The data on Alejandro Kirk are clear: he’s the slowest player in Major League Baseball. Kirk’s so slow that despite being twenty years older than him, I’m (mostly) confident I could beat him around the bases. Admittedly, I could never hit a 98mph slider, which is a prerequisite for running the bases, so that hypothetical comparison can only reveal so much.
(If the world wants to prove its goodness, some thoughtful reader will pass this along to someone who knows Alejandro, and before you know it the Captain and I will be tearing neck-and-neck around the bases.)
In 2016, the Toronto Blue Jays sent a scout to Tijuana, Mexico, to observe players. The scout was supposed to review Catcher #1 but couldn’t take his eyes off Catcher #2: Alejandro Kirk. The scout had a feeling about Kirk, specifically that Kirk could “slow the game down,” which is baseball talk for acknowledging that Kirk could keep his head during difficult situations. For obvious reasons this is a very desirable trait, especially in catchers, who are generally their team’s on-field captains.
The problem with the scout’s feelings, however, is that they were feelings. It’s easy to run afoul of feelings. They’re notoriously squishy and the data on them are mixed. The best we can determine is that 100% of us have them. Everything else is a mess.
When the scout reported back to the Jays, he noted the obvious: Kirk had an impressive foundation of skills, but at only 5’8”, Kirk didn’t look like an elite athlete. 5’8” is data. Here’s more: I am a thoroughly average sized human male. Kirk, however, is not. He’s two-inches shorter than me and eighty-pounds heavier. Depending on the day, those eighty-pounds make up half my body weight. You can plot this data and use it to calculate more data, such as our respective Body Mass Indexes. I’m quite normal, while Kirk is considered obese.
Despite the literally obvious data on Kirk’s physique—one prospect profile gave him a body composition rating of “Chris Farley”—the Jays rolled the dice and signed Kirk for $7,500. When asked, the datanet noted that with $7,500 I could purchase a reliable used car, a laptop for gaming, a weeklong vacation package, or a high-resolution television. It also suggested investing in original artwork and crypto, both of which should make you question the algorithm’s sanity, if you’ll allow such categories to co-exist.
Kirk did well. He sped through the minors and made his MLB debut in 2020. Since then he’s twice been an All Star and won a Silver Slugger award. He’s also the one of the main reasons the Blue Jays almost—almost!—won the World Series yesterday. Sadly, Kirk made the final out, which is just more data for you to remember.
Most of us aren’t professional athletes, but we’re all wading through a world saturated with data. Some of the data is useful, though if we’re honest most of it is simple description that lacks both punch and relevance. In other words (cover your eyes, my tech friends): most of the data we’re aslog in has some obvious limitations. One of those limitations is predicting the success of the Alejandro Kirk’s of the world.
If the data on Kirk had been followed, he never would’ve been hired because he didn’t fit the data-determined demands. If you’ve applied for a job recently, it’s likely you can relate to this. From start to finish, the act of finding work is awash is data. Gone are the quaint, olde-timey days of simple data collection (ten women applied for this position; four of them had BA’s, five had AA’s, and one only her GED); instead, we’re in a new and largely-un-brave world where artificial intelligence not only writes many job listings (to say nothing about coding the platforms on which those jobs are listed), but increasingly does the same for your resume and cover letter.
Step back and think about that for a moment: algorithms create the employment buckets that algorithms subsequently place you inside of.
Uff.
For help on this I turned to—more data. When I asked the datanet how I should think about the above, it acknowledged that “this concept can be difficult to unpack.” Here’s where I can finally disagree with the data: this concept is many things; difficult to unpack is not one. There’s nothing confusing about this process, though we should be stung with questions about the approach’s reasonableness, validity, usefulness, and effectiveness. We should also wonder about time, for if data write the job listings that data subsequently write the resumes to fill, what the hell are the rest of us doing?? In other words: if data are doing all the work, why’s everyone so damned busy?
This could go many different directions, but let’s round the bases and head back home to Alejandro Kirk. If the scout who saw Kirk play had followed the data instead of his feelings, it’s unlikely Kirk ever would’ve played major league baseball. The Blue Jays would have a different catcher. If data can be trusted he’d stand about 6’1”, and would definitely beat me in a race around the bases.
That may feel a little like proving a negative, and you might find yourself wondering: Who cares? For starters, I bet Alejandro Kirk does. I’d care, too, because I enjoy watching Kirk play (I’ve never seen so many three-hundred foot singles!, and that sprint from second-to-home against the Mariners will live in infamy…). Further, without Kirk in the majors, I’d never be writing this essay.
On that latter point, you may find yourself wondering: How can you be sure data haven’t written this essay? Great question. To prove (or save?) my humanity I’ve peppered this with baseball images (round the bases, run afoul), improper grammaticals (ending sentences with prepositions), and neologisms (aslog, datanet, grammaticals), all while knowing that it’s only a matter of time before data are trained to incorporate such details….
Data can be very helpful in some situations, such as when you have a spare $7,500 and need to purchase a used car (the data suggest a 2015 Honda Civic). But there are plenty of times when data are just noise. In those situations it’s best to slow down the game of your life, ignore the superfluous data, and instead listen to your feelings. For some of us that proposition might be scary: as mentioned, feelings aren’t easy to quantify. But outside of certain descriptive parameters, neither is life.