Baseball Prospectus is looking for a Public Data Services Director. Read the description here.

The Cubs really like their president, Theo Epstein, and according to reports just gave him 50-ish million reasons to like them back. On top of that, they also gave his friends (general manager Jed Hoyer and senior vice president of scouting Jason McLeod) a few million reasons each to stick around the North Side as well. The thing about Epstein’s contract is that, if it truly is for five years as reported, it puts his average annual salary at $10 million. That’s impressive, and it’s starting to look like a real major-league player’s salary, despite the fact that Epstein has pitched a total of zero innings this year for the Cubs.

While it’s a major-league salary, immediately after it was announced, it was pointed out that it’s the sort of salary that J.A. Happ, Derek Holland, and Edwin Encarnacion made this past year. Is Theo Epstein worth the same as J.A. Happ?

There’s no really good way to value the contributions of a general manager or of a whole front office. The attempts have been made aren’t entirely satisfactory, mostly because much of the information that we would need to make a reasonable estimate is hidden from view. Plus, I’d argue that a lot of what gets talked about when we talk about “what a general manager does” is really only a part of what a general manager actually does.

The Epstein-Happ comparison is also scuttled by a translation issue. We’re used to thinking of players in terms of on-field value measured in Wins Above Replacement Player during a single year. With front offices, we swing and miss on all three of those. (That’s a strikeout.) With players, we at least get to see what they do on the field and we get a good record of it. With general managers, most of what they do is not a matter of public record. Not only that, but the “replacement level” baseline for players is based on actual observations of what bench/fringe players do in actual games.

We get to see what the utility infielder can do against major-league pitching. We’ll never get to see what the 31st-best GM would have done with the Cubs (or anyone else) this year. And while it’s easy to segment a player’s contributions into individual seasons, it’s harder to do that with a GM. For example, the Cubs picked Kris Bryant in the 2012 draft. This year, he might just win the NL MVP. If we’re going to give the Cubs' front office credit for that, should it be in the 2012 box or the 2016 box?

Still, we do have one easy metric of comparison: dollars. If a win is worth $7-8 million on the free agent market, then we are suggesting that Epstein is worth a little more than one win per year. If we could find a way to translate his work into single-season wins, we can see whether he is a better or worse deal than J.A. Happ. We are not going to walk out of this article with a solid answer on the matter or a ranked list. But I think that even trying to outline the contours of the case is instructive. Let’s see if we can at least make the case that a GM could be worth an eight-figure salary.

Warning! Gory Mathematical Details Ahead!

First things first. We’re going to be using the terms “general manager” and “front office” interchangeably here. The GM is simply an avatar for the cadre of people who work for the team. Maybe at some point, we can try to isolate the contributions of the player development department vs. the analytics department, but not today.

If we’re going to do this in terms of Wins Above Replacement, let’s start, like WAR does, with the core assumption that a player’s (or a GM’s) value can be best defined by what would have happened if he had simply disappeared. In WAR, we assume that the team would replace the player with a bench/AAA/waiver claim. For our GMWAR, we assume that if Theo and friends decided to simply become cabbage farmers rather than run the Cubs, the Cubs would hire someone else to take over. Because GM vacancies come up not infrequently (the Twins just filled theirs), we know who the “replacement-level GMs” out there are. It’s always the same group of people. Most of them are assistant GMs, and they would likely bring along a group of other lieutenants who have previous front office experience.

We are not talking about “callers to sports talk radio” many of whom, despite their braggadocious attitude about their own knowledge and fantasy baseball prowess on The Sports Show with Larry the Big Dog, would have no idea where to even start when they got into the big chair. We’re talking about people who actually know how a front office works because they’ve worked in one before.

We can also dispense with the argument that there is an information gap. Yes, we in the public have an information gap, but if the Cubs were to actually hire the 31st-best GM in the game, that person would have full access to all the appropriate information needed to do the job, including medical histories, scouting reports, and an advanced look at those between-inning Jumbotron things where they ask players what their favorite flavor of soup is.

The nut of value that we want to crack is the question “what did Theo and Friends do better than a replacement-level GM would have done?” It’s hard because only Theo and Friends deal the Cubs specific circumstances. We don’t even know what Theo and Friends would have done if they had been helming a lower-budget team and we don’t even get to see what one of the other 30 actual GMs would have done with the Cubs' circumstances. When we are evaluating players, the end-goals that we care about are comparatively simple. You either made an out or you didn’t. You either hit a home run or you didn’t.

Even if we had perfect insider information, how do you value a specific decision made within a very specific set of circumstances that are fluid and dynamic? Do I congratulate you for that move that turned out to be pretty good or re-write an alternate history and excoriate you for the fact that the same move that you made ended up preventing you from making a move that would have been even better? Do I ding you for the incredibly stupid free agenct signing that you were about to make, but were saved from making because someone else was even dumber (and richer) than you and added the eighth guaranteed year?

We are going to be hamstrung by a severe lack of information. But let’s give it a shot.


Let’s start with free agents.

Usually, this is the first place that fans go when they think about “what a general manager does.” The standard way that we measure free agent signings is to divide salary by WAR, trot out the ever-popular “7 or 8 million per win” canard, and call it a day. Actually, we know there is actually only a moderate amount of repeatable skill in being able to get “good value” on free agent signings, so even if we see a GM who’s gotten a better rate than $7 million per win on his free agent buys over the past few years, it’s not necessarily a guarantee to continue.

Free agents are also a little more complex than that. A GM needs to identify which free agents will make this specific team better. If his team is in dire need of outfield help, a guy like Jason Heyward makes sense. If he already has three decent guys to plug out there, sure, Heyward might be an upgrade, but not as much. Our GM also needs to bid enough that he will be the highest bidder, but not so much as to over-spend the market (offering $100 million when the second highest bidder offered only $50 million) or to over-spend what the player is actually going to produce (which he has to be psychic about, because contracts are given before the merchandise is actually delivered). If you want to get really deep into it, we also have to adjust for what ownership is willing to spend and where a team is on the win curve.

And sometimes it’s about more than just efficiency. A team can run a very efficient payroll (measured in dollars-per-WAR) by spending nearly nothing and having a bunch of replacement-level players on their roster. And they will go 50-112 for their efforts. And no, we’re not going to come up with a calculus equation for this. We want to look at this from the point of view of replacement level. What does the Cubs’ front office do that a replacement-level front office would not have?

Well, in some sense, I think free agency is the wrong place to look. By the time a player gets to free agency, he is about as well-known a quantity as one can get. By definition, we’ve got at least a six-year track record of his performance in MLB and he’s in his late 20’s at the youngest. There’s “a book” on him, and we know his injury history and temperament in the clubhouse. I think that it’s instructive that there are several projection systems in the public space, they largely agree with each other, and that when the general public makes predictions about what a player will eventually get in free agency, they do a remarkably good job. If the general public can do that, it’s not likely that a front office is going to differ grossly from that.

Yes, teams are going to have information that the public doesn’t have that might move the needle in some cases, but remember that our replacement front office is going to have access to that information too. What’s going to distinguish between a good front office and a replacement-level one is the ability to make very fine distinctions within that information ahead of other front offices. The difference is going to be in those little edges.

Let’s consider the catcher-framing craze of the past half-decade. The idea of catcher framing as an important skill is not new, but the technology to really put a number to it and to determine which catchers were good at it didn’t really come about until the advent of Pitch F/X. Teams were much quicker on the uptake than the public gives them credit for, but some were ahead of the curve. Still, we’ve now reached a point where everyone knows about the catcher-framing effect and those skills are now baked into the price of a player. What began as a great coup of knowledge is now simply part of the wallpaper.

For a front office to succeed, they not only need insights like that ahead of the competition, they need a steady stream of them because the ones that they do find will eventually be copied. In baseball, it’s hard to hide an innovation and frankly, there are 29 other teams chasing the same thing you are. A front office also needs a plan to actually act on that insight. Understanding catcher framing is pointless if you’re not willing to go out and get Miguel Montero in a trade.

Now with that said, we’ve seen that there are plenty of little edges to be gained in baseball. Teams who had figured out the catcher framing thing before the others signed or traded for good framers who were considered offensive dead weight. Even in the case of trading for a backup catcher, that can still add a win or so from a roster spot that generally produces nothing. And they got extra value out of it.

I’ve argued in the past for the idea of “emergent WAR,” which is the idea of teams deriving value, not from individual contributions, but from the ways in which the parts of the roster can interact with one another. For example, putting a groundball-heavy pitcher in front of a good infield defense can be worth several hits turned into outs, and turning a hit into an out is worth roughly three-quarters of a run each time you do it. A good GM could have the wherewithal to execute a plan to gather those players together. These specific ideas (catcher framing, putting a groundball staff in front of a great infield) may not be earth shattering any more, but there are more of those out there, and they can easily be worth the amount of value it would take to justify a salary like $10 million per year, even if a GM only has one or two of them in his pocket.


Because we only want to give Theo and Friends credit for the things that were politely handed to them, let’s consider what we should be wary of giving him credit for. One of the things that Theo will be lionized for will be the drafting of Kris Bryant. And hey, having Kris Bryant in a Cubs uniform has been a boon for the team. It just so happened that in 2013, when Bryant was available, the Cubs had the second pick in the draft. It’s not a given that another GM would have taken Bryant (the closest thing to a “public” draft board that we have, Baseball America, had Jonathan Gray ranked as their top prospect and Bryant at no. 3). So, perhaps Theo and Friends get some credit for selecting Bryant when others would have gone with Gray.

But then again, for all we know, the Cubs had Mark Appel at the top of their board and were saved from drafting him by the Astros taking him with the first pick. It’s reasonable to think that some other “replacement-level” GMs (and indeed, some real GMs) might have taken Gray had they been in charge of the Cubs. So, in that case, we could (if we could know these things) give Hoyer some partial amount of credit for “knowing” that Bryant was going to be the better player when a replacement GM would not have “known” that. Or should we?

The evidence shows that, much like getting free agent signings right, GMs (the real ones) don’t have such a great record of getting it right when it comes to draft picks. Once you get past the first round, they are essentially guessing. Now, it’s entirely possible that Theo and Friends have figured out how to tell the good ones apart from the bad ones, or at least to guess correctly more often than they would by simple chance, even if they don’t get it right every time. But we can see that the difference between a season from a solid first rounder like Gray and a bonanza like Bryant can be a big deal, and it can last. Again, even if we only ascribe some small amount of the credit to the GM, there are six cost-controlled years for a player to produce value in and that edge could be worth a lot.

Yeah … But $10 Million?

If there’s a message here, it’s that the advantages that might separate a good GM from a replacement-level GM aren’t at the level of “he signed Dexter Fowler, so he gets credit for Fowler’s four-win season.” That level of information is fairly saturated across the industry. It’s about understanding the nuances a little better than everyone else and having a plan that allows it all to work. But if that’s the case, why is Theo walking away with a $10 million annual paycheck? What about everyone else in the Cubs' front office? (Then again, we have no idea what any of them make.)

It’s not likely that Epstein is the one pouring over the data searching for that little tiny edge or out randomly checking on a high schooler in North Dakota that someone flagged as a possible 30th-round pick. I think that the answer is in the enormity of the enterprise. Let me frame the issue thusly. I am someone who weekly writes articles in which I test some crazy theory about baseball, a lot of which don’t work out. That’s just the nature of research. You try a lot of things, and most of them don’t work and the ones that do work have probably already been discovered in other front offices. Scouts watch a lot of bad baseball in the hope of finding one guy who might be worth a high draft pick or a signing bonus that’s worth mentioning.

But if I had an interesting insight … then what? Let’s say I discover that pitchers from Texas with seven letters in their first and last names are good (hello, Clayton Kershaw!) and no one seems to know yet that it is the combination of the seven letters plus whatever they do to kids in Texas that makes them amazing. I would pass that along up the ladder, because … well, I can’t really do anything with that. I’m going to be too busy specializing in what I know how to do best (or, more to the point, the only thing I really know how to do). I don’t have the skill set to conduct negotiations or to also incorporate scouting information on these players (though, I’ve heard from several scouts that Kershaw is very good). I don’t have the over-arching plan to put all of this information together so that my insight can be acted on. If I did, I’d be … the GM.

The GM has to be the organizing emergent principle behind a front office. The much-discussed “rapidly expanding front office” is actually a nod to the fact that no one person can do everything that needs doing. That’s not a slam on them, that’s just the way things are. There has to be one person who can aggregate all of that knowledge and those little advantages and turn it into a coherent actionable whole. Someone who can understand why the statistical thing is important and why the scouting thing is needed and why we’re even talking about the player development thing. Without that, a front office is just a bunch of people making noise.

With that emergent organizing principle in place, a baseball team is a living, breathing organism that is more than the sum of its parts. That’s where the GM makes his money. A baseball team is such a complex organism and the skill set it takes to run one well, while not impossible to find, isn’t easy either. There are probably plenty of people who can approximate it, but very few who can do it exceptionally well, and the spaces in between really could be worth the kinds of millions that executives are starting to be paid.

Sure, the 31st-best general manager in baseball knows a lot more about baseball than everyone who is currently reading this article. But there’s a lot of room to grow. We see that in the 30 GMs who are currently employed. And yeah, it’s not unreasonable to think that the best GM in baseball is worth a win or two more than that 31st guy.

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 am not so sure about the best GM sitting on the sidelines being worse than the 30th best GM with a team. At that point, you are having faith in the owner to pick the best GM. Were the Twins really run by one of the best 30?

Just like GMs make unfortunate acquisitions—Hello, Ubaldo Jimenez and Yovani Gallardo—owners do also.

At some point on the curve you hit a point where it is a tomato-tomahto difference. Pitchers are more or less equivalent, but one gets more ground balls, and the other more strikeouts. One player hits for more average, but the other has better defense. I wonder at what point the tradeoff starts to happen in GMs if you break down their disparate skills.
When reading about the idea of "replacement level" or "31st best" in GMs, why did Dave Stewart come to mind?
I have always thought that differences in the front office were an exploitable market inefficiency in baseball. While much of what Mr. Carleton said about free agents is true, a single decision like choosing JA Happ over Yovanni Gallardo... well, how much was that worth? If I were an owner, I would find the best GM I could an offer him $10m+ rather than dropping similar coin on a questionable free agent 4th OF.

Also, if I owned a team, I would try to buy the Cardinals' entire player development system. I mean, how much surplus value to those guys churn out every year?