Outfield Training Economy. Should we just focus on reaction?

I say yes, and here is why.

A while back I read this article in Fangraphs that got me thinking if I was doing right by our outfielders. You see, I was basically splitting our outfield practice time almost equally. Routes, reaction, glove-work, speed, they all got roughly the same amount of time and reps. But the article opened my eyes to the fact that maybe some things are more important than others (specifically reaction and burst), and therefore deserve more time. I wanted to know a little bit more about the trade-offs of prioritizing one over the other, so I used it as an opportunity to keep working on my basic data analysis skills.

Let’s start with Outs Above Average (OAA). OAA is the cumulative effect of all individual Catch Probability plays a fielder has been credited or debited with, making it a range-based metric of fielding skill that accounts for the number of plays made and the difficulty of them. For example, a fielder who catches a 25% Catch Probability play gets +.75; one who fails to make the play gets -.25. OAA is how we measure outfielders today.

Catch Probability is the stat that expresses the likelihood for a ball to be caught by an outfielder based on opportunity time, distance needed, and direction.

Based on the ability to quantify each play’s Catch Probability, the following star ratings were created. Plays with a Catch Probability higher than 95 percent don’t merit a star rating.

5 Stars — 0 to 25 percent

4 Stars — 30 to 50 percent

3 Stars — 55 to 75 percent

2 Stars — 80 to 90 percent

1 Star — 95 percent

Tom Tango explains it better with this player comparison:

As you can see, Isbel, although not as fast as Acuña based on Savant’s sprint speed, is quicker to react. And although he takes worse routes, he is a much better outfielder than Acuña.

Kyle Isbel:

Ronald Acuña Jr.:

While route efficiency is important, as Tom said in his tweet, it’s secondary to reaction. An immediate and quick reaction can make up for minor inefficiencies in the route. However, if the reaction is slow, even a perfect route might not compensate, especially on line drives or hard-hit balls.

So, perfect routes, not as important.

Still, when deciding what to practice and how much time and effort dedicate to it, one must also think about how easy it is to move the needle for this attribute. Some things are harder to train than others, and while they might represent the bigger upside, if they are time consuming, maybe we are better off improving the variable that is most easily improved, regardless if it has a lesser impact.

Let’s look at this example:

Some players who have high OAA can struggle with 4 and 5-star catches (those are the potential doubles and triples), either by having bad routes or bad reaction, for example:

High OAA + Bad reaction + Good Routes: Cedric Mullins (29) — 21% of 4/5 Star plays made

High OAA + Bad Routes + Good Reaction: Leody Taveras (25) — 12% of 4/5 Star plays made

So, let’s assume this off-season, they both want to get better at catching these very important hits that are likely to become runs against their team.

As you can see, Mullins needs to work on Reaction, while Taveras needs to work on Routes. But, what would be easier? And how would those improvements impact their capacity to make the tougher outs? But more important: is catching these outs the most impactful thing an outfielder can work on in terms of improving his OAA?

These are the questions that we should ask ourselves in Player Development, when trying to design and plan for practice. Practice time is a valuable asset, it is finite and it needs to be constantly optimized.

The last question, is the easiest to answer. Catching 4/5 star outs has great impact on overal OAA score.

In terms of who will have it easier, if Mullins improving reaction, or if Taveras improving routes, I think we should consider age as a factor. OAA is made of 3 variables, all of which are related to each other, and are related to age. Before we dive into this, let’s address the elephant in the room: yes, we might run into survivorship bias, as the game itself might filter out aging and slower players, or move them to other positions. Nevertheless, age is a factor. 

Up to a certain age, some abilities seem to start a decline. Burst shows an early and steep decline from age 21, and it’s obviously correlated to sprint speed, which of course declines with age. But reaction holds up well up until around age 28 (and it’s not as correlated to speed as burst is), right about the time when route efficiency seems to help cover the gap of age declining skills.

Although, like Fangraphs showed us, route efficiency doesn’t correlate as well to OAA as burst and reaction, there is a certain age where it’s unlikely that a player will improve in either of the two.

Mullins enters his 30 year season next year, while Taveras enters his 26. It’d be easy to make the case that Mullins will have a harder time improving his reaction given his age, and that Taveras, maintaining his current reaction times and improving his routes (a much more trainable and naturally evolving attribute) is likelier to happen. But remember, reaction carries more correlation to OAA than routes.

So, what to do with Mullins? Is it impossible to improve reaction or burst past your 28 years of age? It is not. In fact, Cristian Yelich (31), Mike Trout (31) and Seiya Suzuki (28) improved both reaction and burst between 2022 and 2023, improving their OAA rankings. Yelich went from -5 to 4, Trout from 3 to 4 and Suzuki from -4 to 2.

Yelich had his best defensive season since 2016. Even his throwing max velo from the outfield jumped from 88 to 95 and his average from 81.8 to 84. Trout, who in 2023 played 5 feet deeper in CF, had his best season since 2018 in OAA but also third consecutive season improving this metric. Suzuki is only in his second MLB season, but his arm is already stronger than last year

Brandon Nimmo (31) on the other hand, had a 0.2 improvement in routing, just a 0.1 in reaction and a 0.5 drop in burst, making him go from 6 OOA in 2023 to 1 OAA in 2022.

Yelich worked with Quintin Berry on improving his outfield defense in spring training. In an interview for Brewers Weekly, he spoke about how they kept it simple and just about refreshing concepts. Trying to be locked in during games and aware of pitches and swings, although he highlighted how hard it is at American Family Field with all the windows and shadows. Quintin discussed how they purposely shortened his arm path to make his arm action quicker, and also an emphasis on increasing his throwing volume, throwing every day.

Yelich came into ST wanting to improve his defense. He made it a priority. Was consistent with this goal throughout the season and it showed.

Suzuki on the other hand got in the gym and gained serious muscle. And it shows in his increase in burst.

Although Yelich, Trout and Suzuki saw a decline on route efficiency, their increments in reaction and burst helped them improve their OAA rating from 2022.

However, each of these attributes have different complexities when it comes to developing them. Based on my year-over-year analysis, we can observe that both burst and reaction have an earlier and steeper decline, with burst having more variability in terms of improvement/decline.

The average year-over-year improvements for each variable among players with two or more consecutive seasons are:

• Burst: −0.144−0.144 (a decrease on average)

• Reaction: −0.080−0.080 (a decrease on average)

• Routes: +0.040+0.040 (an increase on average)

So:

• Burst: Although it shows high variability in year over year change, it’s development is pretty straight forward and it happens mostly, off the field. This is where strength and conditioning come into play. Speed and first-step quickness are often developed in the gym through plyometric exercises, sprints, and strength training focused on lower-body explosiveness.

• Reaction: Although it shows a negative average year-over-year change, as we saw with older players, it can be improved, and it’s improvement takes place, mostly on the field through situational practice, live-game experience, and visual reps. Coaches work on improving a player’s initial reaction to the ball off the bat, often using drills that simulate game situations.

• Route Efficiency: The only one of all three metrics with a positive trend year over year. This is also developed mainly on the field, but as we saw, it is mostly the product of acquired experience and also a consequence of slow reaction. However, it can be worked on, and coaches often use specific drills to improve route-running to the ball, including techniques for efficient turns and angles. Video analysis is also commonly used.

So, if burst is something improved mostly off the field, and route effficiency is a by-product of either late reaction or playing experience, by focusing mainly on reaction, which we saw is trainable, less dependant on age than burst, doesn’t naturally improve overtime, and correlates well to OAA, we are helping our outfielders become better everyday. 

So, how can we improve our reaction in the outfield? Here are some things that I think we can leverage:

Training for reaction time can involve various drills and settings, also varying intensities. Working with outfielders can be tricky because you cannot do endless amounts of reps, since those tax the player’s fatigue more so that’s other activities, say like hitting or fielding routine ground balls. There is a large difference in the number of reps that you can do without taxing your body and readinesses to compete, when comparing outfield defense to other skills. Outfield work requires you to run, cover distances, and sometimes do it at high intensity.

• Working on pre-pitch timing is something that can be done almost daily. The ability to hit the ground as the ball hits the bat, which primes you for a quicker reaction (as seen in tennis players returning the opponents serve).

• Working on positioning your body the right way as soon as the ball is hit is also crucial, for reaction and also route efficiency. Most times, routes are bad because of a bad first step. And this can be done even without reading the ball of the bat. I like working reads and directions sometimes even with just tossing a ball and having the player move right or left depending on which hand I catch it with.

• Doing games/ competitions that involve quick decision making and visual queues. I like this rock, paper scissors drill where whoever loses has to turn and run away, and whoever wins has to tag him. Super challenging. Here is Jose Siri, one of the best outfields in the league at making 4 and 5 star outs, working on one of these drills.

• Fungo: when using fungo, you have to mix it up. I believe fungo is sometimes even better than machine work for the outfield, cause it provides the element of reading the ball off the bat. But mixing it up is important. Do drop ball fungo, hit it with slice, hit it with top spin, pull it, hit it from the left side, and hit it on a line drive. Line drives are particularly hard to read and can potentially be doubles or triples if miss played.

• Machine: I use machine work mostly for burst drills. Machine can’t simulate reaction unless you have 2 machines and even still, it reduces the options to 2 directions and players do a lot of guess work in these practice scenarios. I like using machines to simulate different spins though, for that they are perfect. Lately I’ve been using them a lot with different materials, size balls, since they create different spins, speeds and can create more variability this way.

• Live reads: this is, by far, the most important thing. Utilizing live BP scenarios to get real reads off the bat. And I used different settings for this:

1. Straight up catch the ball. No constraints, just be a dude, make the play◦ Simulate runners: call out scenarios with runners on and have them position themselves behind the ball to make a throw

2. Box them on a square made of cones: this was my favorite and also the player’s. You surround the players with 8 cones, and they have to exit that square as fast as possible as soon as the ball is hit. It causes them to sometimes choose weird routes, but it gets them going, and as we saw before, it is more important to react fast than to wait for a good read. Also, the cones provide a visual feedback component, they can see where they exited and see if their initial steps were good or not.

3. Make them call out where the ball is before they can move. This will have them making a decision immediately, wether the ball is in front of them, above them or behind them, and then getting going and figuring out if they were right or not. It is tricky and unconventional, but the purpose is adding an extra degree of difficulty and making those initial reads quick and accurate.

4. Set up teams and make it a competition: set up rules on the spot. 3 points for warning track catches, 2 points for diving plays, 1 point for routine outs. Whatever it is, make it fun, and make it challenging.

Finally, video analysis is of great importance. Going over practice videos can help spot out movement defficiencies that can be tackled hollistically.

There are many other new methods that deserve consideration, like using Stroboscopic Training (wearing special eyewear that intermittently disrupts vision can train the brain to process visual information more efficiently), Neurofeedback (monitoring the brain’s electrical activity while performing tasks. Outfielders can learn to recognize and alter their brainwave patterns for improved focus and reaction times), VR training, Reaction Light Training, etc.

And of course, this analysis is missing a critical component. Not every outfield position is the same, not every direction in which the ball is hit has the same complexity. A dive into those realms is necessary to keep evolving in a field of baseball defense that it’s often considered as the least important, or the one where the least can be done in development. 

I was fortunate enough to be taken out of my comfort zone when I was asked to work with outfielders, and not only it is an intriguing part of the game, I think there is much more room for growth in it.