In Wednesday's column about Terrell Owens' reported suicide attempt, which he says was a reaction to painkillers, not an attempt to hurt himself, I compared Owens' persona with that of Dennis Rodman and made what I thought was an offhand comment that Owens is "a much better football player than Rodman was a basketball player."
Imagine my surprise when several letter writers, both on the site and in my in box, took exception to that idea, arguing that Rodman was, in fact, better than Owens.
I love this sort of argument, and since I don't feel like writing about the boring baseball pennant races, the stumbling of not-even-mediocre teams toward the last few National League playoff spots, that's what we're going to talk about today.
I shouldn't have been surprised. This has come up before, when I ridiculed the trio of economics professors who wrote a book called "The Wages of Wins" for asserting that Dennis Rodman was a more valuable basketball player than Michael Jordan. The authors actually backed away from that assertion by simply claiming they'd never made it, though they had.
One letter Wednesday argued that ">
Rodman was one of the best rebounders of all time because that's all he did. He was a liability on offense. He was certainly an element of teams that won championships, and I'd agree he was a key element of the Pistons' title teams. That was before he became a specialist and actually played the entire game. But would the Bulls have won those three late titles in 1996-98 without him? Well, they won three without him in the early '90s.
This letter argues, I think disingenuously, "Though it's true that Rodman was never a complete player -- at his best, he was a defensive specialist who also rebounded, and was never a scorer -- no football player is really expected to be a complete player. People don't criticize Terrell Owens for his inability to rush the passer."
Using the same logic, I'm more valuable to society than a brain surgeon because I write about several different sports, but a brain surgeon just operates on brains.
Football as played today is a game of specialists. Basketball is not. It's not part of Owens' job, or anyone like him, to rush the passer. It's part of every basketball player's job to score, pass, rebound and play defense. Rodman ignored one of those jobs, arguably the most important.
I think what's happening here is that people are vastly overvaluing rebounds.
First of all, some of Rodman's rebounds were going to happen anyway. During his three years with the Bulls, Rodman averaged 17.7 total rebounds and 6.4 offensive rebounds per 40 minutes. The league average at that time was 6.9 total and 2.1 offensive boards per 40 minutes of playing time.
So what Rodman did better than an average player -- not an average power forward, mind you, but an average player, including guards -- was collect about four extra offensive rebounds and seven extra defensive rebounds per 40 minutes. That's good.
He bought the Bulls four extra offensive possessions with those offensive boards, and in those years the Bulls averaged 1.13 points per possession, so we can figure his offensive rebounding gave the Bulls an extra 4.5 points per game.
It's harder to tell how many points those seven extra defensive rebounds accounted for. It would be 7.91 if we figured that none of those seven would have been grabbed by a teammate if Rodman were an average player, but I don't think we can assume that.
Scottie Pippen, Toni Kukoc, Luc Longley, Bill Wennington and Jordan were all on the '95 Bulls, the year before Rodman got to Chicago, and they were all on the '96-98 championship teams with Rodman. During the three title years, those five players averaged 5.8 fewer defensive rebounds per 40 minutes than they'd averaged in 1995.
Jordan retired after the '98 season, but in 1999, all four of the others -- Pippen and Kukoc for other teams -- got more defensive rebounds per 40 minutes than they'd pulled down in any of the three years they'd played with Rodman. Rodman got a lot of rebounds, but he probably took more of them from his own teammates than he took from opponents.
I think we should figure Rodman added a net one or two defensive rebounds per 40 minutes over an average player when you consider the six rebounds his frontcourt mates weren't getting because of him, but just for the benefit of the doubt I'll say he added four. That would be another 4.5 points a game, for a total of about nine points a game he added with his rebounding over an average player.
Or, as we'll see, about one point less than he cost his team with his lack of scoring.
As for his defense, while it's true that Rodman was first-team all-defensive seven times, only one of those times, 1996, was during his Bulls years. In his last two years in Chicago, he didn't even make third team.
And while rebounding is important, it's not so important that it propels a player who does it well to elite status. Even if he's particularly adept at offensive rebounding. If rebounding were that important, it would correlate reasonably well with winning.
In 1996, the Bulls were second in the league in offensive rebounds. First: the New Jersey Nets, who went 30-52. The Bulls were third in total rebounds, behind the Nets and the Portland Trail Blazers (44-38).
In 1997, the Bulls were third in offensive rebounds. First: the Nets (26-56). Second: the Philadelphia 76ers (22-60). The Bulls were second in total rebounds behind the Nets.
In 1998, the Bulls were third in offensive rebounds. First: the Nets, who improved to 43-39. Second: the Golden State Warriors, who rode their offensive rebounding prowess to a 19-63 record. In total rebounds the Bulls were second to the Warriors.
Rebounding totals don't seem to correlate very well with winning, wouldn't you say? I'm cherry-picking results here, but feel free to look over the rebounding totals from any year. The best rebounding teams and the best teams are not largely the same teams.
The Nets' big rebounder in those years was Jayson Williams. I suppose Jayson Williams was a better basketball player than Terrell Owens is a football player too?
I don't think so, but I do think he was a better basketball player than Dennis Rodman was, though I'll concede a good argument could be made for Rodman.
But it's an argument. They were pretty similar in value.
For every 40 minutes that Williams played in those three years, he had 15.1 points, 15.9 total rebounds and 7.3 offensive rebounds. Rodman was good for 6.1 points, 17.7 total rebounds and 6.4 offensive rebounds in those three seasons. Part of the Rodman argument is that he was a good passer, so let's also consider assists per 40 minutes, which was Rodman 3.3, Williams, 1.2.
So, if you were to trade Williams for Rodman you'd lose nine points and an offensive rebound -- that's right, Jayson Williams was a better offensive rebounder than Rodman the nonpareil -- and gain about three defensive rebounds and two assists.
Let's wildly overvalue rebounds and assists and assume that every rebound, even a defensive one, and every assist is worth two points. That is, every rebound and every assist led to a basket that would not have been scored otherwise, which is of course a ridiculous assumption.
Rodman's two-rebound, two-assist advantage over Williams would translate to eight points -- one fewer than Williams' scoring advantage. In the real world, two rebounds and two assists don't come close to being worth eight points.
Here's another way to look at it. During those three years the average NBA team scored 96 points per game. A game is 240 player minutes (48 minutes times five players), so you could say the job of each player was to score 0.4 points per minute on the floor, or 16 points per 40 minutes. Williams did that. Rodman didn't do half of it. He was 10 points shy.
Of course, Rodman had Michael Jordan, who scored 31 points per 40 minutes, more than making up for Rodman's deficit. Imagine what Jordan would have done if he hadn't had to play in a four-on-five offense! During the first three-year championship run, in 1991-93, without having to beat his own guy and Rodman's, Jordan scored 32.7 points per 40 minutes.
I'm oversimplifying. There are other things to look at. Williams got an extra steal every 15 games or so, and blocked one shot every 40 minutes to 0.4 for Rodman, the defensive nonpareil.
But there's no need to reinvent the wheel. Others have developed ways of measuring how good a player is. John Hollinger's player efficiency rating, or PER, Dean Oliver's approximate value, or AV, and the NBA's own efficiency, or Eff., are three of them.
We'll start with PER, which I think is the best of them. It measures per-game value and sets up so that 15.0 represents the league average. League leaders are in the high 20s -- Dirk Nowitzki and LeBron James led the league at 28.1 last year; Michael Jordan nudged over 31 a few times. Scrubs are around 10 and below.
So Williams was at least a league-average player and often better in those three years. His 18.7 year rightly earned him a trip to the All-Star Game -- where he had 10 rebounds in 19 minutes, by the way. Rodman was a below-average player by this metric.
Incidentally, in his early years in Detroit, before he became a rebounding specialist and was willing to put the ball in the basket to the tune of about 17 points per 40 minutes, PER rated Rodman as an above-average player. In five of his first six years, his PER was above 15, though never by much. He was essentially a standard rebounding forward, a little better than league average, and remember that a league-average player is a pretty good player.
Also remember that Terrell Owens is much better than a league-average player.
Here's AV, which I'll explain after the chart:
Rodman comes out better by this measure. Here's Oliver's explanations for the relevant scores:
Six or seven indicates an average bench player or a good player playing under 1,500 minutes, which is about 18 minutes a game. Nine indicates an average regular or a good sixth man. Ten indicates an average regular or a very good sixth man. Eleven indicates an above-average regular or an excellent player playing about 1,800 minutes, or around 22 minutes a game.
Approximate value measures total season contributions. The NBA's efficiency rating, like PER, measures a player on a per-game basis.
Not sure why the NBA system downgrades Williams' '96, but never mind. The point is Williams and Rodman were roughly comparable players in 1996-98. Some metrics show Rodman to be the better player, others Williams.
None of them indicate that either player was significantly better than a league-average contributor. Rodman helped the Bulls win three championships. Williams helped the Nets be lousy twice and mediocre once. If they'd been traded for each other, it's doubtful their teams' fortunes would have changed much.
And one of these guys was better at basketball than Terrell Owens is at football?
Terrell Owens is an elite player. In the nine seasons he's played at least half of his team's games, he's gone to the Pro Bowl five times. He's 22nd on the all-time receiving list for catches, 20th for receiving yards and fourth for receiving touchdowns.
The more sophisticated metrics at Football Outsiders show that Owens was consistently among the top half-dozen receivers in the league during his prime, in the early years of this decade. Considering that in those days there were 62 starting wide receivers in the league, I'd say that's significantly better than league average.
And those stats "measure only passes thrown to a receiver, not performance on plays when he is not thrown the ball, such as blocking and drawing double teams."
There's little dispute that a good deal of Owens' value is in drawing the attention of the defense away from other receivers and the running game. Is there any other way to explain Todd Pinkston and Freddie Mitchell suddenly turning into good receivers in 2004, when Owens arrived in Philadelphia? And how are they doing now? Answer: They're not.
Oddly, Owens' ability to draw extra defensive attention is the exact opposite of Rodman's main liability, the fact that opposing defenses had no need to worry their pretty little heads about him.
As for all those rings Rodman has, well, Steve Kerr has a bunch of rings too. Was he a better basketball player than Owens is a football player? Shall we talk about three-time World Series winner Luis Sojo, and whether he was better than one-time winner Rogers Hornsby?
Owens is one of 11 men on the field at one time, and it's less than half the time. If you want to argue this way -- and I don't -- let's not forget what happened to the Eagles, who failed to make it to the Super Bowl year after year, the one full year they had Owens.
Plenty of basketball players much better than Rodman didn't win as many rings, and plenty of football players not nearly in the ringless Owens' class -- or even in Rodman's -- have won them. Let's not have any more of this nonsense about Rodman being better than Owens, or even better than just OK.
Tomorrow: Nina Simone was a better singer than Dave Schultz was a hockey player. Discuss.
This story was corrected after publication.
Previous column: T.O.'s trip to the E.R.
- - - - - - - - - - - -