Somewhere at my parents’ house, there’s a Starting Lineup figurine of Jose Canseco, depicting him during his Bash Brothers days with the Oakland A’s. I got it for Christmas one year back in the days when Jose Canseco was my favorite player. I would have been nine or ten at the time and he was … let’s just say the words “Jose Canseco” evoked a different image back in the late 80s/early 90s than they do now. Canseco had won the 1986 Rookie of the Year award at 21 and the 1988 MVP at 23, hitting 40 home runs and stealing 40 bases in the same year. At the time, Canseco seemed like the guy we would all look back on some day and tell our kids that we saw him play.

I have yet to mention Jose Canseco to my daughters.

For those who don’t remember Starting Lineup figurines (they haven’t been produced since 2001), they were small action figure-like toys that were in the form of pro athletes. They didn’t really do much other than stand there and look like Jose Canseco or Kevin Mitchell or Pat Tabler–if Canseco and Mitchell and Tabler were plastic men who stood about four inches tall. I had a few of them, but Canseco was the pride of the collection, because he was the best player in baseball and he was hard to get. The figurines were mostly sold regionally and I lived in Cleveland, so at the local toy stores, there were plenty of Tablers, but no one who played in another market. Fortunately, my father happened to “know a guy.”

I wish I still had that Canseco figurine. And the Tabler one too. Not because they’d make me rich, but because I think they might actually help me to be a better researcher.


Regular readers of my column know that my formal training is in child and adolescent clinical psychology. I no longer see patients, but during my graduate training, I did. Most of them were kids, and for a while, I had plans to become a practicing child psychologist. At some point, I realized that while it’s a very good and noble career, it is the wrong career for me. Still, there are a few good lessons to take from those days. “Therapy”–especially when working with kids–can take many forms, and perhaps not the form you might be thinking of. Most people think of therapy as a place where one talks to a counselor about whatever’s going on, and that certainly is one form. The thing about kids is that they don’t always have the vocabulary to talk in the same way that adults do. Kids are more likely to express themselves through playing.

If you go into the office of a child therapist, there’s a good chance you’ll see a dollhouse, strategically stocked with toys that might evoke playing out particular scenarios. In fact, half the time, the therapist puts things into the dollhouse before a session to encourage a child to engage a specific issue. A child might not be able to talk about the parent who yells too much, but if you see them using the doll family to play out scenes where a parent is always yelling, it’s probably their way of “talking” about what’s going on at home. Sometimes, it’s just a more comfortable way to bring up the topic, because they can say that they were “just playing” if addressing it out loud is too overwhelming.

Adults do this too, by the way, just in more sophisticated ways. (“Oh … I was just joking.”) Rule no. 1 of child psychology is that it applies throughout all of life.


There’s something interesting about play that we often don’t give enough credit to. Play is a great way to explore new ideas. Marla Montessori (yes, the one for whom the schooling method is named) was fond of saying “play is the work of the child” and there’s a good reason for that. Kids don’t have a wealth of experience to draw from. Even simple social situations are new to them, so they need to rehearse what they’re going to do out in the real world. In the dollhouse or the sandbox, you can set up whatever situation you like and play it out a bunch of different ways. You can get feedback on which responses work best.

The good therapist has a wide range of items that might be used for play therapy. There are always the human dolls, but sometimes kids will prefer to play with toy animals instead. It might give them a little more psychological distance, but it might also serve as a metaphor for a specific character trait in a person. A tiger might be a stand-in for a ferocious person. A dog might be a playful and loyal person. Pat Tabler might be a stand-in for a guy who inexplicably sticks around for 11 years and yet only puts up 3.0 career WAR.

When one plays in a dollhouse or a sandbox, you have to think things through for all of the characters involved. What would the dinosaur think about this situation? What would Abe Lincoln’s response be to the dinosaur? It’s an exercise in perspective-taking, which is a skill that just about everyone can use more of in their lives.


We live in an era of screens. You’re looking at one right now. And there are plenty of people who will tell you that screens are evil and ruining the world (or that they are amazing and will save the world). I will spare you both sides of the issue because neither is true. What is true is that computers do change that way in which we engage information. There’s wonderful research that’s been recently published about how people solve problems on screens versus the way they solve the same problems on paper. Researchers had participants play a game in which they had to design a strategy to contain an outbreak of a simulated deadly virus. The game could be played either on a computer screen or on paper. What the researchers found was that the screen players (who had been randomly assigned) tended to focus their efforts on responding to the immediately present “cases” of the virus, which turns out to be a different strategy than one that actually limits the spread of the virus.

The people who received the paper version of the game (again, randomly assigned) were more likely to hit on the right answer. The researchers chalked up the discrepancy to the idea that screens tend to prime people to look for quick answers, rather than to deeply engage the problem. Fans of Daniel Kahneman’s work on the two “systems” in the brain for thinking (fast and slow) are probably already nodding.

It’s one of the nice things about computers. You can do a lot of #GoryMath quickly. In baseball, it’s led to people asking questions that prior to the computer age, couldn’t have been solved with just pen and paper. But that speed comes at a cost. I can come up with an interesting question, represent it mathematically, and have the answer to that question within a fairly short amount of time. But the cost is that it’s easy to focus on the fact that “I got an answer!” But unlike Abe Lincoln in the dollhouse, it doesn’t leave a lot of time to take the dinosaur’s perspective.

To take a simple example, there is something of an eternal battle within the sabermetric movement around the use of the closer. There are, mathematically, better ways to use the best reliever in your bullpen than to protect a three-run lead in the ninth inning, but no team seems to be willing to try even minimally invasive improvements. The standard story is that the “new” models of closer usage require that a pitcher will be used at different times in different types of games, and that pitchers prefer certainty about those sorts of things. It’s not the sort of thing that shows up when you plug away in a spreadsheet. This may not be efficient, but it is reality.

The thing about baseball research is that it’s most often mediated through a computer screen, and yes it’s something of an old critique that maybe researchers ought to pull away from the spreadsheet once in a while. There’s something to be said for that. Yes, you can figure out the mathematically optimal bullpen usage quickly, but why not slow down and think about how the dinosaur feels about pitching in the seventh inning. (Yes, I know that you can tell him to shut up and pitch when he’s told. He might bite your head off.)


Play is a slow process. It’s the opposite of the computer. But it draws a person into a different space and a different way of thinking than does working on a computer. If working on a screen encourages quick and immediate thinking, play requires that a person consider a situation more deeply. Consider a team thinking of implementing a five-man infield. On a computer, it’s easy to generate the kind of batted ball profile and spray chart of why a particular hitter might be a good candidate for such an infield, but imagine using a couple of Starting Lineup figurines to represent the idea in the sandbox. Jose Canseco could be the batter while a few other stars of the 90s manned the five infield positions. Abe Lincoln and the dinosaur could patrol the outfield. In setting up the infield though, you have to envision where the five plastic men would actually stand. Would they overlap? If the ball were hit to one of them, would he likely be fielding the ball in a “weird” position on the field? What about the batter? Looking out what would he see, and what would be the logical thing for him to try to do? Maybe he’d just try to hit a fly ball toward the outfield.

In reality, if a team were to actually execute a five-man infield, the physical act of setting it up would take 30 seconds. To play it out might take 30 minutes. Playing it out involves moving physical objects through space, moving around to physically take the vantage point of various points on the diamond, and “stopping time” and thinking about what might happen. Quite the opposite of a machine that can simulate an entire season of games within minutes. Play engages multiple senses and forces the person playing to slow down and engage in some of that deep thinking. It’s a nice counter-balance of forces, each with its own strength.

Yes, all ideas have their downsides and yet some are still worth it. The point is that if you slow down a bit, you have a better chance of thinking of some of those downsides and preparing for them.


A proposal: I think all baseball researchers should spend some time in the sandbox. Find some old Starting Lineup figurines and maybe a few other random action figures from back in the day. Bring a Superman one because at some point, you’ll construct some weird hypothetical with a Superman-like character playing center field. We all dream of trading for Mike Trout.

Create a small baseball field for them to stand on and some representation of a house or some living space, because players do that too. And once in a while, test out your ideas on this pretend field. Maybe teams ought to have little sandbox annexes next to their Stat Caves, too. It’s better than thinking up these crazy ideas and then trying them with actual real people, although that might make an interesting book. You aren’t going to prove whether your idea is brilliant in the sandbox, but by engaging with whatever problem you are working on in a different way, you might have a chance to see nuances in the problem that you hadn’t considered.

It might seem strange, because it’s decidedly low-tech, but technology will only get you so far in baseball (and life). I think the next time I’m visiting my parents, I’ll see if I can find those old Starting Lineup figurines in the attic and bring them home with me.

Thank you for reading

This is a free article. If you enjoyed it, consider subscribing to Baseball Prospectus. Subscriptions support ongoing public baseball research and analysis in an increasingly proprietary environment.

Subscribe now
You need to be logged in to comment. Login or Subscribe
I love this. I've got an old Al Toon that can surely handle LF.
Russell, it feels like you've found another axis around which to spin a "scouts vs. stats" distinction: quick vs. deep analysis. What a terrific article.