On the eve of tonight's all-star game, the Science Channel aired an excellent documentary, "Baseball's Secret Formula." What's the formula? It's the integration of the ideas of Bill James and sabermetrics into managerial and player-personal decisions. Who's practicing the formula? According to the documentary, the Oakland A's of Moneyball fame and the Boston Red Sox, a team that not only employs James on its payroll, but that has moved several players--impatient hackers Nomar Garciaparra and Shea Hillenbrand (both currently hitting above .300), to name two--who don't fit James' philosophy.

What's this all about? Think American League baseball, or, at least a variant of the stereotype of American League baseball. James's philosophy posits that preserving outs, rather than advancing the runner, matters most. So steals, sacrifice bunts, and intentional walks are no-nos. The occasional cry, "a walk is as good as a hit" is taken perhaps too literally among the Jamesites, who discard batting average for on-base percentage. The RBI, one of the big three in batting statistics, gets downgraded because where one hits in the batting order, rather than how one hits, plays a major role in the accumulation of them (but isn't where you hit in the order determined by how you hit?).

Perhaps the most interesting admission of the program was James's concluding statement that statistics can only tell us so much. There are too many situational variables to judge a player solely by the back of his baseball card--no matter how many of the nerdish baseball fanatics featured in the documentary wish it were so. How do statistics quantify Derek Jeter's clutch performances? Don't defensive statistics lie in placing Doug Mirabelli at the bottom of catchers when he actually does a masterful job catching knuckleballer Tim Wakefield? Where do statistics measure how the mere basepath presence of a Ricky Henderson, or a Lou Brock, or a Jose Reyes for that matter, sends pitchers into frenzies and puts defenses out of position?

In addition to forwarding the controversial ideas that teams would be better served using their closers in the seventh inning than in the ninth and that batting order really doesn't matter all that much, James offers in the Science Channel program his all-time all-star team based on his statistical analysis. They are:

C Yogi Berra
1B Lou Gehrig
2B Joe Morgan
SS Honus Wagner
3B Mike Schmidt
LF Ted Williams
CF Willie Mays
RF Babe Ruth
P Roger Clemens

No Ty Cobb? Barry Bonds? Hank Aaron? Johnny Bench? Rogers Hornsby? Even the most scientific, most well-thought-out statistics can't settle a baseball debate.

That team is hard to argue with. I can't say that I would take any of your suggestions over those he picked w/ the sole exception of Hornsby over Morgan. Aaron cannot beat out Ruth for the RF position; Mays beats Cobb in CF; Berra beats Bench and to throw Bonds name in here is a joke, despite the fact that he would beat Williams if we ignored reality.

I would think the easiest to argue w/ is Clemens as your sole pitcher. First of all, I would choose a Lefty if I could only choose one pitcher, and that would be Koufax (despite his short career . . . I can be swayed to go w/ Spahn). Even if I am picking a righty (and of course that side has the greatest pitchers of all time so maybe I should) then I would pick "The Big Train" Walter Johnson over Clemens. I think his crunching of stats must value the strikeout a bit too much for my taste.

As for the "screw advancing the runner" and "no such thing as a productive out" crowd, a.k.a., the American league, that is a degenerate form of baseball and is both frustrating to watch and the cause of a decline in character and moral fiber in those who do watch it.

The sac-bunt builds character, end of story.

No Nolan Ryan?

Roger Clemens key stats: 341 career wins (22 years through 2005); 3.12 career ERA; 7 ERA titles; 7 Cy Young Awards; 1 MVP. Nolan Ryan: 324 wins (27 years); 5714 strike outs (no 1 alltime); 2 ERA titles; 3.19 career ERA; 0 Cy Youngs; 7 no-hitters.

Nolan Ryan probably does not crack the top 10 pitchers of all time (although still a very good pitcher). His problem, was that often he was not the best pitcher on his own team (0 Cy Young's is a glaring number over a 27 year career).

How many gold gloves did Williams have? How many stolen bases? Ted Williams was a fantastic, top tier Hall of Famer. But Barry Bonds has it hand-down over him for the LF spot, current controversies notwithstanding.

That said, I think it is very hard to argue with Bill James's statistically chosen line-up. But I also agree wholeheartedly with Dan's thoughts about the "intangibles" and how they must be figured in (esp. by GMs), even if they can't be statistically measured.

Williams sat out 5 years in his prime serving his country in 2 wars. In searching out reasons for picking the best players I would factor that into statistics. He had little opportunity for golden gloves as they didn't exist until 1957. Batting average, slugging percentage, on base percentage, Bonds more than doubles Williams on strikouts. This all with him missing 5 of his best years. I'm going with Ted. And this doesn't even address Bonds rule-breaking steroid enhanced performance of the last 6 or 7 years.

Much of what you say is true, Webster. But even though it is a tragedy to baseball (but not the country) that Williams missed 5 prime years, the fact is those stats aren't there. Some stars had their careers prematurely ended via dramatic injury, but we can't project forward. Not quite apples to apples, I know, but we can only take our "factoring in" so far.

There may not have been a Gold Glove Award, but Williams was never considered better than an average outfielder with the glove.

Bond may have had the end of his career boosted by drugs (and also better understanding of nutrition, the value of exercise and weight-training, video, etc, etc, etc), but it is also increasingly evident that he did it against pitchers that were similarly artificially enhanced. And it is absolutely impossible to determine the effect one-way or the other. In the end, players must be evaluated against their own era, and Bonds is a colossus in this era, against many who we do know did in fact use steroids.

Anyway, I line-up Williams’s and Bonds's batting stats and all things considered there, I find it a virtual wash. But then you take Bonds stellar fielding for most of his career and his 400-plus stolen bases (he was a 400HR-400SB guy well before anyone accused him of chemical enhancement, after all - by far the only one in the history of baseball), I have to give Bonds the hefty edge of the all-time LF spot.

Come on now Eric,

Williams is attacked for his defense but that is really exaggerated . . . just as Bonds defense is exaggerated. Williams absolutely mastered playing the Green Monster, and had a lifetime fielding pct of .974 compared to Bonds's .982 (that diff is neither negligible nor cause for exaggeration). Bonds always had great range but a really lousy arm, which is why he was moved from center to left, and that shouldn't be forgotten.

You want to dismiss the fact that Bonds spent the last 8 years of his career juiced AND dismiss the fact that Williams sacrificed almost 5 full seasons, between the ages of 24-26 and 32-34, because supposedly some D-list pitchers were on roids as well? You simply can't draw conclusions about Bonds hitting stats since he turned 33 unless you devise some method of adjusting for the cheating. Look at Williams numbers since he turned 33 . . . and ask yourself if Bonds continues at the same level as Williams w/o cheating.

Speed is obviously Bonds's edge, and I agree that that should be more of a factor than Sabermetrics makes it out to be. But since this lineup is clearly built as an "all-star" type line-up rather than a actual working team you could put on the field speed just isn't that important. How would one put these guys in a batting order that would make Bonds speed noteworthy?

Throw in Joe Dimaggio, put Ruth at DH.

it is a tragedy to baseball (but not the country) that Williams missed 5 prime years, the fact is those stats aren't there.
To me it is a tragedy that Bonds's stats are there because they confuse his place in baseball. He was fast and a great fielder, but he is also a cheat. Can't use a cheat. My all-time team consists of those I want to bat, run, field, and pitch at their best for one game. Those players don't include Bonds. Pitchers, though, are much harder to pick. There are Gibson and Koufax to consider and a host of others.

Acknowledging their greatness, I think Sandy Koufax and Joe DiMaggio are two of the most overrated players among great players. I'm not questioning whether they belong in the Hall. I'm just questioning why their names are included when discussions move to the best of the best. DiMaggio's lifetime line: 361 HR, .325 BA, 1537 RBI, and 2214 H. Statistics do him an injustice as the war cut his career short. But there are at least six outfielders I'd put ahead of him, and even if I carried an extended roster that allowed for a third-string of outfielders I'm not convinced he would make my team. Koufax was a rather mediocre pitcher for most of his short career who had four great years to end his career. Let's get real: he won 165 total games. Just as we adjust Bonds' numbers for the hitters' era we live in, we should adjust Koufax's (and Gibson's for that matter) numbers for the pitchers' eras they played in. But even if you weren't to adjust Koufax's numbers, you're left with the realization: they're just not impressive. If Pedro were to retire today, he would have had a better career than Koufax. Ditto for Maddux, Clemens, and Johnson.

I threw Koufax out there as the Lefty I would choose though it is close w/ Spahn who had a greater career but wasn't as gifted and dominant at his best as Koufax was at his. I didn't explain it above but I was thinking that if you had to pick one pitcher then I might prefer to pick a great lefty over a righty for the simple fact of more batters being righty. I wouldn't claim that koufax was better than Clemens however.

Let's ignore Dan Flynn's anti-Bronx, anti-Brooklyn diatribe ... sounds pretty senseless to me.

We all know that Pedro isn't worthy to lace up Sandy's baseball shoes. It's a simply question of character.

And I say that as a Yankees fan with three generations (father, grandfather, great grandfather) coming out of Mickle Avenue in the Bronx (*not* Brooklyn).

Statistics like those Dan cited are scheisse.

I second Brian's suggestion, though, of Walter Johnson as pitcher ... righty or lefty. I would take him over Koufax any day.

I don't see how he could be beat out by anyone, let alone Roger Clemens.

DvT

In the entire Hall of Fame, there are but two starting pitchers with fewer lifetime wins than Sandy Koufax. There are actually several hundred pitchers in the history of MLB who have more wins than Koufax. Wins is a pretty significant category for starting pitchers. Longevity is also important. Sandy Koufax had four years as an elite pitcher. They were four incredible years, but four years is still four years. He was a mediocre pitcher for the better part of the rest of his short career.

Despite the fact that Pedro pitches in a hitters' era (and in his dominant years in a hitters' park, Fenway), and Koufax's best years were in a pitchers' era (and in his dominant years in a pitchers' park, Dodgers Stadium), Pedro boasts about 40 more wins, a slightly better lifetime ERA, and about 600 more strikeouts. Anyone who says that Martinez couldn't lace up Koufax's cleats knows very little about baseball.

Willie Mays said Koufax was the best he ever faced. No question.

It's hard to compare the pitchers today, who very rarely pitch complete games, with those who pitched many marathon games in the past. If your criterion is that he had a 15 excellent year career, you are with Dan. If your criterion is who had the best fastball or curve, or who was most feared by batters, then you are not with Dan.

I guess if the question goes something like, "What pitcher, on his best day, would you want taking the mound?," then I don't laugh at an answer of "Sandy Koufax." But if the question is, "Who is the best pitcher of all time?," the answer "Sandy Koufax" is a joke. The guy had four great seasons, but a forgettable career otherwise. Here's another way of looking at it: If you took out Sandy Koufax's best season, would he still make the Hall of Fame? Maybe, maybe not. If you took out the best season of Pedro, Clemens, Randy Johnson, or Greg Maddux, they would still not only make the Hall of Fame, but would do so on the first ballot.

Dan, I agree. Koufax's rep is based on two other things. 4 no-hitters and the sad early career end. Both draw attention. He retired when he was well established as baseball's most fearsome pitcher, and that sort of drama just never happened before. Even Jim Brown was nearing career end when he retired.

Agreed, and good point about Brown. As much as people like to trump up his early retirement, they forget that he played a game, and a position, where players have a short shelf life. Nine years is a healthy career for an RB.

And I think Williams and DiMaggio benefit from the "what might have been phenomena" too, because of war service. Perhaps as well they should. Injury, I think, is a different bag of marbles.

In Koufax's defense to his mediocre early years that I pointed out above, the Dodgers, by virtue of signing him for such a large amount of money, were apparently obligated to keep him on the major-league roster when he was young and should have been in the minors. I wonder to what extent prematurely being a major leaguer led to his later injuries, a la Todd Van Poppel and other rushed pitchers. My suspicion is not as much as the 4-man rotation and 25 complete-game seasons, rigors that, as someone above pointed out, are not endured by current pitchers (For instance, Koufax boast more CGs than Roger Clemens despite Clemens having double the career).