Username:	Password:

Remember me

Forgot your username/password?

Print Email

Home>Articles

Willfulness At Work

By Steven Goldleaf

August 28, 2015

<div> A little personal history before I present my critique of George Will’s lionizing of Tony La Russa.  As I’ve noted, this is an excerpt from a book I was writing in the early 1990s, which I called Myths of Baseball, the premise of which was that much baseball writing was making up legends, often self-serving ones, in violation of the researchable record. I wrote several hundred pages of this book, and was pitching it to various publishers, when I noticed that new, often deeply flawed, books about baseball were being published much faster than I could write my critiques of them, so eventually I gave up on adding to the book, and finally gave up the book when I realized that it was fated to be hopelessly out of date by the time it would come out.</div> I can tell you more about the various chapters of this project, but for now I’ll stick to this one chapter, of which I found a print-out in my office very recently, and in which I’ve cleaned up a few typos. Apparently (it’s amazing how much I’ve forgotten about his project, which consumed my life for a few years), the chapter below is a truncated version of the complete Chapter Two, the full version of which gives many more examples of Will’s self-serving partial truths about baseball –this truncated version is limited to his misguided praise of La Russa’s stated philosophy of managerial principles. I sent this chapter, along with a truncated version of Chapter One, which concerned the sportswriter Tom Boswell’s work, as sample chapters, along with an outline of each of the book’s eighteen chapters (one for each half-inning?), to potential publishers along with my lengthy query letter. This query letter, the outline, and the two sample chapters are all I have at hand of the entire book, though I do have some floppy disks from the early 1990s that probably contain more of the actual text, if I could find a way to access a floppy disk at this point.<o:p></o:p> Basically, it’s a book of what Bill calls "Tracers," and a book heavily influenced by Bill’s methods (minus Bill’s diligence for crunching the numbers). I haven’t checked any of the math below in this millennium, but I did check it pretty carefully at the time, and the flaws in Will’s reasoning that I point out still seem pretty evident to me, even the political gibes at the beginning of the chapter. (This was composed only a few years, I’ll remind you, after Ronald Reagan left office in 1989, as ancient as this history may seem now.)  The other thing I remember about this chapter is that I was planning to deliver it as a talk before some quasi-academic panel that met in Cooperstown in the summer of 1992—oddly, I don’t remember giving this talk at all, other than vague memories of wearing a sports jacket to the Otesaga Hotel, but I seem to have done so. My c.v. lists "Cooperstown Symposium on Baseball and America Culture, June 16, 1992 ‘Willfulness and Political Argument in George F. Will's Sportwriting" and I trust my c.v. more than I do my memory at this point.  (For a variety of personal reasons, the summer of 1992 was one of the most eventful of my life, a downright crazy time, and if I gave this talk, it was probably not among the 20 most memorable events of that summer to me.)  In any event, here it is, while it’s still longer than this prefatory note: <o:p></o:p>   From Chapter Two, "Willfulness At Work"<o:p></o:p>   Through George F. Will's career as a political commentator, he has padded Ronald Reagan's accomplishments, just as Reagan's tailors pad the shoulders of his suits, making him seem a far more gigantic figure than he actually is. Although Will's praise of Reagan may strain his audience's belief, most of Will's listeners lack the data to rebut his detailed arguments. Will carefully, even pedantically, criticizes liberals, while ignoring Reagan's notorious manglings of facts. Will props Reagan up as an authoritative ideologue, and then uses him as a false authority for many of his own conservative positions. Happily, when Will sets up Tony La Russa in Men At Work as his ideological authority on a subject I do know something about, baseball, Will's over-reliance on a false authority is much easier to spot.<o:p></o:p> I'm not going to draw explicitly the extended analogy between Reagan and La Russa that I might, but Will's "treatment of Tony La Russa underlines a confidence in control and benign leadership" (Strauber and Antczak 4), a description that also applies to his treatment of Reagan. According to Will, both Reagan and La Russa are dynamic, imposing and articulate speakers. But, as I see it, the stats that Will calls baseball's "remorselessly objective measurement" demonstrate that La Russa's inspiring principles, to say nothing of Reagan's, crumble under the provable facts.<o:p></o:p> La Russa's leadership is primarily inspirational. Will notices this quality when he observes that La Russa "constantly refers to one intangible: intensity." Since that ability to inspire may be a manager's most important quality, the inconsistencies I'm about to note don't address the question of La Russa's ability to manage a ball team. His job, after all, isn't to analyze; it's to plot strategies and to keep his players focused on winning. He's obviously gifted at his chosen metier: he just isn't able to apply his own stated theories, nor is Will capable of noticing how wildly inconsistent La Russa is.<o:p></o:p> Will quotes La Russa asserting confidently that<o:p></o:p> there is . . . a psychological advantage in getting the lead and then increasing it. That builds an expectation of defeat in the other team, especially if you have a bull pen that can hold leads. So La Russa thinks scoring runs one at a time is important because scoring frequently is important.<o:p></o:p> Will approves of La Russa's saber-rattling position that threatening to score often is better than threatening to score a lot of runs at once, even if the same number of runs results from each strategy.<o:p></o:p> Although La Russa lectures Will on the crucial importance of scoring the first run by whatever means necessary, actually, on the playing field, contrary to Will, La Russa tries less than your average manager to score a single run early in the game: in 1989, the first season statisticians John Dewan and Don Zminda ran such stats (and only one year after La Russa told Will his philosophy of scoring early runs), his A's sacrifice-bunted fewer times than any other American League team but one. Those 1989 stats also contradict La Russa’s supposed practice of bunting early in the game: his players bunted only 3 times in the first inning. Eight other American League teams bunted more often in the first inning, only four bunted fewer. So La Russa avoids practicing his own preaching on the early sac-bunt.<o:p></o:p> Every year, La Russa claims in Men At Work [44], "the same statistic comes out showing that clubs that score first [in a given game have a] . . . winning percentage in those games [of] anywhere from .550 to .650." But scoring first means much less than La Russa thinks: to begin with, all teams, of course, have a collective .500 record. That's a pretty fair base to begin with. And an average team, winning 81 games and losing 81, can expect to get shut out 10 times. That means in games where they score even one run, the average team's record is 81-71, or .533. A goodly chunk of the advantage in scoring first, then, is simply in scoring at all. To advise major league ballplayers not to get shut out doesn't exactly tell them something they don't know. <o:p></o:p> La Russa's stat about scoring first is even more scewed than I have suggested so far. When he says "clubs that score first" he's including clubs scoring more than one run first. If the A's get a three-run homer in the top of the first inning, those three runs count as scoring first. That's a big-inning strategy, of course, but La Russa counts it exactly the same as if they'd eked out a single run. If they follow that three-run homer with 8 more runs in the first inning, all eleven runs will go down as the first runs scored in the game. It's pretty easy to win once you've gotten off to an eleven-nothing lead.<o:p></o:p> To validate his stat, La Russa needs to discuss only those games in which the game's first run is the only run scored in that half-inning. To eliminate the problem caused by games won on shutouts, La Russa would need to further limit his data to games in which the opposing side scored at least one run. If scoring the first run of a game is a valid top priority, then games in which you score a single run first and your opposition at some point eventually scores a run themselves will still be games that you win far more often than not.<o:p></o:p> I haven't done such a study myself, but it should be obvious that scoring a single run determines the game's outcome much less often than scoring the first five runs does. And, since the advantage in scoring first in all games is less impressive than it seems at first blush, the advantage in squeezing (or stealing) only that first run across is at best dubious. What I have done is a quick study, using post-season linescores for 1988, 1989, and 1990, the three years La Russa's A's played in the post-season. Of the 22 post-season games played by both leagues in those years where one team scored a single run first but did not shut out its opponent, that team won exactly 11 times, or exactly .500. The edge may be a little higher in a whole season of play, but I doubt that it's a big edge. Assuming it exists, it could be as low as .510, or one victory in a hundred, hardly the stuff pennants are made of. <o:p></o:p> La Russa gives "another fact Jim LeFebre pulled out. He took last year's club [the 1987 Athletics], a pretty good offensive club, and looked for the stat of how many innings were scoring innings in a game. It didn't matter how many runs we scored. When we scored at least three different times in a game, our winning percentage was at least. 600, even if the scoring was just a one and one and a one." The 1987 version of the A's went exactly 81-81, including games in which they were shut out. Since they obviously lost all the games they were shut out in, and lost almost all the games they scored only one or two runs in, isn't it colossally obvious that a .600 record when they scored three or more runs is the least you'd expect from a .500 team?<o:p></o:p> Three or more runs is an understatement, too. Because a team scoring "at least three different times" will usually score more than just three runs. At least one of those innings is likely to be a multiple-run inning, so La Russa's talking about games in which his team scores over 4 runs. Actually, the real figure is a whopping 5.4 runs (the number of runs in an average scoring inning is around 1.8, and we're multiplying that by 3 scoring innings).<o:p></o:p> How surprising is it that when the average team scores more runs than usual their winning percentage jumps from .500 to .600? Not very. It's surprising they don't do much better, when you stop to think about it.<o:p></o:p> If a team scored 5.4 runs a game, they would score 875 runs every 162 games, a figure that in almost any year leads any league by a wide margin. (In the 1980s, for example, the average league leading figure was nearly a hundred runs lower.) It might not seem like much, but scoring in at least 3 separate innings per game is fantastic hitting. Once you recognize that fact, La Russa's stat translates to: if I score far more runs than any team in the league (playing in the Oakland Coliseum, a park that depresses scoring), I'll play .600 ball. Big deal.<o:p></o:p> La Russa is arguing, in effect, scoring 5 runs in 5 different innings creates a better offense, somehow, than scoring 5 runs in only one or two innings. Bill James disagrees with this conclusion, asserting "it is safe to say that every sabermetrician who has studied the issue agrees with Earl Weaver." James, by the way, studied run distribution in 1982 and found the most nearly average manager as far as emphasis on big-inning and little-inning strategies was Tony La Russa. <o:p></o:p> James's analysis directly contradicts Will. Will blandly accepts La Russa's self-serving self-analysis. La Russa just doesn't know himself, and Will doesn't know any better than to take him at his word. La Russa talks a far riskier running game than he manages. It's fine that La Russa spouts this stuff. I just hope he doesn't believe it. Maybe he's spouting it so George Will will print it, and the other American League managers will be fooled into defending against La Russa's non-existent go-go attack. Will is impressed with La Russa’s facade: "La Russa talks the way he manages," Will mistakenly believes, "and the way he wants his team to play, controlled but intense." Deceived by the sizzle, Will eats the potato and says "Great steak!"<o:p></o:p> Some of Will's own figures in Men At Work are more misleading than false: Bob Gibson, he claims, "would have been [30-1 in 1968] if the Cardinals had scored even four runs in each of his starts" [105]. That word "even" implies that four runs per game is a pitifully low figure. But in fact it works out to 648 runs over a season. In 1968 only one National League team scored above 612 runs. The champion 1968 Cardinals, for example, only scored 583, or around 3 and 1/2 runs per game, a hair under the league average.<o:p></o:p> So, as with La Russa’s misleading figures, Will's figure, far from low, is significantly higher than average. A team scoring four runs a game in 1968 with a perfectly average pitching staff would have won over 93 games, so Will is clearly not describing an inadequate level of offense here. And that's only the beginning.<o:p></o:p> In positing what would have happened to Bob Gibson's record with this significant improvement in his run support, Will –like the liberal economists he often mocks-- assumes that Gibson's run support would be distributed in the most favorable (and least likely) way. Four runs each game-- no more, no fewer-- means that not once all year, in a league featuring Juan Marichal, Ferguson Jenkins, Don Drysdale, Tom Seaver, all at or near their peaks, would an opposing pitcher have shut out Gibson's team-- or even held them to 1 or 2 or even 3 runs. That's a wild hypothesis, and one at violent odds with probability. Whether relying on authorities or winging his numbers on his own authority, Will proves gullible or just unreliable, the very adjectives I'm tempted to apply to his political arguments.<o:p></o:p>

COMMENTS (11 Comments, most recent shown first)

OldBackstop
""And, since the advantage in scoring first in all games is less impressive than it seems at first blush, the advantage in squeezing (or stealing) only that first run across is at best dubious. What I have done is a quick study, using post-season linescores for 1988, 1989, and 1990, the three years La Russa's A's played in the post-season.""

In looking at the regular season numbers for those years, the As were above average in SBs, SF and SH every year, in fact, top three every year with the exception of one or two.

In the 1988 ALCS, the As swept the series, scoring first in each game.

In the 1988 WS, the As scored first once, and won the game. The Dodgers scored first the other four games, and won all four.

In the 1989 WS, the As scored first 4 times, and won all four times.

Anyway, I wonder if it is fair to judge a regular season strategy on a few dozen playoff games..as Billy Beane once pointed out, his regular season "shtt doesn't work in the playoffs..."

3:36 AM Sep 2nd

steve161
While we're doing the fact checking, the dinner was after a Cards@Mets game I attended in May of THIS year. The restaurant is about a 10-minute walk from Citi Field and was very good, but I don't remember what it was called. Ask Steven if you're hungry in Queens.

There are a few details in the article I would quibble with (the fact that La Russa eschews the early sacrifice doesn't necessarily mean that he doesn't value scoring first) but the main point is a lot more obvious to us now than it was 25 years ago, especially to dabblers like George Will: sometimes what players and managers actually do stands in direct contradiction to what they intend to do.

For the record, I liked Men at Work. It's not The Long Season, but it's a good read. I don't really think about his politics when I read his baseball writings, but I agree with raincheck that, set alongside the current state of the Republican Party, he looks like the voice of sweet reason.
2:35 PM Aug 31st

shthar
You'll never get rich writing about george will.
12:23 AM Aug 31st

shinsplint
It seems to me that better teams would tend to score first more often, and therefore they would win more often because they are better. And I would think this is true regardless of whether the team scoring first scored 1 run or 7. So it would take a winning percentage well above .500 to indicate that there is any strategic advantage in scoring the first run because with the better team scoring first there's probably an expected .520 (pulling that out of my a**) winning percentage to start with.
6:22 PM Aug 30th

flyingfish
Steven: I agree with this from your reply: "people make stuff up, but in baseball, a self-contained system with copious records, we can examine closely what people write, and show how we not only can disagree with their conclusions and opinions but show how the 'facts' they use to justify those opinions and conclusions were wrong." That's exactly what drove Bill James, if I understand him correctly, to start with his Baseball Abstracts and the rest of his baseball writing career. It's why I give him credit for changing the way people look at and think about baseball.

9:59 AM Aug 30th

Steven Goldleaf
tommyr--nothing wrong with lionizing players or managers, just don't do it by citing inaccurate figures or by faulty reasoning. If you've got a man-crush on Tony La Russa, just say so and I won't give you a hard time--Will has a way of reading from his thesaurus and entangling his reader with convoluted logic that takes a lot of effort to untangle, so most people give him a pass on his reasoning. I just tried here to show where his reasoning fell apart, which it did like a rack of well-done ribs.

And flyingfish, publishing this old piece isn't about Will's errors of 25 years ago as much as it was meant to display my thesis which is: people make stuff up, but in baseball, a self-contained system with copious records, we can examine closely what people write, and show how we not only can disagree with their conclusions and opinions but show how the "facts" they use to justify those opinions and conclusions were wrong.

I agree with your objection that this article is out of date--that's why I never published it in the first place. At best it would have gotten published in 1995 or -6, and would have been a review of a book published in 1990, while other, newer, possibly wronger books would have come out in those intervening years. But I was having dinner with steve161 after a Cardinals' game last year, and when he said something nice about La Russa's managing, I mentioned this chapter of this book. Steve161 asked to see it, so when I got this gig, he reminded me about my offer--I don't plan to publish further historical artifacts from my early career, and am eager to resume my "Thrown from a Moving Vehicle" series soon.

But I like my old thesis--so much of what we've all read about baseball is predicated on "facts" that never happened in the first place--personally, I think that when you argue a point, your evidence should be true rather than false, and we too often just accept someone's observations as accurate.

Oh, and mike--I appreciate your interest, but the events of 1992's summer were just personal--suffice it to say, though, that my divorce was probably the 3rd or the 4th most traumatic event of that summer, to give you an idea. Gives me chills even thinking about the stuff I was going through as I wrote this piece, and I'm glad to have moved on with my life.
5:18 AM Aug 30th

flyingfish
I'm not a huge fan of George "Why Use A Small Word When A Big Word Will Do" Will. I think he's probably become worse, not better, in the past 25 years, but I agree that taking him to task ONLY for something he wrote a long time ago isn't perhaps as interesting as it might be.
3:32 PM Aug 29th

raincheck
I also have been unimpressed with Will's work on baseball. He is a fan, a hobbyist, who is published on a subject he loves only because he had had success and a degree of fame writing in another arena. I would think that, like most thoughtful fans, Will has probably changed many of his beliefs in the face the onslaught of analytics since then. LaRussa as well. Critique of their 25 year old opinions seems a bit like an easy target, even if the critique is as old as the target.

On politics, on the other hand, he can often be quite thoughtful and is refreshing in that he is not knee-jerk partisan, often disagreeing with the predominant wave of his party on significant issues (he has long been substantially less apt to propose military force as a reasonable choice in international fairs, for example). On both side commenters tend to toe the party line, even when it changes rapidly, and Will, for better or worse, keeps his own counsel, which I respect.
10:40 AM Aug 29th

tommyr
I have no problem with some baseball writing lionizing players or managers. I am as big a fan of sabermetrics as anybody, and I count Bill James as a hero of mine. But this isn't chemistry or nuclear physics or biomedical engineering. It's baseball, and its enjoyment is not just served by sabermetric analysis and search for absolute for truth.

I read Men at Work and remember enjoying it. George Will is a political analyst, first and foremost. Other non baseball dedicated writers have written about the sport. On some level, that's a good thing, isn't it?
10:29 AM Aug 29th

mauimike
Ouch, sorry Professor. I'd like to know what happened in 1992.
1:26 AM Aug 29th

Riceman1974
Reaading this excerpt, it's not surprising it was never published. This is just awful analysis. I was far more interested in your mysterious exploits in the summer of 1992 than your tedious critique off Will, LaRussa and Reagan.
7:59 PM Aug 28th

Willfulness At Work

COMMENTS (11 Comments, most recent shown first)

Leave a comment

Report inappropriate comment


Type of Abuse:
Comments: