Baseball is the most historical of American games, the one featuring constants in its basic numbers. It's always been three strikes per out, three outs per inning, and nine innings per game, all of it playing out in diamonds defined by familiar 60'-6" and 90' distances. Experts have been breaking down the statistics produced within that framework for at least 150 years, with every generation adding new insights along the way. There shouldn't be all that much room for new numerical analysis.
There is, though.
Nearly three decades after Bill James revolutionized baseball with systematic, evidence-based approach embodied in ‘sabermetrics', the game still has plenty of room for innovative new views. Bill Felber's ‘The Book on the Book' is proof.
Felber, the editor-in-chief of the Manhattan (Kansas) Mercury, starts off ‘The Book' with fresh statistical perspectives for on-field performances, then moves on to present new data on the managerial decisions and executive strategies that also shape the on-field action. The project's a treasure trove on competitive advantages, one that's equally insightful in examining hitter's situational hitting and managers' play-calling as it is in illuminating budget decisions and fan economics. All throughout its pages, baseball's conventional wisdom (‘the book') receives a set of thorough, convincing alternatives.
Bill Felber has managed to produce some very new analyses for a very old ball game, and he recently discussed them:
What did baseball mean to you while you were growing up?
I was born and raised on the south side of Chicago, and baseball was a part of my earliest memories. We played every day in back yards and sandlots.
One thing that was unique about Chicago was that we could come home from school every day and see a game on TV. It would be the seventh inning, but it would be on, and could follow it and learn from it. So, I was very lucky in that baseball was something I played and studied a bit from a young age.
When did you first get interested in writing a book about the game?
I was an associate editor with The National Pastime [periodical] in the mid-1980's - 1984, 1985, 1986 - but I was primarily focused on my newspaper work well into the '90's. I just started kicking around some ideas, more and more, so in 1998, I approached John Thorn. John got behind my notions and, eventually, my drafts, until [the book] started to take on a life of its own.
The notions centered on what you call 'The Book' on baseball play. Can you explain what that is?
It's 'conventional wisdom', in a phrase, when it comes to putting together lineups, pitching strategies, and on-field tactics like hit-and-run.
Let's face it - there's a lot of conventional wisdom in the game, and in two different areas. First and foremost, you have the sort of old-time, inside baseball folks who've grown accustomed to doing things in certain, safe patterns. Second, there are a lot of guys in SABR [the Society of American Baseball Research] and [research] groups like that. I love them, but they can have their own echo chambers.
The book was based on a desire to get away from all that, in a detached fashion, and find out what really works. It was an effort to put these old ideas up against the real-world data and draw some conclusions. By the late 1990's, computers had made it possible to gather a lot more baseball numbers, relatively affordably and efficiently. One of the challenges, as it played out over time, was to constantly keep up with new research and data that kept popping up.
Basically, I wanted a new look at baseball, one that wasn't in the form of a talk-show rant or a scholarly treatise. Something that was fun but serious, too.
One of the first things that leapt off the page was your rough treatment a couple of very respectable modern statistics, in Park Effects and Win Shares. Can you talk about your thinking in that regard?
In the book, I tried to make the simple point that park effects numbers are, essentially, tools for fantasy baseball. They're relative numbers and they can change significantly due to outside factors. They aren't predictive at all.
I regard Bill James to be a friend but, you're right, I don't consider his Win Shares concept to be a universal, unified measure of ball player performance. From reading the book, you know why - there's an absolute quality there, in the fact that every team is assumed to total up the exact same number of Win Shares, every time, every year. Bill's enough of a mathematician to recognize the role of random chance, but in the Win Shares concept, that's simply not allowed.
More than that, you've very skeptical about most closers' numbers.
Most fans look at most numbers, I believe, to make projections about what's likely to happen in the future, but I guess I have a lot of skepticism as far as that goes. Most often, you don't have sufficient data to be comfortable, short of a couple of hundred or so data samples per player.
So I'm very skeptical of closers, only because that few of them pitch sufficient innings to project what they're going to do. As a matter of fact, if you look at most closers' performance numbers, there are wide variations from year to year. In any given season, they may be way up or way down.
Think of it this way - you wouldn't make a judgment about a starter's season based on, say, his first 18 innings pitched. Just because [Phillies set-up man] Randy Wolf, for instance, performs in a certain way over fifteen appearances, why assume there's enough of a database for us to reliably predict how he's going to do on his sixteenth appearance?
What's happened, frankly, is that relief pitchers have signed up some very good agents who have done a very good job in convincing team executives and fans that their clients are worth a lot of money. After a stopper gets $5 million per year or whatever, then managers basically have no choice but to use him as a be-all and end-all.
What about someone like Mariano Rivera?
I have only one quarrel with Mariano Rivera - the Yankees should pitch him more often. Instead of pitching him 70 or so innings per year, they should take him out for 85 or so. He's solid in every respect and deserves his reputation
The problem is that every team thinks that they have a Mariano Rivera or an Eric Gagne. What they really have is a Jose Mesa, a guy who's up and down in one year to the next. Predictions on closers are, generally, very questionable, so I'd argue, spend the $5 million on a starter or a position player instead. Throw a live young arm out there as a closer and spend the money on someone with solid numbers. Meanwhile, by all means, find durable workhorses among starters and do your best to maximize their effective innings pitched.
With that kind of skepticism of closers, you turn around and apply numbers to managers, and rating them in their tactical decisions on bunting and steals and such. What was your thinking in that?
Well, you can gather some data on managers. (chuckles) I don't know why someone hasn't sooner, but then again, maybe I do - gathering up all the relevant information was a tale that could fill another book. (chuckles)
I do try to preface those tactical results by saying that on-field game calling probably isn't the most important part of a manager's job. Most of the job is in clubhouse leadership, in dealing with different personalities, making sure they devote themselves to team goals and stay motivated. I don't deny any of that, but, unfortunately, that doesn't lend itself to any kind of an objective analysis, any more than player 'hustle' or 'grit' or something can be measured.
What I did was look at the manager's averages, in usage and in success. From there make some conclusions as to how often they tried to steal, for example, and if they fared relative to their competition. You might not have fully reliable data, because of relatively small samples, but it's a start.
I've never seen anyone try to rate General Managers in any systematic way, either.
I don't think anyone's done that, and it's strange. Even those who don't necessarily agree that pennants are won with off-season moves can agree that the front office is incredibly important to a baseball team. That's fairly incontrovertible, as far as I can tell. Yet, those who take stabs at assessing General Manager performance rely on vague info and platitudes. If you look, historically, a lot of people didn't even know who the de facto GM of a club might be - for many years, the job of putting together a roster was divided among the owners and various executives and the manager.
At any rate, my GM ratings came about because I got kind of tired of hearing that 'so-and-so General Manager is a genius' but, then again, 'so-and-so is a genius', too, and so is another guy. It can't all be true - they can't all be the greatest general manager in the world, right? In Pete Palmer's formulas, I found a pretty good basis for some solid conclusions and, after that, it was just a question of setting some parameters and running the data.
Trying to look at the data in an objective way, I came across surprises. Terry Ryan [of the Twins], for instance, came out very strong in his short-term rating. His team didn't make the playoffs in 2005, but he signed Johan Santana, who deserved a Cy Young Award, so the team was significantly better off than it otherwise would have been. Kenny Williams, on the other hand, had a very average number for the year, though he'd laid a foundation for the World Series championship in previous years.
To me, your approach was great because it put some concrete foundation into the contention that 'the Yankees front office is dumb, but they have money' or 'the A's are brilliant, but they can't compete financially'.
Sure. I wrote that 'squandering money is the easy part', and that might be true for some big market teams, more than the media, much less the team executives, will admit.
A favorite profile is Walt Jocketty. You look at the data on what he's done over the years, and it's pretty easy to see why St. Louis is a perennial contender. Between Jocketty and [manager Tony] LaRussa, you've got a Cardinals team that really knows how to acquire talent and use it.
Before I picked up 'The Book on the Book', I thought that most franchises were almost identical in allocating their budgets.
Can you explain your approach to another new concept, in team building analysis?
The thought was that teams do have identical roster slots, but they can have wildly different priorities in the way they can allocate their money between stars and reserve players, between their pitching staff and position players.
The origins for that were, really, in my budget decisions here as an editor with the [Manhattan] Mercury. I know that the salaries I pay have to fit into the paper's overall budget, and they have to be in line with what I see as the most important needs for the business. It's the same with baseball - different GM's will make different calls about what's important, and allocate their budgets accordingly.
Here, again, you have patterns. John Schuerholz consistently puts a good portion of his budget into pitching, and it's had its payoff. Others have tried to put disproportionate dollars into one or two superstars, and that almost never works out.
That brings up another interesting question. Is Alex Rodriguez overpaid?
(chuckles). That depends on what you mean by 'overpaid'!
As you know, in 'The Book on the Book', I tried to do something new in systematically apply performance to salary data and then, based on league averages, try to judge which teams were getting the most production out of their contract dollars.
So, is A-Rod overpaid? Definitely, as far as on-field production. As great as he's been over the last few years, his mega-contract was so big that he's been overpaid by about $10 million per year, something like that.
Here, again, as an analyst, I have to acknowledge that the stats don't tell me everything. Maybe A-Rod's being paid because he puts extra fans in the seats, or because he's been marketable for the Rangers and Yankees franchises, or he helps establish new media contracts, and so on. OK, I don't know what that sort of thing is worth, exactly, but it's there. You can't assume that it's all about winning baseball games, period. In the real world, there are different factors that come into the equation.
One of the interesting things in the course of the book is trying to establish why ball clubs do really strange things. For instance, I mentioned Terry Ryan - his first few years as GM were terrible. Based on the team's final standings, he should have been fired. The fact that he wasn't fired, and actually came back with strong performances, indicates to me that a small market team knew that they had a smart guy who could build for the long term.
It's much the same thing with the performance/salary breakdowns. One thing I've tried to convey is that losing teams sometimes offer nutty contracts because they're looking beyond the numbers, to that off-field stuff. For example, Detroit's contract to Travis Fryman in the mid-'90's. Some might say the Tigers offered him a big contract simply because they didn't know what they were doing. It is possible, however, that they wanted to make a splash and look like they were doing something rather than conceding that they were completely lost.
The way you describe it, plenty of Major Leaguers are overpaid for their performance, but then you point out how many are dramatically underpaid, too.
I firmly believe that almost no one in the Majors matches production and salary. Free agents get too much money on the market, but young stars don't have that kind of leverage before their seventh year in the League, so they have to settle for hundreds of thousands or a few million. It's a lot of money by any reckoning, but far below their market value as free agents.
Are you optimistic about baseball's competitive balance?
It's difficult to draw definitive conclusions at the moment. If you look at the data, there was a relatively strong correlation between payroll and standings in the late ‘90's - something like 70%, 80%. The richest teams almost always finished ahead, but then things changed since 2000 or so, to the point where the correlation can dip to a relatively modest level, 30% or so. A lot more small payroll teams were winning.
What happened? Maybe expansion had something had something to do with it. Maybe general managers became more savvy. Perhaps the salary structure isn't as skewed as we've been led to believe by the media - after all, only the Yankees and Red Sox dominate their division, the AL East, by spending a lot more than everyone else. It could be any number of things.
It does strike me that there has been a cyclical element over the years, where money has been more important at certain times and less important in others. As for baseball's future, I honestly don't know. There are a lot of possible factors at work.
Have you been happy with your work's reception among critics and readers?
I'd love to have the whole world shouting about it, I guess, but I feel very good about the reaction out there. I'm certainly looking forward to my next book, about the 1897 season, when the bad-guy Baltimore Orioles and the good-guy Boston Braves battled through one of the great pennant races in baseball history.
The greatest thing is, I've learned a lot about the game in the process of writing the book. For instance, I used to think that it was important to exercise patience, patience, patience at the plate. The cliché is, take the first strike, but I wanted to do a breakdown by pitch count. After studying the data, I learned that, on 0-0 counts, hitters put up something like a .340 batting average with .520 slugging. Now, I'm much more inclined to follow the Richie Hebner rule - take a whack at the first nice pitch you find.
After writing 'The Book on the Book', are you more of a baseball fan, less of a fan, or about the same?
I may be a bit too analytical now, trying too hard to figure it out instead of sitting back and enjoying it (chuckles). But, no doubt, I'm even more of a fan.
The complete Table of Contents for the ‘Baseball Men' interview series can be found here.