused you, not to worry. These numbers might be easier to digest:

Given that in at least one way fielders of either handedness seemed to have an advantage to their gloveside, these results provided enough backing for my hypothesis for me to look into their implications. One way that directional fielding might influence strategy is in terms of outfield positioning. For instance, if a fielder performs better moving to their left than to their right, a team might place them more toward their right than the average fielder. This way, they wouldn’t have to cover as much ground to make a play on a ball hit to their weaker side.

I found in my last article that on the whole, outfielders were generally positioned pretty close to where fly balls were hit. To review some more:

Interestingly, in the 240-man sample from my previous piece, the most popular position for lefties was left field; that’s where 41% of them played. Left fielders were positioned on average to the left of where their fly balls were hit, meaning lefties typically had to move to their armside.

However, lefties playing left field also have the option to cede armside fly balls to the center fielder. The least popular position for righties was right field, but perhaps it should be favored more; this way, they too could yield to the captain of the outfield when it comes to armside knocks. Possibly understanding this, teams on average had right fielders who were better moving to their left and left fielders who were better moving to their right:

It’s unclear if teams deliberately aligned their fielders such that center fielders could pick up the slack on their corner outfielders’ weaker sides. Perhaps hits to the gaps in left- and right-center that the center fielder doesn’t get to are just typically harder to track down, or they go for hits more because they don’t have a chance of going foul, and these things drag corner outfielders’ numbers down in the gap directions.

Lacking conclusive evidence, I turned to individual positioning data. Sorting by position (LF/CF/RF) and handedness did not yield any differences in starting spot. Nor did sorting by position and whether or not the players were better moving to their right. But teams did seem to play their best backpedallers a bit shallower than normal:

Even these differences were small, and not present for left fielders. And there are other important factors that should hold weight when it comes to not only outfield positioning but overall alignment. For example, nearly 10 years ago on this very site, Dave Cameron examined how important it is to have your best defensive outfielder in center field based on the amount of action he’ll get. With today’s more precise data, I recreated his study.

But first, I went traditional, looking at putouts per outfield position to establish a baseline. Cameron’s theory was that the distribution of outfield tries is generally 30/40/30 for left/center/right. For 2022 putouts, that largely held true: the distribution stood at 28.6/39.0/32.4. Cameron also posited that balls in the air (fly balls, line drives, and popups) that were fielded by an outfielder would have a similar distribution regardless of whether they were putouts or not. For 2022, it was a little tighter at 31.1/36.0/33.0, but still in the ballpark of 30/40/30.

But, with the ability to calculate spray angle from Statcast data, we can do Cameron one better; we don’t have to settle for the ultimate fielder of the ball. We can instead look at the area the ball was hit to, which will remove the bias created by hawkish center fielders taking charge on balls hit to the gaps. Splitting the 90° baseball diamond into thirds, the distribution was 26.1/45.3/28.6. This is a bit further from the 30/40/30 distribution, and interestingly, center field gets even more play; perhaps it’s the corner outfielders that are taking balls away from the center fielder after all.

One explanation for this is that there are a lot of gap-doubles that end up past the center fielder and picked up by one of the outfield corners, but there are fewer down-the-line doubles that end up past a corner and fielded by the center fielder. Check out the distributions when we look only at extra-base hits to the outfield:

This doesn’t change the fact that center fielders are all but certain to lead outfielders in putout opportunities. They also seem to lead outfielders in difficult putout opportunities: the only meaningful difference in xWOBA on contact for batted ball types across outfielders is for fly balls, where center fielders receive significantly tougher opportunities:

Given that more flyballs are hit to center field than any other batted ball type, and more fly balls are hit to center than either corner by such a wide margin, center fielders also lead the corners in easier putout opportunities; 9,391 sub-.300 xWOBACON flies were hit to center, but neither corner cracked even 7,000. Whether it’s to ensure a snag on a routine play or to give themselves the best opportunity to squeeze a tougher hit, teams ought to play their best outfielder in center.

How should they decide who plays each corner, then? Directional OAA is one way. Additionally, right fielders get more opportunities on flyballs, edging their left field counterparts by at least 4% whether you look through the spray angle or ultimate-fielder lens. On the other hand, while line drive rates are always within 1%, and there are too few outfield popups (73) to really say, left fielders seem to get more ground balls. So it’s settled: after choosing your center fielder, put your best grounder-fielding outfielder in left and your best flyball-catching outfielder in right.

But this entire time I’ve talked about putout opportunities and glossed over assist opportunities. Moving from first to third on a single is usually much easier for a baserunner when the right fielder is making the throw from a longer distance. As a result, right fielders will get more and longer throwing opportunities. So typically, after placing their best overall outfielder in center, teams will put their best arm among the rest in right. Traditional baseball has already gotten one win today, as it favors the center fielder argument. Can it get two?

Back in the 2000s at The Hardball Times (RIP), John Walsh evaluated outfield arms using five important situations: a single with a man on first and second unoccupied; a single with a man on second; a double with a man on first; a flyout with a man on third and less than 2 outs; and a flyout with a man on second, third base unoccupied, and less than two outs. Walsh based his overall evaluation of an outfielder’s arm on how often runners were thrown out, held, or advanced in all of these situations for any given fielder. He then compared this number with league average rates.

Now, we have publicly available metrics like rARM and ARM, which are components of Defensive Runs Saved and Ultimate Zone Rating, respectively. Since those metrics are proprietary, we don’t know exactly how they are calculated, but while Walsh’s five kinds of arm opportunities aren’t the only ones considered, they most likely make up a decent portion of the ones that are. Here is how they are distributed among outfielders:

And by spray angle:

Once again, corner outfielders ended up taking a big bite out of center field chances. In some cases, the corners might take the opportunity because their momentum is carrying them in a better direction for the throw. For example, on a single that’s hit to right center where a runner is trying to move from first to third, a center fielder’s momentum is carrying him in a direction almost opposite third, but a right fielder’s momentum is carrying him closer toward third.

Additionally, this discrepancy might be because 48.2% of these arm opportunities were after line drives, which had a very similar discrepancy between the spray angle distributions and the ultimate fielder distributions. This in turn might be because center fielders make an initial attempt at the ball but don’t come up with it, so the corner outfielder in question is listed as the ultimate fielder.

Either way you slice it, right fielders get more throwing opportunities than left fielders. And even if center fielders actually could get more throwing opportunities by more aggressively cutting off singles or sitting back on clear hits, avoiding missed initial attempts, it still makes sense to consider efficacy on putout attempts over throwing ones. Throwing opportunities represent a considerably smaller proportion of balls in play than putout opportunities, and while there is some overlap, there were 30,586 outfield putouts last year against only 16,258 arm opportunities.

So do teams actually place their best putout-ers in center and their best-of-the-rest arms in right? OAA, which evaluates fielders solely on the catch probability of their putout opportunities, shows that on a rate basis, center fielders are clearly superior. Among the 240 outfielders with at least 250 attempts — roughly a season’s worth — since Statcast began tracking OAA in 2016, those who primarily played center netted 0.015 OAA per attempt, the only positive figure. Right fielders and left fielders respectively cost their teams 0.002 and 0.011 outs per attempt. OAA doesn’t consider grounders fielded by outfielders, so this doesn’t tell us whether teams are putting their best-of-the-rest grounder-fielding outfielders in left. But the superior OAA of right fielders indicates they may be putting their next best air-ball outfielders in right.

As for my evaluation of arms, I took the total rARM and ARM at each outfield position and divided by the total number of arm opportunities (at that position). While not perfect, this served as an approximation of a rate-based measurement of arm skill:

The results were inconclusive. Left fielders had easily the best arms per attempt when looking at rARM, but according to ARM, they had the worst. When I upped the requirements for inclusion to at least 20 arm opportunities per fielder at each position, right fielders ended up being the worst across the board, and center fielders the best:

Both rARM and ARM try to account for the level of difficulty associated with each opportunity, so even if right fielders have tougher opportunities on average, this shouldn’t be a major factor. Maybe teams just primarily care about optimizing in accordance with putout opportunities, not arms. After all, no one outfield position has as clear an edge in throwing opportunities as center field does when it comes to putout opportunities. Or maybe there’s another explanation.

To come full circle: do right-handed outfielders perform better per throwing opportunity in right? And should handedness be a more salient concern to teams when evaluating the positioning of their arms than the amount of opportunities at the position itself?

Maybe it should be. Right-handers performed better across the board but racked up the most extra outs in right. Lefties, on the other hand, performed best in center but, of the corners, fared better in left.

Why is this the case? My guess would be angles. Anecdotally, I’ve noticed that most outfielders make plays on flyballs and line drives to their backhand; such a catch sets up a right-handed right fielder or a left-handed left fielder for an easy transfer that won’t require a spin to make a throw to the infield.

About 2,000 words later, this finding gels nicely with the idea that outfielders move more deftly to their gloveside. Right-handed right fielders and left-handed left fielders get better throwing angles and have the opportunity to cede more armside fly balls to the center fielder. If teams decide to implement these findings and employ more righties in right and lefties in left, the role of the center fielder becomes even more important, and their putout-savvy more crucial, due to the extra ceding situations. This is in addition to center fielders already receiving the most putout opportunities in their vicinity. But once a team chooses their center fielder, in the absence of more conclusive evidence on outfield throwing frequency, handedness might be the single most important factor in deciding who plays the corners.