It’s died down a bit, but last season there was a lot of hand-wringing by some baseball folks about “Analytics” gaining too much power in baseball. I want to discuss this topic and hopefully convince you that both “gut feel” and analytics have an important place in the game. In case it’s not obvious, I have no gut feel, having played exactly 1 inning of baseball in my life and being somewhat recent to view the game from the inside. Still, I see the limits of analytics.
Why Analytics Are Necessary
Putting it as simply as possible, our eyes lie to us. Routinely. Our interpretation of events is typically biased by our current reference frame (e.g. where you are sitting, what is happening in your life, etc) as well as the narrative we’d like to see proven. I sometimes say “Confirmation bias is one of the most powerful forces in the universe.” Understanding that about one’s self is key to being effective in the game, yet I find that very few people in general understand this human tendency.
I live near the Idaho border. There are some real things at play here: the region of Idaho nearby is more rural than where I live and folks from Idaho visiting the big city are often impatient due to being used to only competing with cows for the road. Still, my perception of Idaho drivers being maniacs is fed by confirmation bias. If I see a Ford F-250 Pedestrian Mutilator on the road, and especially if it’s moving fast or recklessly, I check for an Idaho plate. If I see one, I chalk one up for “Idaho drivers are maniacs.” If it’s a Utah plate, I promptly forget the whole thing.
I believe this is human nature and most humans don’t recognize it. I’m convinced the world would be a better place if they did.
Like all baseball teams, there’s a dad that hangs around my son’s team to provide the benefit of his immense knowledge of the game (and other games). I call him Uncle Rico. I’d been friendly with him for a long time and he spent quite a bit of time throwing balls to my son and others on the team. His hitting philosophy is firmly rooted in 1986 and he believes there is a “right” and “wrong” way to swing.
Last season, my son was struggling at the plate. He eventually traced the issue to lunging, but Rico was sure that Sam had one of those new “launch angle” swings and was not “flat through the zone”. He sent me a video to prove it. I was listening. The video contained 2 frames with the bat in the zone. In both frames, the bat was at the same height. I pointed that out. He continued to talk about the need to be flat through the zone. That was the last time I spoke with that guy. He wasn’t interested in objective information, but only in defending 1985 hitting. His politics are similar, but I digress.
I’ve seen a lot of high school coaches say “that fastball cut” when it lands glove side or “that slider backed up” when it lands arm side. A Rapsodo or Trackman will tell you that it did no such thing. Analytics, whether they be measurements, metrics, or models, don’t care about your preconceived notions. They are objective, and that is why they are valuable. Unfortunately, they can also be wrong, which leads to …
Why Experience Is Important
Technology, especially when it is new, can be flawed. Unfortunately, many are bamboozled by new technology and place more faith in it than they should. That’s where the old school has a leg up.
Many, and maybe most, baseball guys I’ve met are very skeptical of any analytics information that clashes with what their eyes and experience have taught them. I have little doubt that this tendency is fueled by at least one example of analytics being wrong, which they will never forget.
My favorite example is the early days of Trackman spin rate measurements and their use by the Yankees. They noted a correlation between fastball spin rates and strikeouts and doggedly pursued more spin and high targets. Unfortunately, the picture was not complete at the time, since gyro spin had yet to be measured. Gyro spin does not lead to whiffs high in the zone, and a pitch with a lot of gyro plays better low in the zone. We know that now, but didn’t then. The analytics guys were acting on incomplete information. I’m sure the old school guys enjoyed this failure.
Secondly, there is the fact that we are dealing with human beings. Analytics may show that an athlete is performing sub-optimally, but the coach has a much better understanding of the cost of correcting the issue as well as an understanding of other issues that may be more important. Many coaches fear that a player may become fixated on metrics rather than performance. Focus on metrics may eclipse the desire to compete.
The Sweet Spot
Analytics and experienced coaches each have information. And more information is better, right? Not always.
The internet came along in the 90’s and immediately provide us with more information than we’d ever had access to before. Quickly, those of us around at the time realized that information and correct information are two separate things. Filtering out the bad became critical.
If you don’t understand what I mean, Google (or better yet, DuckDuckgo) “how to improve your hitting.” There is tons of information on hitting on the internet. Most of it is junk. The power comes from being able to select out the good stuff.
I teach measurements, and one idea that I constantly push is that one should never trust a single source of information. I believe that is also true in baseball. If a coach has an idea, respect their opinion and experience, but check the idea based on available data. Understand your tech sufficiently that when it says something surprising, you can determine if it’s something new, or simply a bad reading. Quite often, metrics have assumptions baked into them that need to be understood to be properly interpreted. If you know you don’t understand these nuances, do not blindly follow the recommendations based on that metric. Know your limitations.