In a column posted a couple of days ago I tried to explain why I see the issue of signing Barry Bonds in 2008 very differently than do most of my friends and colleagues in the analytic community. I think we are making some progress in understanding one another here, so I thought I would pick up the cudgels and go at it again.
Rick Shaneyfelt’s comment that the Hall is a different discussion, better left for another time, is certainly correct.
Gary somebody (Gary, Indiana?) posed the question, “What are the criteria that make sense for a team to take on Bonds?” and suggested three criteria:
1) That it be an over-.500 team, within striking distance of the championship,
2) That they have a need at positions Bonds can play, and
3) That Bonds not be obstructing the development of a young player.
I think that’s right, generally; I think that you have to get past those three gates before you can reasonably discuss signing Barry Bonds, and I would point out that very few teams do. But I see it differently in this way: that this argument assumes that Bonds would help that team, given those three conditions. I’m far from convinced that this is true. After all, we have a historical example of a team that met all three of those criteria in spades—the 1935 Braves. They were in position to contend, they had a huge need for power at the corner outfield positions, and they had no young player to fill that role. They signed Ruth, at a moment when Ruth was exactly where Bonds is now except not under indictment, and it was a monumental disaster. Doesn’t that give you just a moment’s pause, before you assume that Bonds would help? If you’re looking at the possibility of a very marginal gain, against the risk of a monumental disaster, doesn’t that lock your knees just a little bit?
The core of the disagreement is the argument advanced by the tag team of Blackadder (which I think is a pen name for Kevin Towers) and Tangotiger (which, of course, is a pen name for Tiger Woods.) Let me take Blackadder’s comments, and work in my own responses as if this were a dialogue:
Blackadder: According to work by Tom Tango, a typical player ages in expectation .5 wins every year after his “peak” age.
James: OK, well. . name two other respects in which Barry Bonds is a typical player.
Blackadder: Maybe (you) think he should be aged by something like four wins, but I really just don’t see any evidence for such an extreme discount.
James: You would see it if you looked. Players who are similar to Bonds—and Bonds is the archetype, so this is more true of him than of the group—players who are similar to Bonds typically degenerate, having reached a point comparable to this, at an extraordinary rate of speed. You’re not seeing that for three reasons. One, you haven’t looked. You’re looking at generalities, rather than specific cases. Two, you’re confusing last year’s stats with this year’s skills. Three, you’re confusing OPS with ability.
OPS is a central skill, certainly, and, putting aside the quibbles with OPS as a measure of offense, let us assume that OPS measures offensive ability. There is still a variable relationship between OPS and overall ability. There are other things that are also important, namely defense, baserunning and the ability to play every day. Some players—Yadier Molina, Adam Everett and Jose Reyes—their OPS represents a fairly small percentage of their value. Other players—Bonds, Ted Williams in 1960—their OPS has a very high relationship to their overall value.
Old players whose OPS represents virtually all of their ability age, from that point, very, very rapidly. In most cases, their ability essentially disappears overnight.
Blackadder: The slots issue is a bit of red herring. Adam Dunn will face the exact same “slot” issue this offseason, except that he is a worse hitter and fielder than Bonds, and I guarantee that someone will sign him.
James: That’s simply not true, for this reason. Well. . .small point. Adam Dunn has driven in and scored 100+ runs in each of the last four years, except one year he had 99 and 92 or something like that. Barry Bonds has not driven in or scored 80 runs in any of the last three years.
Moving past that, the team that signs Adam Dunn this winter has a reasonable expectation that Dunn will continue to be the player that he is for another four or five seasons.
Perhaps the biggest mistake that you can possibly make, as a major league organization, is to fall into the habit of fixing problems in such a way that you have to fix them again the next year. A baseball team needs about 16 to 18 productive players to win—9 regulars, 5 starters, 3 relievers, more or less. Sixteen to eighteen “slots”.
Every winter, even if you have a very good team with a very good organization, you’re going to have to solve three or four problems in those 16 to 18 slots. This is difficult. If you fall into the trap of solving problems in such a way that they have to be solved again next year, you find yourself heading into the winter with 9 to 12 problems that you need to solve. This is impossible.
The essential problem of Barry Bonds is that he represents the ultimate short-term gamble—and short term gambles are for suckers. I’m not saying that nobody should sign him. I am saying that I would be very careful about it.
Blackadder: I think the last paragraph of the article is a little misleading.
James: What paragraph is that?
Blackadder: (reading) “The end came for Babe Ruth, and the end came for Joe DiMaggio, and the end came for Ted Williams, who had the highest OPS in all of baseball in his last year, and the end came for Stan Musial, who had fifth-highest OPS in baseball in his next-to-last year, and the end came for Mickey Mantle, who had a top-25 OPS in his last year, and the end came for Willie Mays, and the end came for George Brett, and the end came for Mark McGwire, who had the highest OPS in baseball one year before he quit, and the end has come for Barry Bonds. Give the man a round of applause, get off his ass, and let’s move on.”
James: Why is that misleading?
Blackadder: The guys on that list who still had performance that justified a major league spot, they all left the game because they just didn’t want to play anymore. I am pretty sure that if Ted Williams had wanted to keep playing, he could have signed a contract.
James: Yes, of course he could have. Mantle could have returned, DiMaggio could have returned, McGwire could have returned, even Musial could have returned if he had wanted to. It would have been misleading if I had suggested otherwise, but I didn’t.
But those players were free to return because they had earned the respect and, to a degree, the affection of their employers. Bonds hasn’t. What does that have to do with anything?
Blackadder: A player of Bonds’ current calibre being forced into retirement is, as far as I can tell, historically fairly unique.
James: Everything about Bonds is unique, and has been for many years. So what?
Bonds’ calibre LAST YEAR is not his “current” calibre. It’s last year. The point that I was making is this: that when a player reaches the point where ALL that he does is hit, he is normally very near to the end. Ted Williams realized this, and walked away after having had a good season. Stan Musial didn’t realize this, stayed around one year too long, and had a difficult last season. Bonds, whose understanding of “gracious” appears as if it might be a little bit limited, has not only not understood this, but has resisted it.
But if you look at old players who have a very high OPS and essentially no other skills, what happens to them is that they suddenly collapse. They go from “valuable” to “out of the game” or “still in the game, but worthless” in one year. A few examples, if you’ll indulge my stretching beyond reason the fiction that this is a dialogue.
Mark McGwire in 2000 had the highest OPS in baseball. One year later, he was finished, and he retired.
Mike Schmidt in 1986 was the National League’s Most Valuable Player, and was still arguably the league’s best player in 1987. In 1988 he was finished. He hit .249 with 12 homers in 1988, retired early in the season in 1989, hitting .203.
Willie Stargell in 1979 was the National League’s Most Valuable Player. Like Bonds, he was an old left-handed hitter who was limited to about four games a week, but he could still crush a baseball. His career totals after that season: 335 at bats, 14 homers.
Henry Aaron in 1973 was limited to 120 games and 392 at bats, but he had the highest OPS in baseball. In fact, the experience of Aaron, at the end of his career, is almost as perfect a parallel to Bonds as is Ruth. The Braves didn’t want him any more. Like the Giants, they had built around him and built around him, waiting for him to break the home run record, and once that was done they were anxious to get him out of there and get on with their lives.
As the Brewers saw it—exactly the same way you see Bonds now—Aaron was still a valuable man. He had the highest OPS in baseball just one year ago (1973). In 1974 his OPS was still in the top ten in the National League—and they’re going to give them to us for free! How can we not do that? But Aaron was finished, and the Brewers didn’t get better, they got worse. Their win totals dropped from the 70s to the 60s.
In 1971 Willie Mays had the fifth-highest OPS in baseball, in large part because he drew 112 walks in 136 games. In very early 1972 the Giants decided that he was finished and they had to move him out of there—and, in fact, he was finished. It worked out OK for the Mets, who stumbled into the World Series with 82 wins in ’73, but Mays was a part-time player and not very good. My point is that the Giants hustled Willie Mays out of town after a season in which he had the fifth-highest OPS in baseball--just as they did Bonds.
In 1962 Stan Musial, who had announced his intention to retire that spring, had the sixth-highest OPS in baseball. He came back for one more year—but all of a sudden, he was finished.
Ted Williams in 1960 had the highest OPS in baseball—and retired.
In 1954 Jackie Robinson, batting 386 times, had one of the top ten OPS figures in the National League. In 1955 he was done.
In 1950 Joe DiMaggio had the third-highest OPS in baseball. In 1951 he was finished, and he retired after the season.
In 1946 Hank Greenberg led the American League in Home Runs and RBI, had the second-highest OPS in the American League, behind Ted Williams. That winter he got into a contract dispute with the Tigers, and the Tigers traded him to Pittsburgh because Pittsburgh was the only team that stepped forward to meet his contract demands.
It was a disaster. Greenberg hit .249, although, because of his walks, he still had a high OPS. Pittsburgh finished last for the first time since 1917. Greenberg retired after the season.
In 1945 Mel Ott had the third-highest OPS in the National League. The entire rest of his career consisted of 72 at bats, with a batting average of .069.
In 1944 Indian Bob Johnson had the highest OPS in the American League. He retired after the 1945 season.
Edgar Martinez in 2003 was still a top-level DH. In 2004, all of a sudden, he was done. He retired after the season.
Jason Giambi in 2006 had an OPS of .971, one of the highest in baseball. In 2007 he suffered a dramatic dropoff. He may not be finished; maybe he’ll come back, but. . .it wasn’t a good experience.
The normal rule, throughout all of baseball history, has been that when you have an aging superstar whose skills narrow to the OPS categories, and particularly when that player has to start taking two days a week off, he goes from “still has a very high OPS” to “finished” in a very, very short period of time. This is the way it has always been, back to the time of Dan Brouthers and Sam Thompson.
Yes, the exact scenario by which this plays itself out is different for Bonds than for many other players—but that scenario is normally painful, and normally controversial for the team. Babe Ruth’s exit from New York was excruciating, for Ruth and for the Yankees. Joe DiMaggio in 1951 was ripped in a national magazine—I think it was the Saturday Evening Post—and retired in embarrassment. Jackie Robinson was traded by the Dodgers to their hated rivals, the Giants, and refused to report. The Giants were ripped for trading Willie Mays, while Mays himself was humiliated by his inability to play defense in the 1973 World Series. Aaron didn’t appreciate being pushed out of Atlanta. Greenberg’s exit from Detroit was a massive controversy, and left everybody unhappy with the outcome—the Tigers, the Pirates, and Greenberg.
The specific scenario of Bonds’ exit is a little unique, but then, everybody’s is, and everything about Bonds is. Bonds hasn’t had the grace to walk away with his head held high, as Ted Williams did and DiMaggio did. He doesn’t have the organizational support to stay on with diminished skills, as Musial did or Mantle did. He is leaving as a free agent, rather than as a marked-down, discontinued item in trade, as Mays did and Aaron did, because this is the free agent era rather than the era when a player belonged to a team until the team didn’t want him anymore.
But a) it has always been extremely common for aging superstars to make a very quick trip from “high OPS” to “out of the game”. And b) history shows that when players don’t accept that end, and when some team then steps forward to provide a showcase for the superstar to continue his storied career, it turns out to be very disappointing experience for both the player and the team—in fact, I would argue that “embarrassment” and “disaster” are the only outcomes on record for that scenario. “Mskarpelos” argues that that includes the San Francisco Giants of 2006 and 2007—they provided a showcase for Bonds to continue his personal quest, and it was a disaster for them, as it was for the Braves in 1935, as it was for the Pirates in 1947, as it was for the Brewers in 1975-76.
Maybe this time will be different. Maybe, if you’re the team that steps forward to allow Barry Bonds to continue his career, maybe this time it works out for you. God bless you, and I hope that flies. I ain’t lining up to be the test pilot.