September 23, 2013

Baseball Therapy

If One Win is Worth $5 Million...

by Russell A. Carleton

the archives are now free. All Baseball Prospectus Premium and Fantasy articles more than a year old are now free as a thank you to the entire Internet for making our work possible. Not a subscriber? Get exclusive content like this delivered hot to your inbox every weekday. Click here for more information on Baseball Prospectus subscriptions or use the buttons to the right to subscribe and get instant access to the best baseball content on the web.

Subscribe for $4.95 per month Recurring subscription - cancel anytime. Subscribe for $39.95 for one year a 33% savings over the monthly price! Purchase a $39.95 gift subscription a 33% savings over the monthly price! Already a subscriber? Click here and use the blue login bar to log in.

Lately, there’s been a lot of writing among baseball analysts and thinkers about how teams might leverage small investments to their benefit. You know the ones. They all start with “Well, if a win is worth five million dollars, then spending fifty thousand on this has to net just one hundredth of a win to be worth it.” (I’ve written a few of them myself.) Then they continue with an exasperated tone, wondering why teams don’t seem to do much of this sort of thinking. (Or do they?)

The reason can be summed up in four short words: “It’s not that simple.” It never is.

It’s such a tempting argument to make, because changing one strikeout to a walk is worth something like half a run, and if you can do that a grand total of once each year for each guy in the lineup, the net benefit is half a win. And signing up everyone on the team for the Coffee of the Month club costs only $19.95 a month. All it takes to justify that expense is a quick explanation of how the coffee will give everyone on the team a little extra focus, and that once in a while, it will allow them to be sharp enough to avoid swinging at ball four on a 3-2 pitch.

It’s entirely possible that some of the ideas that have been floated out there really would be cheap ways to make teams better, and substantially enough that they would be worth the cost. But a simple model of anything often ends up being overly simple. As a vehicle for illustration, allow me to review one of the most harebrained proposals of all, which has the added benefit of being one of my own. I wrote about it a few months ago, suggesting that teams should, instead of providing starvation wages (sometimes literally) in the form of per diem to minor leaguers, they should instead invest (i.e., spend a lot more money) on providing full catered meals to their players at all levels of the minors, both at home and on the road. I call it the “Food for Kids” program.

Let’s examine “Food for Kids” with a more critical eye, examining of the issues that have to be overcome after we’ve had the crazy, yet brilliant idea.

That’s a lot of money…
In my original piece, I suggested that the cost of providing catered meals over the course of a year would be something on the order of $1.3 million per year. The exact number isn’t as important as the order of magnitude. It would be something around a million dollars. Believe it or not, even in baseball a million dollars is still a lot of money. Yes, I know that a million dollars buys you a 32-year-old utility infielder who hits .250 with barely passable defense (if you’re lucky) and that even the low-budget teams carry payrolls around 80 million.

I once had a conversation with my brother, who has a master’s degree in finance and put things in a different light. He pointed out that the business model of a baseball team (and a sports franchise in general) is awful. The large majority of your input costs (player salaries) are set in advance and are “hard” costs. You can release a player if he doesn’t perform, but you still owe him his salary until the end of the contract.

On the other hand, a good chunk of your operating income is based on ticket sales, which is variable income. You can estimate how many people will buy tickets, but when you commit to the payroll, those people haven’t shown up yet. And maybe they won’t. Worse, the thing that fans most seem to respond to is a winning team, and we know that there’s lot of luck in baseball. Go on a streak of bad luck, and suddenly, the fans stop showing up. Not only that, but the dollars that you’re chasing are discretionary dollars. The nice thing about running a grocery store is that even when times are bad, people still need to eat. They’ll cut back on taking themselves out to the ballgame instead. There aren’t big fat piles of extra cash sitting around front offices. The profit margin could disappear too quickly for that.

Could money be re-purposed from the MLB payroll into my “Food for Kids” program? Sure. But before you make such a big investment, there are some other factors to consider.

The Second Move Advantage
Imagine for a moment that you are playing the game Rock-Paper-Scissors. The key to a good game of RPS is that both players need to show their hand weapon of choice at the same time. Now imagine that instead of simultaneous movement, your opponent has to go first, then you get to pick which of the three you’d like to use. You’ve just experienced second move advantage.

Let’s first note that the effects of the “Food for Kids” program are a mystery, because no one has yet tried it. We don’t know exactly what would happen. It might make everyone in the system amazing. It might make them all fat. It might have some small, but meaningful effects that are big enough to make the program a success but not enough to shout about it from the housetops. Right now, I could make a reasonable case to expect modest, although meaningful results, but it’s a risk. A risk that costs a lot of money.

Now, let’s say that some other team tried it. It wouldn’t take much effort to figure out that they were doing it. (Every time we play one of their affiliates, a big catering truck shows up.) Plus, people talk. And if they’re getting good results, then we can simply copy them next year. They took the risk to find out if it works and since no good deed goes un-punished, they don’t get to reap the benefits from it for more than a couple of years or so. This is the difference between getting an absolute vs. getting a relative advantage. “Food for Kids” might make everyone in the system a better baseball player, but if everyone else can get that advantage once they know about it, then the team that started it returns to being no better off. Most arguments starting with “If a win is worth $5 million…” focus on the absolute advantage to be gained. To win in baseball you have to be better than (i.e., have a relative advantage over) the other teams.

If the program fails, then the other team is out the money, and everyone else gets to find out that it doesn’t work—for free.

So, everyone is stuck. It doesn’t make sense to be the first person to risk a lot of money on something that might not work unless you believe that in that short window where it’s just your little secret, you can reap benefits that make it worth it. Then there’s the fact that if it works, you know that other teams will copy it and will negate your advantage, but you still need to spend that million per year, just to keep up with the Doug Joneses. For a low-budget team, you’re down a million dollars on a payroll of 80 million while other teams are down a million on 200. Maybe it’s best not to start that arms race. For big-budget teams, why help the low-budget guys identify a cheap source of improving team-controlled talent, the kind of resource that they’ll need to compete with you, when you can just out-spend them?

Someone has to run that program
It often surprises people to learn that MLB front offices are actually rather small places. Not just physically, but when we speak about “the front office,” it’s actually a very small cabal of people making the big decisions. Sure, like any place of business, there’s the accounting department and the marketing department, but the baseball operations department isn’t usually teeming with people. It’s usually a small inner circle of mostly men. Some of that is probably by design. Too many cooks spoil the broth and some teams may feel there is more value in keeping the inner circle closed. But more than that, it’s not like teams can create new positions at will.

If a team were to actually implement my “Food for Kids” program and commit a million dollars to it, ownership would want to make sure that someone was watching that money. Plus, someone has to be in charge of making sure that the catering truck makes it to the park in Albuquerque. That won’t be the farm director. He’s got other things to do. So, you might hire a director of culinary logistics with the skillset to know about catering, negotiations with food service companies, and sports nutrition. But that comes with a cost in salary, benefits, and overhead. It might be worth it, but there’s always the tendency to see costs like that as top line expenses rather than investments.

For some other programs (“teach everyone the knuckleball!”), it might make sense for the farm director to oversee such a program, because it directly relates to what he’s doing anyway. However, it’s one thing to say that and another to sit down and write a program manual about the gory details of how it’s going to work, then implement the program. Going from crazy idea to actual practice means building an entire structure within a culture to support that idea, and that’s harder than you might think, whether we’re talking about baseball or real life. Front office personnel aren’t generally selected on their ability to design and implement systemic interventions. Some guys probably have that in their back pocket and maybe teams should pay more attention to that sort of thing, but it’s not usually in the job description.

I expected it to be… bigger
Let’s say that our team decided to green light the “Food for Kids” program. The next question would be “Is it working?” It’s a simple question, but it takes some doing to answer it. It might be easy to select some criteria for whether players are getting better. Does the team have the data infrastructure to measure those things and an analyst who knows how to operationalize that?

Past that, there’s another problem. About the best research framework you’re going to get in this case is a simple pre/post design. This is the amount of development that we saw last year before the program went into effect. This is what kind of improvement we saw this year with the new program. Was the improvement due to better nutrition? Was it random variance? Was it the new hitting instructor who makes all the hitters recite the Gettysburg Address as they walk to the plate? A baseball team is not the place to run nice, neat randomized controlled studies. Even a really good researcher is going to have to say “Yeah, I can’t be totally sure…” when evaluating these programs.

Then there’s the matter of effect size. Most of these proposals are framed as, “Well, all you need is a really small movement in strikeout rates and the program pays for itself.” Here’s the problem. I suggested that if each regular hitter replaced one strikeout with one walk, it would be worth half a win. The problem is that over 600 PA, one extra walk would represent a tick upward of 0.16 percent. Yes, walk rates are quick to stabilize, but that doesn’t mean that they are free from noise. Maybe the player’s true talent level really did improve, but answering that question with that small a change is going to require a really big data set (for maximum power) that might not be available. If you don’t have that, you’ll need bigger effect size. Things have to be going so well that it’s just plainly obvious that it’s working. “That could just be randomness” is such a damning critique because it’s true.

Maybe the “Food for Kids” program really would make a positive, if small, difference. But it’s going to be hard to show that it’s working and it might be a few million dollars spent on a big bust. Yes, teams do spend millions of dollars on ideas that don’t work (they’re called failed draft picks). But then front office folks are in the business of knowing about baseball. If they can mess up on something like a draft pick in their ostensible area of expertise, how much less willing will they be to take a risk on something that isn’t in their domain of expertise?

What kind of idea would work?
All told, these factors probably make baseball teams more conservative than they should be with trying new ideas, but also not as foolishly stubborn as they generally get blamed for. There are real structural obstacles to starting some new plan. (And we haven’t even talked about getting “buy in” from the coaches and staff if they’re the ones who are going to have to actually implement the ideas.)

Maybe part of the critique is that with all of these obstacles in place, teams are missing out on some killer ideas. But for now, let’s focus on what ideas will have a better chance of making it as things stand right now.

• It should be cheap (or better, free) to implement. The idea of the infield shift has become common, and more teams are starting to adopt it. The early adopters are starting to lose the advantage that it gave them relative to everyone else, but the costs associated with it are the costs of gathering a good spray chart for each hitter and telling the shortstop to stand over there. But let’s remember that cost is not the only thing in the way of getting teams to adopt an idea.
  
  
• It should be supported by some sort of previous data, whether that’s from baseball or not. The case for my “Food for Kids” program is based on plenty of research on sports nutrition generally as well as work on eating behavior more generally. But with that said, there’s little data from a baseball-specific context from which to draw. That makes it a bigger risk, because you have to start making assumptions. If you want to pitch an idea, assume that the burden of proof is on you.
  
  
• It should be hard to copy. That can be done through near-paranoid levels of secrecy (or just a plain old non-disclosure agreement). Alternately, if the idea involves a very particular set of skills that are relatively rare, then a team could simply corner the market on people who know how to do that. Ever wonder where most of the really good PITCHf/x analysts went?
  
  
• Most importantly, it should have the potential (perhaps “very strong likelihood” would be better) of producing a large effect in a short period of time so that whatever benefits are gained before everyone else catches on can justify the cost.

The idea doesn’t have to meet all of these criteria, but it had better meet a few of them. So, before you come up with a grand idea and are tempted to start a sentence with “Well, if a win costs $5 million…” be prepared to talk about some of the other issues that might pop up along the way to making it a reality.

Russell A. Carleton is an author of Baseball Prospectus. Follow @pizzacutter4
Click here to see Russell's other articles. You can contact Russell by clicking here