The post “Kobe Makes Pau Gasol Unhappy” has apparently made fans of Kobe unhappy. Such fans are quite interesting. As many people have learned, any suggestion that Kobe is not the greatest basketball player in both this world and all future worlds is quickly met with a great deal of anger in the Internet. And this anger is often part of a package that includes misspelled words and poorly constructed arguments (which readers such as Simon, jbrett, ilikeflowers, and others are doing a great job of adddressing).
Unfortunately, I wish to tell another story today that involves Kobe. My story is actually focused on Clyde Drexler, but I think it’s a good idea to begin with a direct comparison of Clyde the Glide and Kobe.
Table One: Comparing the Career Performances of Kobe Bryant and Clyde Drexler
As Table One reports, Drexler’s career averages top Kobe’s marks with respect to shooting efficiency, rebounds, steals, blocked shots, and assists. And yet Kobe is considered by many to be the better player.
Why is Kobe Better?
There appear to be three explanations for why Kobe is thought to be the better player. First – as Table One notes – Kobe is the more prolific scorer. Of course, this is because Kobe leads Drexler in field goal attempts.
Another issue is that Kobe spent his career with the Lakers while Drexler played for Portland and Houston. In general, players for teams located in LA and New York tend to get more media exposure and therefore are thought of as better players.
And then there is the issue of championships won. People tend to think players on championship teams are better than those who toil for teams that tend to lose in the playoffs. It’s easy to point out the absurdity of such logic. Teams win championships and one can pick up a ring just because you happen to have the right teammates. After all, does anyone think Luc Longley (three titles) was a better center than Patrick Ewing (0 titles)? Or that Robert Horry (seven titles) was a better forward than Dominique Wilkins or Karl Malone (0 titles)? Despite such obvious arguments, people will note that Kobe’s four titles must mean he’s a better guard than Drexler (1 title).
Drexler Second Best
Drexler did appear in the NBA Finals four times. But his team came up “second-best” three times. And that appears to be the story of Drexler’s career. He often toiled for the “second-best” team.
Back in 1990, Drexler and the Portland Trail Blazers finished the season with league’s 4th best efficiency differential (offensive efficiency minus defensive efficiency). After defeating the Phoenix Suns – the team with the best differential – in the Western Conference Finals, the Blazers faced the Detroit Pistons in the NBA Finals. The Pistons differential of 6.2 was only slighly better than the Blazers mark of 5.9. But after seizing homecourt advantage in the series, Portland went on to lose three consecutive games at home.
The next season the Blazers were even better. When the regular season ended the Blazers differential of 8.3 was only bested by the 9.2 differential posted by the Chicago Bulls. But in the Western Conference Finals the Blazers were defeated by the LA Lakers.
In 1991-92 the Blazers again posted the best differential in the Western Conference. Their mark of 7.0, though, lagged far behind Chicago’s 10.6 differential. Hence it was not a surprise that the Blazers were defeated in the NBA Finals.
Across the next two seasons the Blazers slipped. Then in 1994-95, Drexler was traded to the Houston Rockets. And despite the fact the Rockets were underdogs in all four rounds in the playoffs, Houston managed to repeat as NBA champions.
The next season the Rockets were swept in the second round of the playoffs, while the Chicago Bulls won 72 games and the NBA Title. At this point it looked like Drexler was never going to contend for another NBA championship.
What Might Have Been in 1997
But then the Rockets made a move that looked to return this team to contending status. In August of 1996 the Rockets traded for Charles Barkley. This acquisition gave the Rockets an impressive trio of Hakeem Olajuwon, Drexler, and Barkley. These three produced 45.2 wins in 1995-96, so it wouldn’t take much more for the Rockets to contend in 1996-97.
When we look at what the Rockets did that season, though, it seems like the “much more” never arrived.
Table Two: The Houston Rockets in 1996-97
As Table Two indicates, the Rockets won 58 games in 1996-97. But the team’s efficiency differential – and corresponding Wins Produced – is more consistent with a 52 win team. After the aforementioned trio, the Wins Produced of the rest of the roster was only 14.1.
So there wasn’t much after Hakeem, Clyde the Glide, and Sir Charles. Plus, Hakeem – who was 34 years old, saw his production decline. Had he maintained what he did in 1995-96 the Rockets would have been five wins better.
Such an improvement, though, would not have been enough to catch the Chicago Bulls and the Utah Jazz. There was a move made, though, that should have made a difference. A month before the Rockets acquired Barkley, Houston signed Brent Price. In 1995-96, Price produced 9.8 wins and posted a 0.230 WP48 [Wins Produced per 48 minutes] for the Washington Bullets.
As Price’s bio at NBA.com reveals, his 1996-97 season went off the tracks before it began:
Price was expected to step in as Houston’s regular point guard after he joined the Rockets as a veteran free agent, but a broken left humerus suffered in a preseason game on Oct. 24 ruined those plans. He did not make his debut until Dec. 28, and just when he seemed to be playing his way back into top form he suffered a torn anterior cruciate ligament in his right knee in a game against the Los Angeles Lakers on Feb. 25. He was placed on the injured list on March 3, underwent surgery on March 15 and sat out the remainder of the regular season and postseason…Price was on a roll when he got hurt the second time. In the eight games prior to his torn ACL, Price had averaged 10.1 ppg and 4.9 apg while shooting .482 from the field and .462 from behind the arc…That spurt included a season-high 20 points on 8-for-11 shooting, plus a season-high six assists, in a 106-97 win over Vancouver on Feb. 11.
Because Price was hurt the Rockets had to turn to Matt Maloney. An undrafted rookie out of the University of Pennsylvania, Maloney started every game in 1996-97 and produced 2.7 wins with a 0.053 WP48. Had Maloney’s minutes gone to the Price we saw in 1995-96 (and Maloney took Price’s 1996-97 minutes), the Rockets would have been 11.3 wins better. Coupled with Olajuwon maintaining what we saw in 1995-96, the Rockets would then have been transformed into a team with 69.8 Wins Produced.
This mark might have allowed the Rockets to challenge the Chicago Bulls (who won 69 games in 1996-97). And it’s possible Drexler might have won a second title.
Of course, Drexler, Olajuwon, and Barkley would have still had to defeat Michael Jordan, Dennis Rodman, Scottie Pippen and company in the NBA Finals; and even with Price and Olajuwon performing as we saw in 1995-96 the Rockets were not really better than the Bulls. So Drexler might have still finished second best to Jordan.
So even this “what-if” tale is consistent with the story of Drexler’s career. Even if the breaks did go his way, he was still not as good as Jordan. But that really isn’t such a shame. Second to MJ is still pretty good.
After all, Kobe Bryant – how is also pretty good — is also second to Jordan. But Kobe is not – as the above table indicates – as good as Drexler (yes, I could have let it go but I didn’t).
– DJ
The WoW Journal Comments Policy
Our research on the NBA was summarized HERE.
The Technical Notes at wagesofwins.com provides substantially more information on the published research behind Wins Produced and Win Score
Wins Produced, Win Score, and PAWSmin are also discussed in the following posts:
Simple Models of Player Performance
What Wins Produced Says and What It Does Not Say
Introducing PAWSmin — and a Defense of Box Score Statistics
Finally, A Guide to Evaluating Models contains useful hints on how to interpret and evaluate statistical models.
Wages of Wins Journal 
tywill33
March 17, 2010
I have to make this quick comment.
You’re absolutely right about the existence of what I call the “Kobe Defense League”. The mere suggestion that Kobe is not God’s gift to basketball and they come at you as rabidly and vigilantly as the JDL.
Funny story. One time on the old Bucks Diary I posted something that suggested — very mildly –that Kobe was a second tier star. That he was good but probably not as good as popular opinion holds.
Wow, did the roof come in on my comment section. One guy sends me back what stands as my favorite comment ever, and it said in relevant part “Kobe Bryant is a great man, you are not a great man, how dare you compare yourself to his greatness!”. He claimed I was somehow comparing myself to the “Great” Kobe Bryant.
In my mind that stands as the quintessential Kobe Defense League comment.
Joe
March 17, 2010
Didn’t Drexler play a lot of his career minutes at the 3?
I suppose Bryant hasn’t decline yet, though, either so his WP number is artificially high.
I guess things just even out to Kobe being a consistent top 20 NBA producer-“ish” over the course of a decade…. still impressive.
Alvy
March 17, 2010
I understand the Robert Horry rebuttal and why it’s necessary, but usually, (at least I think so), people mean X player has more titles than Y player because of the production X player has contributed to a championship team, and not just a reference to some nominal total. I dunno, maybe I’m wrong and it’s just an internet-NBA thing I don’t get.
BTW, fans have probably forgotten how good Drexler was (something fairly typical in NBA culture for some reason). In fact, when Portland was in the Finals against Chicago, the match was sometimes labeled as …Michael Jordan, and the Jordan of the West, so Drexler wasn’t some obscure figure, nor is the “media exposure” reasoning entirely necessary.
Or have I forgotten what the 90’s NBA media was like?
Pat
March 17, 2010
While Drexler may have finished some games at the 3 while playing with Porter and Ainge, he played the bulk of his career in Portland at the 2, with Jerome Kersey at the 3.
Italian Stallion
March 17, 2010
I also think the “number of titles” argument is pretty much bogus in a team game. A great player cannot win a title without a lot of help, but a mediocre player can win a lot of titles if he happens to be on a team of very productive players.
However, at the margin, when we are talking about great players that have actually been in a position to win multiple titles, I tend to give some extra weight to those that actually captured them.
I doubt I’ll get much agreement here, but I believe there are issues of competitiveness, determination, heart, experience, ability to perform under pressure, ability to rise up to a higher level for a short period of time when required etc… that sometimes separate athletes.
These kinds psychological and personality attributes are less apparent and measurable in a team game, but IMHO they exist.
So at the margin, if you show me two players that are very similar statistically where both were often on title contending teams, but one won the title multiple times and other not at all, I am going to prefer the proven winner.
That would go double if I could identify instances of the non winner having trouble performing well with it all on the line.
Italian Stallion
March 17, 2010
One thing that I hope makes it into the public record soon is a better breakdown of lifetime shot locations.
I know there are a few web sites that track that kind of thing currently, but I’d like to see it in a more detailed format and for lifetime past performances.
My view is that some decent outside shooting is a requirement of a successful team basketball, but for multiple reasons it’s a less efficient way of scoring than post scoring, slashing, and other general inside scoring.
It seems to me that whoever has a greater role in the outside shooting will be penalized statistically relative to those that score more on the inside and also draw more fouls.
If you could get away without outside shooting, that would be totally appropriate, but if you can’t (and I think you can’t) then IMO we should also study outside shooting efficiency as a separate matter from overall shooting efficiency.
Granted, much of this gets captured in positional adjustments. However, IMHO there are clearly differences between players that play the same position when we are talking about how they do their scoring.
I think this may be especially true for SGs because some of them are hyper efficient slashers and some of them are more in the Reggie Miller mode of outside shooter off screens etc…
It seems to me that the slashers still need a good outside shooter to be teamed with them in order for the team to excel.
Chicago Tim
March 17, 2010
I’ve now finished The Wages of Wins and Stumbling on Wins on my Kindle. I liked them and recommend them. I’ll admit that I don’t follow all the math, and I don’t understand the disputes between various sports statisticians. But you and your co-authors do a good job of making your theories sound plausible to non-experts like me. And although the statisticians seem to disagree about precisely which statistics to use, they tend to agree that GMs in all sports persistently make many mistakes based on conventional wisdom.
I like to have the theories you cite in the back of my head when I read sports stories, so that I can question the conventional wisdom. Maybe it is okay that the Chicago Bears don’t have a first-round draft pick. Maybe the Chicago Blackhawks don’t need to spend big money on a goalie. Maybe the Chicago Bulls don’t need to spend big money on a coach. And maybe the Chicago Cubs and White Sox can improve their farm system by drafting more college players and position players, and fewer high schoolers and pitchers.
In Stumbling on Wins, your latest book, I like the way you link sports stories to the broader economy and even, perhaps, to broader theories of social and political science. Maybe people don’t act rationally, because they can’t compute all the variables and rely on conventional wisdom. Maybe they act irrationally repeatedly, decade after decade after decade. Maybe some of our heroes don’t deserve their fame, and some forgotten people should be remembered and given their due. If highly-paid experts make these mistakes in sports, a relatively-closed environment with many objective measurements of performance, how much more likely is it that we make such mistakes outside of the narrow world of sports?
I would have liked even more stories, and a more detailed discussion of sports like hockey and soccer. But I guess I’m just greedy.
I will say that having read this blog pretty regularly, there were some stories I recognized and already knew. However, in the books you do present the arguments more systematically and thoroughly, and there is much in the books that I do not remember from your blog. I also found some redundancy between the first book and the second, since you couldn’t assume that everyone was familiar with the first. But you put the more complex equations in the appendices, where I was free to skip or skim them.
Finally, I like your dry sense of humor, of which there are many examples in both books and in this blog. If anything, you undersell your theories. Some of them are pretty shocking, such as the possibility that black quarterbacks are systematically undervalued.
One nitpick: in most cases “use” can be substituted for “utilize.” In rare cases, “utilize” is more appropriate, that is, when you utilize an object for a purpose not originally intended, such as picking your teeth with your knife.
Lior
March 17, 2010
The players who are remembered over time are those who got the media exposure. This means things like winning championships, featuring in highlight reels, having all-NBA team and all-star team selections. Playing in NYC and LA counts for a lot. That’s what the media remembers, so that’s what the general public remembers (especially since most of the public wasn’t around to watch these players at the time).
Lior
March 17, 2010
IS: Yes, the average SG has a lower shooting percentage than the average inside player. All this means is that “above average” shooting percentage means something different for a SG than for a forward. The position adjustment in PAWS and WP48 is exactly about these issues. This is also why the tables posted by Prof. Berri always compare players to the average at their position and not to the league average.
Evan
March 17, 2010
Dave, admit it, that last comment reviewing your books was you. Just in case you get another deluge of Kobelovers, you put the comment up to try and sell them some books. :)
unfortunately, i don’t think the Kobelovers are persuaded by things like science, math, statistics and logic.
John Giagnorio
March 17, 2010
Just keep posting this stuff, Dave. It’s fun to watch the Kobe Bryant apologists squirm :)
dberri
March 17, 2010
I like the review from Chicago Tim. I think that needs to go into a post by itself. And no, that wasn’t me. I would never nitpick with myself.
too many steves
March 17, 2010
At least in his prime, Kobe was a much better defender than Drexler ever was. I know, WoW is predicated on the idea that nobody is better than anybody else on D, but the rest of us don’t have to believe that.
So, I’d give Kobe a very slight edge offensively and a big edge defensively, making him the better player overall. Drexler was better than most people remember, though.
By the way, Henry Abbott has a very good appreciation/critique of WoW today on TrueHoop.
VH
March 17, 2010
It doesn’t help that Drexler was consistently dominated by another player at his position. Although you can make statistical arguements against Kobe, there has never been a situation where Kobe was dominated by an opponent the same way that Jordan did to Drexler. Those things hurt your legacy.
ilikeflowers
March 17, 2010
too many steves,
It’s hard to see where anyone could reasonably give an edge to Kobe offensively based upon the objective evidence. Other than FT%, he only leads Drexler in usage stats: attempts, points per game, and minutes per game. I can only guess that you’re talking about the offensive skills aesthetic that is sometimes presented as a measure for being offensively superior. It’s interesting to note that Drexler has better ‘defensive’ stats as well: more steals, blocks, and rebounds (I don’t know how to value fouls defensively since there are arguments both ways about whether more or less fouls are a useful proxy for defensive ability and/or effort). These are really our only available objective measures of defense. Of course, given the lack of more specific defensive stats it’s reasonable to come to whatever conclusion one wishes on defensive ability. Clyde does appear to be definitively superior to Kobe offensively however.
John Giagnorio
March 17, 2010
If Bryant is so obviously better, why is the difference in rebounds, steals, and blocks so pronounced? Seems like these things should have at least some relation to defensive ability.
For that matter, a statement like “Kobe was a much better defender than Drexler ever was” should be very easy to support (if true), and yet you don’t bother to provide any evidence at all. Why?
John Giagnorio
March 17, 2010
“It doesn’t help that Drexler was consistently dominated by another player at his position. Although you can make statistical arguements against Kobe, there has never been a situation where Kobe was dominated by an opponent the same way that Jordan did to Drexler. Those things hurt your legacy.”
Huh?
ilikeflowers
March 17, 2010
Two things that would be very interesting to see would be :
[1] player productivity stats against the top 50th and 25th percentile teams
If Kobe’s stats are superior to Drexler’s against the more elite teams then he can definitely be considered a better player yet at the same time be less productive. Using top 50th and 25th percentiles should be sufficient sample size especially across multiple seasons.
[2] player productivity variance.
It’s entirely possible that Kobe has a higher performance ceiling and variance than some of the other top SG’s. If this were the case then both sides of the Kobe is Jordanesque debate could be correct. Maybe Kobe can produce at MJ levels but just hasn’t done it as consistently.
VH
March 17, 2010
Jordan dominated Drexler every time they matched up. In the finals, in the regular season. That makes him look bad in retrospect. Is that clear enough for you?
mick
March 17, 2010
there will certainly be disagreement over the claim that Drexler was better than Kobe but lets agree on one thing: Clyde “The Glide” is a far better nickname than “Black Mamba”
too many steves
March 17, 2010
I’m looking now at the Basketball-Reference pages for both guys, and they seem to back up my initial thoughts. I don’t like comparing career numbers because Clyde’s career is over and Kobe’s isn’t, but … they do look pretty similar. Kobe has a slightly higher ts%, since he’s a better 3-point shooter. Drexler’s a better rebounder and has a slightly higher assist%, but also a higher turnover rate (maybe that’s because Clyde looked at the ball while he dribbled.) I would bet that Clyde got way, way more easy shots, though. The styles were so different — Portland in 91 scored 115 a game, while the Lakers in their championship years never averaged more than 100 or so. That’s a lot of fast-break dunks for Clyde.
It’s true that defense is harder to quantify (and I don’t think steals and blocks are very good measures at all). I remember Kobe back in the day as a great defender, and I don’t remember Clyde that way, but they had the same career defensive rating, 105. Drexler’s best year was a 103, though, and Kobe got all the way down to 98 once and 102 a couple of times.
Anyway, it’s a tough call, and anybody who says one guy is obviously better is fooling himself. If I had to pick one guy to build a team around, I’d take Kobe. I think he is/was better defensively and more versatile on offense — he can create in the halfcourt in ways that Clyde couldn’t, and he has better range. Clyde was a clearly better rebounder and probably a better finisher in the open court. I’d say they’re even as passers. But like I said, it’s close.
Megalo Arenas
March 17, 2010
What the fuck? Barkley was the most productive rocket? I don’t doubt your analysis, professor, but man I did not see this coming.
Then again, Barkley getting more rebounds might explain why Olajuwon’s production declined. Solid post.
ilikeflowers
March 17, 2010
too many steves,
Basketball Reference has Drexler and Kobe with identical turnover rates, 2.9 per 36 min.
ilikeflowers
March 17, 2010
I looked at Kobe’s and MJ’s best WinScore seasons (02/03 and 88/89) to look at the variation in an effort to see if it would help to explain the perception of Kobe vs MJ in some quarters. In the end all it did was really hammer home yet again how much better MJ was than Kobe.
________WS48Avg_WS48Std_WS48Max_WS48Min
88-89_MJ___0.39____0.16____0.93____0.07
02-03_Kobe_0.24____0.17____0.73___-6.50
MJ has 4 WS48’s greater than Kobe’s best and Kobe has 10 WS48’s lower than MJ’s worst. Kobe even has 4 games in the negatives (with greater than 32 minutes in each).
Maybe Kobe = MJ can be better understood by taking into account variation when comparing his best with MJ’s typical?
Or maybe it’s so rare for SG’s to produce in the upper ranges that Kobe can reach that people are less appreciative of how much consistently MJ produced at those same levels?
too many steves
March 17, 2010
ilikeflowers,
Not turnovers per 36 min, turnover percentage (tov%). It’s an estimate of turnovers per 100 plays. Better than TOs per game or per minute, because not everybody handles the ball the same amount.
ilikeflowers
March 17, 2010
too many steves,
We know how many minutes they play. How do we know how many possessions they play? If it’s being estimated then how can it be more accurate than using minutes which are measured?
Italian Stallion
March 17, 2010
Lior,
“IS: Yes, the average SG has a lower shooting percentage than the average inside player. All this means is that “above average” shooting percentage means something different for a SG than for a forward. The position adjustment in PAWS and WP48 is exactly about these issues. This is also why the tables posted by Prof. Berri always compare players to the average at their position and not to the league average.”
I understand you, but this is the trouble I have with the idea conceptually.
Suppose team “A” has a SG that is primarily a slasher that’s great at getting to the hoop, drawing fouls, and scoring very efficiently.
Suppose team “B” has a SG that is primarily an excellent outside shooter that uses screens and other open outside shots to score.
It is likely that the SG from team “A” will measure as the more efficient scorer and better player.
However, if you were building team “A”, I don’t think you could add another two of those more efficient slasher types at the PG and SF and expect optimal results. You’d need some really good outside shooting from somewhere.
The exact opposite would be true of team “B”. Since you already have the solid outside shooting, you might want to add a slasher type at the PG or SF.
I guess what I saying is that positional adjustments go a long way towards compensating for these things, but they don’t go all the way because within each position, there are players with very different skill sets.
If we had shot location data, I think we could then start comparing subgroups within the position.
We could ask how much better Jordan was than Wade or how much better Reggie Miller was than Alan Houston, but we wouldn’t bother comparing the first two to the 2nd two because they have different but required roles on their respective teams even though they all play the same position.
(of course, there are also many guys in between too)
too many steves
March 17, 2010
ilikeflowers,
I don’t know the specifics of how possessions are measured — I assume it’s just adding a team’s shots with its turnovers. It shouldn’t be that difficult to measure.
Raw turnover #s don’t mean much. You’ve got to relate it to the player’s usage rate or at least the team’s total possessions. If I touch the ball 10 times a game and turn it over twice, that’s a lot worse than a guy who touches it 40 times with 3 turnovers.
brgulker
March 18, 2010
Not necessarily.
If I touch the ball ten times, turn it over twice, take eight shots and make 7 of them, I’ve done more to help than hurt my team (I think).
If I touch the ball 40 times, turn it over 3 times, take 26 shots and make 10 of them (a recent Kobe shooting performance), in spite of my relatively low TO%, I’ve still hurt my team badly.
I agree with you insofar as I don’t think you can take TO/Game in isolation, just like you can’t take PPG in isolation. But you can’t say definitely that 2 out of 10 is worse than 3 out of 40 unless you have more context.
Anon
March 18, 2010
IS: My view on what you said about needing the shooter is that it ultimately doesn’t matter much. I agree that you need someone who is able to shoot the ball. I don’t think anyone disagrees that you need a variety of different skills on the court. But the argument that a team of slashers fails won’t be successful I believe is based on the theory that the defending team can simply clog the paint and the slashers will struggle. But if you put a shooter or 2 with those slashers then the defense is forced to make a decision: stay with the shooters or help on the slasher. if they move away from the shooter, a GOOD slasher should realize that and kick out for a high percentage shot. if they stick on the shooters, the slasher should be able to get a decent shot near the basket using his skills.
Just like you can’t have a team of all slashers, you can’t have a team of all shooters either. both of those teams will make it easy on the defense. The two skillsets clearly help each other to succeed.
So i don’t think that i agree that the shooter will necessarily be less productive in a good offense. if the shooter is good and getting a fair amount of good shots, he’s going to have a great true shooting percentage.
Jason Kidd I think is a decent example of this. he’s not even really a great shooter, but because the offense does a good job of setting him up for open jumpshots, he looks super efficient even as his skills decline. A jumpshooter who uses his skills to take contested 3s has very little value imo.
and clearly the most productive players are the ones who can do multiple things. for example being a good slasher who can also shoot a bit and grab some rebounds or something.
a slasher (or shooter) who does nothing else is probably not going to get any minutes unless he’s on a terrible team (or he’s really really really good at what he does).
Just some thoughts off the top of my head, not as thought out as they could be.
kjb
March 18, 2010
Wait, Jordan dominated Drexler EVERY time they faced each other? Umm, not so fast.
When looking at their numbers head-to-head, it’s apparent that Jordan was more productive, but it’s inaccurate to say he “dominated.” Unless, of course, by “dominated” we mean that he played a little bit better.
Jordan did outscore Drexler by 11.6 points per 48 minutes in their 15 regular season matchups, but Drexler had more rebounds and assists, and committed fewer turnovers. Drexler also shot better from the field. Jordan had more steals and blocks, got the FT line more frequently and shot a better percentage from the line.
Net? Jordan had a per minute Win Score of .269 to Drexler’s .244. On a per game basis, Jordan’s Win Score was 11.1 to Drexler’s 9.6. Comparing Win Score’s game-by-game indicates that Jordan “won” 9 of the individual matchups; Drexler “won” 6.
In terms of “dominance” — I looked at games in which the per minute Win Score difference was .200 or greater between the two players. Jordan had 4 games where he “won” the matchup by that margin or more — including a +.431 whopper on 2/7/93. Drexler had 3 games where he “won” the matchup by .200 or greater.
Drexler’s teams went 6-9 in the regular season against Jordan. It’d be interesting to add the playoffs into the mix, but b-r.com doesn’t have that data available.
So, at least in the regular season, Jordan’s “every time” dominance didn’t manifest itself. Unless we want to redefine the meaning of “dominance.”
mrparker
March 18, 2010
My theory is that Jordan lowered the bar for NBA fans. Because he was the best player and the best scorer fans who watched the NBA in Jordan era have come to equate the two. Before Jordan high scorers were considered dumb losers for lack of a better phrase.
Is it just me or does anyone else read all the Kobe Mafia comments in the Stephen A. Smith voice?
ilikeflowers
March 18, 2010
Adding Drexler’s best season to the mix:
________WS48Avg_WS48Std_WS48Max_WS48Min
88-89_MJ___0.39____0.16____0.93____0.07
88-89_Drx__0.28____0.15____0.67____0.05
02-03_Kobe_0.24____0.17____0.73___-6.50
At least here there’s a better argument if one assumes that Kobe was a better defender. Kobe booms a little higher than Drexler but busts much lower and more often.
jbrett
March 18, 2010
I’m heartened to see the conversation has mostly left Kobe-vs.-Clyde and morphed into a Clyde-MJ comparison; I’d like to think it means Clyde over Kobe is a no-brainer. Memory may deceive, but I think it was clear, back in the day, that Drexler was hands down the best SG not named Jordan; there wasn’t any real debate about anyone being better than Mike, but if there had been any conversation, Clyde is the only one who wouldn’t have been laughed out of it. Wins Produced backs me up; it’s always nice to have an opinion supported by numbers.
too many steves,
To let you know where I’m coming from, personal-baggage-wise, I’ve been a Laker fan for 42 years. (I’m 46.) I would like nothing better than for Kobe to be the best in the game, and I’d be ecstatic if he were better than MJ, who I grudgingly appreciated but could never warm up to. But he isn’t. It’s not close. I won’t waste any more time on the subject; I’ll move on to how much better Clyde was than Kobe is, was, or ever will be.
I realize I’m paraphrasing myself over and over, but the case for Kobe is invariably one of aesthetics; he LOOKS like he’s better than (blank). If you want to tell me that he can do things Drexler (or someone else you choose) couldn’t do, I’m willing to discuss that. The hole in the argument is that, while Kobe can do more things than Clyde could do, Clyde actually DID more things, consistently, all the time. He produced more, and that’s what ultimately matters. While Kobe may wow you or me with the sheer physical skill he exhibits, fact is there are no extra points for pretty, no multiplier for degree of difficulty. If you can convince the NBA to add judges (and if you think the officiating is suspect, just wait!), who will take into account how impressive each play was, then you’ll be watching figure skating–and I’ll go back to football, or some other game with a real scoreboard. I’m willing to give Kobe a boost in the debate for clutch performance; I believe it is significant that he is willing to win a game at the buzzer, and does so, again and again. But the numbers–the actual production–indicate that with Clyde on the court instead of Kobe, the game wouldn’t have NEEDED a buzzer-beater; it would have been over with a few minutes left in the fourth quarter, and Clyde would have been icing down on the sideline. (And MJ would have already been in street clothes, dammit, but that’s another story altogether.)
khandor
March 19, 2010
If you conducted a poll which asked only elite level basketball coaches to choose between having, either, Drexler or Bryant on their team, the results would be a landslide victory for KBB.
Although Clyde The Glide was also a terrific player in his day, an authentic assessment of his actual basketball talent, in comparison with Michael Jordan and Kobe Bryant, would yield the following rankings:
1. MJ [by a rather wide margin]
==========
2. Kobe
3. Drexler [close but still no cigar]
Fortunately …
Analyzing basketball properly involves both Art & Science.
ilikeflowers
March 19, 2010
Here’s something else that can’t be demonstrated. I bet if we used WP48 to pick 10 players from the last 20 years and this team played against a similar team chosen by the elite level (whatever that is) basketball coaches using their authentic assessments (as opposed to the counterfeit methods presented here) that the WP48 team would win more games and have a better efficiency differential.
khandor
March 19, 2010
mikeflowers,
re: “Here’s something else that can’t be demonstrated. I bet if we used WP48 to pick 10 players from the last 20 years and this team played against a similar team chosen by the elite level (whatever that is) basketball coaches using their authentic assessments (as opposed to the counterfeit methods presented here) that the WP48 team would win more games and have a better efficiency differential.”
In the world of “fantasy hoops,” where the only things that mattter are production numbers, I suppose anything is possible. LOL
khandor
March 19, 2010
ilikeflowers,
Please accept my sincere apology for butchering your moniker.
dberri
March 19, 2010
Khandor,
As I have pointed out before…
Analyzing basketball properly requires that you know how to do analysis. And although I am not sure what you are trying to do, I am sure it is not proper analysis. Just stating opinion after opinion with no supporting evidence is not analysis (and yes, relying on the opinions of other is also not a substitute for actual evidence).
ilikeflowers
March 19, 2010
khandor,
Actually, having re-read what you posted I’m not sure why I posted a reply. Nothing you said is of relevance to whether or not Kobe is more productive than Drexler.
You are probably right about the poll. But it’s not relevant to whether or not Drexler was more productive than Kobe.
An ‘authentic assessment of his actual basketball talent’ isn’t relevant to productivity measurements when we’re evaluating veteran players. And of course there are talents which are difficult to see with the eyes whose effects may turn up when you start measuring actual results.
‘Analyzing basketball properly involves both Art & Science.’
This is definitely true. I doubt anyone here would take issue with that. If I’m constructing a team and two players at a given position achieve similar productivity in dissimilar ways then I’m going to choose the one that I think complements the skillset of my other players. Also, I’m choosing consistency over peak ability and I’m choosing team defensive acumen over individual defensive acumen although I’d like both.
khandor
March 19, 2010
ilikeflowers,
Although some might prefer to phrase the question in the following way,
“Was Clyde Drexler a more ‘productive’ basketball player than Kobe Bryant?”
The fact is …
The question is actually phrased like this, instead,
“Was Clyde Drexler a ‘better/superior‘ basketball player than Kobe Bryant?”
In which case, simply stating that according to a specific metric … which is based exclusively on stats-based measures … fails to answer that exact question in an adequate way.
——————
re: “Also, I’m choosing consistency over peak ability”
What I can tell you for a fact is that an elite level coach … across any of the major team sports … would never ever agree with your position on that specific subject.
An elite level coach … e.g. like Phil Jackson, Pat Riley, Bill Belichick, Bill Walsh, Joe Torre, Earl Weaver, Scotty Bowman, Mike Babcock, etc. … would always choose the player with ‘peak ability’ over the player with ‘consistency’ because what these coaches happen to know, both, instinctively and from their own first-hand experience is that:
i. Great coaching is indeed capable of taking the athlete with the ‘peak ability’ and, through proper training, allow this ‘peak ability’ to be put on display with increased regularity;
while,
ii. Even great coaching is not necessarily capable of taking the athlete with ‘consistency’ and, through proper training, allow a higher level of ‘peak ability’ to be put on display with regularity.
khandor
March 19, 2010
David,
Trust that I am someone who actually does understand what constitutes proper basketball analysis … both, when I provide it, myself, and when I see or hear it from others.
jbrett
March 19, 2010
khandor,
It is possible that this may be the last time anyone responds to your wildly subjective, data-free, analysis-free, logic- and reason-free proclamations about what ‘the people who know’ accept as fact, supposedly on your say-so. Your insistence on re-framing the question where facts are meaningless or, for your purposes, a hindrance, is nearly as unpleasant as your frequent claims to know the inner workings of every coach since Naismith. I don’t know if you’re geek enough to remember the DC Comics’ character The Phantom Stranger, but his catchphrase was, “Accept the word of one who knows!” Clearly, he was a pompous blowhard, as well.
This is, at least, the last time I will engage with you; it seems likely to me that it will not take long for most others to realize the futility of attempting useful debate with someone who is, in my conservative estimate, crazier than an outhouse rat. I don’t have any facts which would prove that, mind you, but if you were to poll the elite-level thinkers on this blog, I believe you would find a solid consensus. Seeing as how you invariably prefer opinion to fact, perhaps that poll would convince you to just shut up–and if you’re going to post ideas that boil down to ‘I’d rather have the guy who COULD play well than the one that DOES,’ shutting up would be the smartest thing you could do.
dberri
March 19, 2010
Khandor,
jbrett is a bit harsh but essentially on the mark. Speaking as someone who actually does analysis — and teaches others how this is done — I can assure you that you do not know how to do analysis. Proper analysis requires that you provide some objective evidence. One un-supported opinion after another is fairly pointless.
What I think is interesting is that you must think what you are doing is useful. Do you find when you discuss issues with people outside the confines of the Internet that your approach is persuasive? Again, unsupported opinions tend not to persuade most people who comment in this forum. Then again, maybe the people in this forum are not representative of the people you interact with.
ilikeflowers
March 19, 2010
khandor,
I think of the ‘best’ player as being the one who produces at the highest rate while being highly used. One can reasonably argue about the sweet-spot between usage and efficiency but the argument between Kobe and Drexler is really just one about whether or not Kobe is a superior enough defender or strategically more valuable when compared to Drexler to overcome his lack of relative production as measured. The box score is imperfect but so are the eyeballs.
I did choose a poor word in ‘ability’ since it can be used to mean some innate maximum (as you have interpreted it) as well as being a measure of what actually is demonstrated. I’m talking about peaks and consistency of ability as demonstrated. Specifically Kobe, MJ, and Drexler. Sorry about the ambiguous wording.
BTW, should we discount Kobe’s innate ability vs Drexler’s since Kobe has benefited from better coaching and yet produces less as measured?
ilikeflowers
March 19, 2010
I feel like I’ve entered some basketball stats version of Brokeback Mountain.
Pointless-Internet-Posts I wish I could quit you.
khandor
March 19, 2010
jbrett,
Here’s just one example of a problem which exists with what you just wrote.
Please re-read it and then explain what your definition is for the term, “play well.”
khandor
March 19, 2010
David,
Please note that in what you wrote initially you used the specific term “basketball analysis”; yet, in what you just wrote the second time around you used the generic term “analysis”.
There’s a basic problem which exists in some of what you happen to write on this otherwise fine blog concerning the game of basketball:
You seem to think that what constitutes legitimate “basketball analysis” must be rendered in the exact same style and structure as other forms of quantitative “analysis” used in scientific experimentation.
If you’re interested, take a crack at this instead:
Using your preferrred methodology, please explain which work of art is superior to the other?
A. Michelangelo’s “David”, or
B. Shakespeare’s “King Lear”
And, secondly,
Which artist was, in fact, superior to the other?
A. Michelangelo; or,
B. Shakespeare.
I won’t actually critique what you write, either way, so there’s really no need for you to worry. I’m just interested in reading what you might have to say, based on your preferred method of “analysis”, in general.
[In fact, you don’t even need to provide that answer in this forum. Simply send it to me, directly, via email, so that I can read for myself and not have it clutter up this thread.]
Just because Clyde Drexler amassed higher totals in “individual production numbers” over the course of his career, this does not necessarily mean that he qualifies as being a “better” basketball player than Kobe Bryant.
BTW, what I just put into bold isn’t just my opinion, unsupported by meaningful evidence … it’s a logical statement of fact.
Please forgive me for what might appear to be simple obstinance, on my part.
I can assure you, however, that it is not intended in that way.
I agree with many of your ideas about the game, itself, and the idiocy of certain GM’s, coaches and executives who conduct themselves as if they actually know very little about how the game really works … e.g. when they offer huge contracts to strictly marginal players like Andrea Bargnani, etc. … but, for me, or anyone else, who really does understand how the game of basketball actually works … and, how stats work, as well … it would simply be sacrilegious to allow some of what you sometimes have to say here about the game, and those who play/coach it, to go unchecked, as if it makes sense, from a practical basketball standpoint.
If I happen to incur your wrath for being willing to do that, on occasion, then, so be it.
Doing the right thing is worth the risk.
khandor
March 19, 2010
ilikeflowers,
re: “I think of the ‘best’ player as being the one who produces at the highest rate while being highly used.”
Hopefully, when you employ the term ‘highly used’ you are not simply referring to the nonsensical stat category commonly known as “usage”, which fantastically asserts that a “Turnover” should actually be credited to a player as a “positive” measure toward establishing his/her real level of “usage” in a game of basketball … while, for example, ignoring how many effective passes, cuts, screens, picks, close-outs, and proper defensive transitions, etc., a player completes in a game, when attempting to assess accurately their actual level of “usage”.
Conversely, I think of the ‘best’ player as the one who actually performs the many different skills which are involved with playing the game of basketball properly, at the highest level, while encompassing an array of different measures, including [but certainly not limited to]:
i. Simple game stats,
ii. Interpretive artistic merit,
iii. Interpretive degrees of difficulty,
iv. Team success/failure rate,
v. Functionality,
vi. Longevity,
vii. Consistency,
viii. Peak level of performance,
ix. Leadership, and
x. Teamwork.
ilikeflowers
March 19, 2010
khandor,
Usage as in playing a lot of games and a lot of minutes per game. Something that Kobe does well.
‘ii. Interpretive artistic merit’
Now I know you’re just trolling for blog hits.
"The Phantom Stranger"
March 19, 2010
David,
Accept the word of one who knows!
All of the best teachers of art and literature have informed me, either via mental telepathy or channeling (if deceased), that Michelangelo is undoubtedly the superior artist. Using only the first ten of my infinite number of criteria, my intuitive analysis clearly shows that, while Shakespeare may have produced more work overall, he loses out on degree of difficulty. At the end of the game, it will be clear that Michelangelo won, and moves on to face the winner of the DaVinci-Pollack showdown–oh, wait, you know what? They don’t have a scoreboard in the fine arts, so there’s no way to pick a winner; I’m now being told that basketball allows one to know at a glance which team produced the most points. I guess I’m just a pompous blowhard, after all. Nevertheless, Michelangelo is clearly better. Kneel before Zod! (Oh, wait, wrong blowhard.)
t4inc
March 20, 2010
Since Kobe won his 1st title in 2000, I’ve been telling my friends that I thought Kobe’s career numbers would end-up more like Drexler’s than Jordan’s.
Although I respect Kobe’s accomplishments, I just never bought into the “Kobe is the next MJ” thing. For one thing, Kobe was never the lock-down defender MJ was.
Also, and this is just a matter of preference, really, MJ was always a bit more graceful. The differences are subtle, really: Kobe’s drives are a bit too herky-jerky and end in a lot of fading-to-nowhere shots whereas MJ always seemed to get to the rim and finish with either hand.
Meanwhile, Kobe has to employ all these awkward clutches, fakes and Tony Parker-type floaters. Also, MJ’s jumper was smoother; Kobe’s follow-through is too stiff. Still too much wasted motion on Kobe’s part. Besides, I don’t think I ever saw MJ pull a floater; instead, MJ had this double-handed clutch where he stayed aloft long enough to shake the shot-blocker off and capped it off with a bank-shot.
MJ also employed a lot more pivots and curls on the low block, although Kobe has improved tremendously in that area in the last 2 seasons.
I’ve always told my friends that it seems that whereas MJ employed all these acrobatic moves to get higher-percentage shots, in Kobe’s case, he seemed to employ all these moves to look good; style points over substance. I’m pretty sure Shaq would agree with me.
Also, I’m pretty sure that Kobe enjoys his box score a little too much. MJ, on the other hand, still hates the 63-point game at the Boston Garden.
That said, while I love Lebron’s game, Kobe’s moves are still a lot prettier than his.
Anyway, going back to my original point, it turns out, based on Dr. Berri’s numbers, that I was wrong: Kobe wasn’t even as good as Clyde Drexler.
Wesley
March 20, 2010
My question on this topic may take it in a different direction, but I think it’s an important direction:
Is it fair to compare Drexler and Kobe?
Is it fair to compare Bird and Kobe?
Is it fair to compare Robertson and Kobe?
I think there is a statistically viable and analysis-based way to take a player from an older generation of the game and compare him to a player in today’s game. I think the WP48 stat is great for that.
But recently I’ve been arguing with sum buds that think superstars of yesteryear would be mediocre at best in today’s NBA. They argue that the larger pool of talent and competition is the reason why they couldn’t. I argue that if you took Oscar Robertson and raised him in today’s game (from youth rec leagues to the NBA) that his game would be as elevated as it was in his era and that he’d still be a superstar.
Am I wrong here? Is is irrational to think that older NBA stars – if given the chance to train and compete in today’s basketball climate – would be just as dominate?
Patrick
March 20, 2010
I don’t post on this blog very often, but I wanted to say that:
“ii. Interpretive artistic merit”
made my entire day.
I don’t know if this question has already been answered somewhere else, but I don’t understand the value of an assist. I mean, I understand that passing is valuable, but every other simple box score stat represents a tangible benefit (either points or extra possessions in all cases), yet assists do neither of those things. Can anyone here provide me an answer?
Patrick
March 20, 2010
*By “value” I guess I should clarify that as “statistical value”. Like, if you removed assists from the box score I think you would still get an accurate reflection of shooting efficiency and number of possessions, which seem like the meaningful things to me.
mrparker
March 20, 2010
Prof,
I was wondering if you could do a small study on Luke Harongody’s absence at Notre Dame towards the end of the season. Seems it would give or not give anecdotal evidence that ball hogs can be detrimental to a team even when they are productive.
jbrett
March 20, 2010
Patrick,
There is a section in The Wages of Wins explaining the value assigned to assists–or more accurately that there is a reason to assign a value. I wish I could remember more precisely, or had a better handle on the analysis; maybe someone can offer a more helpful post.
khandor
March 22, 2010
ilikeflowers, David, “Phantom Stranger” and others,
The facts are these:
1. No trolling for blog hits by me. Unlike others, perhaps, I don’t need blog hits to justify the work I do on-line.
2. The reason there will be no response coming from David, or anyone else for that matter, to the specific question which I asked above … except, possibly, in jest [e.g. see the “Phantom Stranger”] … is due to the illogical nature of someone even attempting to provide a coherent answer as to which of those two “artists” is actually “better/superior” to the other predicated upon a quantitative stats-based metric of some sort.
Yet, when it comes to trying to understand how the game of basketball actually works … and, therefore, properly evaluating the respective “artists” who engage in this field of endeavour … some in this specific environment are under the mistaken impression that they can accurately make this distinction/determination predicated upon just such a metric.
[… and, what’s even more funny is that these same individuals would dare suggest that someone else who would oppose the use of this essentially illogical method of evaluation in this specific sphere of life should somehow be characterized as being the “nonsensical” one in this type of discussion about the subject.
Hint: Just try to follow this type of [il]logical rationale for a moment.
i. Basketball is fundamentally a TEAM GAME for which a cumulative “score” is kept for each of the 2 opponents in a specific contest.
ii. Team Wins and Losses, Number of Championships Won, etc., are recorded for each player, coach and executive.
iii. Yet, Team Wins and Losses, Number of Championships Won, etc., should not be included in the proper evaluation of these individual players, coaches and executives.
iv. Only “individual stats-based production numbers” based on Points Scored, Field Goal Makes-Attempts, 3PT Field Goal Makes-Attempts, Free Throw Makes-Attempts, Offensive Rebounds, Defensive Rebounds, Total Rebounds, Assists, Steals, Blocked Shots and Personal Fouls should be included in a proper evaluation of the individual players, coaches and executives … in what is fundamentally a TEAM GAME.
v. As such, other individual skills which are also associated with playing the game properly/well, on an individual basis, e.g. cuts, slides, screens, picks, non-assist passes, dribbles, dribble attacks, offensive footwork, defensive footwork, deflections, rotations, helps, close-outs, box-outs, defensive transitions, offensive transitions, positioning, penetrations created, penetrations allowed, etc., are not to be used as sound/important means for evaluating the individual work/ability of players, coaches and executives … since they are, either: A. Not currently tracked at all by game statisticians; or, B. Currently tracked poorly by game statisticians, who do not have the proper training required to make these types of skill-based assessments accurately, in the first place.
If you were asked to evaluate the sensibleness of that specific step-by-step process of thought, applied to a different sphere of life, or “science”, for that matter … Would you really say that it is sound?
My best guess is that … No, you would not.]
3. Based on the available data …
Although it is perfectly reasonable to observe that Michael Jordan had a “higher” Points Scored Per Game Average than Clyde Drexler and, in turn, that Clyde Drexler had a “higher” Points Scored Per Game Average than Kobe Bryant, etc. … using Points Scored Per Game as just one example … it is not perfectly reasonable to assert that Michael Jordan was, therefore, a “better/superior” basketball player than Clyde Drexler and, in turn, that Clyde Drexler was, therefore, a “better/superior” player than Kobe Bryant.
To engage in a process which looks like that is, indeed, faulty science … which confuses:
I. Partial quantitative research [i.e. concerned with “most”, “average”, “least”, “standard deviations”, etc.];
and,
II. Comprehensive qualitative assessment [i.e. concerned with “best”, “better”, “worse”, “worst”, etc.].
brian
March 22, 2010
khandor, I agree that the current WP48 system is imperfect, and that it neglects a lot of things like screens, tapped rebounds, etc. It only attempts to approximately predict the impact of certain actions (rebounds, steals) on team wins, and I have no doubt that the picture will somewhat change if we were to account for these other aspects of the game.
But when doing science, people make approximations all the time–sometimes it’s because we don’t have as much data as we would like to have (actual detailed numbers for tapped rebounds and drawn charges, for example), sometimes it’s because we’re lazy, and sometimes it’s because research is just plain hard. You can’t just claim that because a model has a few approximations that it automatically sucks, as you’re doing here–if you want to say that the model’s approximations disqualify it from accurately predicting player contributions to wins, you must prove it with some empirical evidence, or a better model based on empirical evidence. Again, this is an imperfect model, but it’s been tested time and time again against the empirical evidence (team wins), and we see from there that the predicted win totals from the model come pretty close to the actual number of wins accrued by a team.
In response to iv, the “individual stats-based production numbers” are all accumulated in the presence of 5-on-5 basketball, as I’m sure you realize. So I don’t see how they’re irrelevant to predicting who will and who will not win in a game of 5-on-5.
Finally, as I mentioned, people would love it if the data you mention in v were recorded and accessible, and I’m sure that the story told by WP48 might change a bit (for the better) if that stuff were incorporated into the model. However, the model already predicts win totals reasonably well as it is. Furthermore, dberri showed in a few of his journal articles that a much simplified version of the model neglecting a good number of the box score stats still told mostly the same story with regards to the allocation of wins produced. So for these reasons, it’s unlikely that the model predictions would change appreciably upon the addition of the aspects of the game you mention in v.
Not sure if it’s worth continuing this conversation with you, but if you were to actually read up on the model and learn about regression analysis, you’d probably be receiving fewer flames. You continually defend your opinions with other opinions, without ever really citing empirical evidence.
brgulker
March 22, 2010
khandor,
Strawman. No one is making the claim that ONLY statistical production should be included in player and/or team evaluation. Dr. Berri makes the point in his book and repetaedly on this blog. You are arguing with yourself on this point.
brgulker
March 22, 2010
Man, serious html fail there. My bad, everyone.
Christian
March 22, 2010
Kobe’s career isn’t over, so your comparison to Drexler or Jordan is premature. I have some questions for anyone that throws rocks at Kobe: how many guys have hit 6+ game winners in a season with half a hand? How many guys have played with more pain and excelled while doing so? How many guys have scored 81? How many guys have 4 rings by the age of 31? How many guys played their hearts out while trying to litigate a bogus rape charge? How many guys have wanted to win more? These are questions that go beyond FG% and Points/48 Minutes.
Sure, you can say all you want. But when his career is over, I predict he will be known as the GOAT. He just needs to outscore Kareem and win 7 titles. I know it’s a tall order but that’s the only thing that will shut you boys up.
One thing I’ll add, and I find this very interesting as a Lakers fan, is that Kobe’s legacy is in jeopardy this year if the Lakers lose to the Cavs in the Finals. However, if he and the Lakers win the title against LeBron and the Cavs, he will have beaten one of the best players in history. If LeBron wins, the same can be said for him. Kobe has more to lose because he has less time on his side.
brgulker
March 22, 2010
Wow, Christian. That was very original. I’ve never heard any of those arguments before!
Christian
March 22, 2010
Wow brgulker, sarcasm, how very original and thought provoking! You’re like the best thing since sliced bread.
brgulker
March 22, 2010
Christian,
Did you even read the comments? Every single argument you offered in favor of Kobe has been addressed directly, if not debunked completely — especially the point total argument and championship ring total argument.
The only people who will think that Kobe is the GOAT are people such as yourselves, Lakers fans.
khandor
March 22, 2010
brian,
The fact is …
1. At no time have I ever said that David’s metric “sucks”.
2. What I’ve said at different times is that …
i. Any statistical-based metric is incapable of generating a proper answer to a question which concerns a qualitative appraisal like, “Who is the best Off Guard of All-Time?”
ii. What a statistical-based metric is capable of generating, however, is a proper answer to a question which concerns a quantitative appraisal like, “Which Off Guard amassed the highest total of ‘individual production numbers’ of All-Time, or compared to another player, according to that specific metric.”
IMO, there should be no doubt in anyone’s mind that empirical data such as this is a useful tool when attempting to determine a player’s actual worth/quality with a high degree of accuracy.
3. The problem comes about, however, when “production based stats” are used to make observations and draw final conclusions about the relative quality of individual players, coaches, teams and executives, etc., in a broad sense [i.e. Player X is “the best” of all-time] for which they were not designed in the first place.
——————
brgulker,
If the goal is simply to win games and championships, as defined by the final score of those games and championship series, etc., then …
Please explain the reason The Great Bill Russell is not unanimously recognized as the “best” basketball player of all time? … since THE FACT is:
1. The teams he played on have the highest regular season win percentage, and;
2. Won the most NBA Titles;
in league history.
[i.e. when comparing the ability of one Centrifugal Force player with another]
——————
brian and brgulker,
1. I think David is perfectly capable of talking for himself on his own blog and doesn’t need anyone else to do this for him, in an effort to respond to me.
2. Just because I might be willing to criticize David’s use of a certain bath-tub does not mean that I think it needs to be thrown out with the bath-water.
mrparker
March 22, 2010
i just find it interesting that the rest of the unquestioned all time greats measure up to that status according to wins produced. Only Kobe does not. Used to love the guy before the anointed status. Since then, I find it hard to stomach all the praise even though he deserves praise its not to the level that he receives it.
brgulker
March 22, 2010
khandor, the point that you seem to be missing — one that is made clearly in the book — is that the “statistical production” of individual players is directly linked to team wins.
Once one has understood this connection, there is no dichotomy between your points i and ii.
Purely IMO: I’m not much for making cross-generational comparisons of individual players, for a variety of reasons. It’s easier to make the Kobe vs. MJ comparison, because they’re only one generation apart. So, IMO, the “Greatest of all time” debate is a non starter for me.
Regarding Bill Russell’s greatness, I doubt you’d find a regular commenter here that would argue against Russell as being an all-time great. Unfortunately, we don’t have the data available to us that allows us to compare his statistical production with contemporary players.
Um, what?
You challenged anyone to answer the question you initially asked of Dr. Berri. So I did.
Your words exactly:
Christian
March 22, 2010
Brgulker: You didn’t address anything I wrote, and then questioned whether I read through the comments. Kettle, meet black.
I haven’t seen anyone address Kobe’s play with injuries, his criminal defense, game winners this season, the 81 point game, his mental and physical toughness, his style, etc.
My point is, the stats don’t tell the whole story. They are important, but there’s a lot more to be considered. So, to make myself clear, I’m not saying your reasoning is invalid, I think it’s incomplete. If you’re going to compare the players, a holistic approach is best.
And IF Kobe wins 7 rings and scores 40k points, those stats will speak louder than anything else you can compare him to.
brgulker
March 22, 2010
Christian,
Are you serious?
Re: injuries. Everyone plays with injuries at some point in their career. MJ did it. So did Clyde. So what? How does that affect who’s better and/or more productive than the other?
Re: “criminal defense.” You didn’t see the in-depth conversation about defense in this very comment thread?
Re: the 81 point game, point totals have been discussed at great length in this comment thread (and time and again on this blog generally). Regarding the specific 81 point game, who cares? What does it establish? It’s one game. It’s impressive, sure, but what does it have to do with his relative status to Clyde and MJ? What weight does one game carry when compared to entire careers?
Regarding his “toughness,” well, I don’t know what can be said. Those aren’t quantifiable categories, so I haven’t the first idea how one would compare Kobe’s “toughness” to MJ’s or Clyde’s. I don’t know if you ever watched MJ play or not. I did, quite a bit. I certainly saw “toughness” on display time and again. But I have no idea how to compare that to Kobe.
The championship ring argument, again, was discussed at length above. Best I can tell, you didn’t offer any substantive responses when it was debunked above. If rings make a player greater than another, then Robert Horry is greater than Kobe Bryant.
As to point totals, well, that’s a dead horse.
brgulker
March 22, 2010
One last thing, why is it that the 40k state — that stat in particular — that will establish Kobe as the best player ever in your eyes? To me, that’s a very telling statement about how you understand the game.
Christian
March 22, 2010
Everyone plays with injuries? Yes, you’re right, everyone gets tweaked. Who shoots like Kobe with a mangled mit though? There’s no one tougher in the NBA than Kobe. He plays with more pain than anyone, so much so that his teammates are inspired by it. That’s leadership.
How is 81 irrelevant? It’s one of the finest displays of scoring, EVER! It’s a part of his legacy. Aside from that, MJ and Clyde never accomplished it.
I didn’t see anything written about Kobe’s rape trial, but I think it should be addressed, because how many players can concentrate with that stuff going on? Stu Lantz always says that Kobe is the rare player that plays well when he’s angry. i agree.
Kareem scored 38k+, so if Kobe scores more than Kareem, he will be the all-time greatest scorer, hence the 40k benchmark. It’s a nice round number.
brian
March 22, 2010
Christian, in the grand scheme of things, a single (albeit very impressive) 81 point performance means nothing. If Kobe could pull that off every night while shooting a high percentage, then sure, he would and should be considered an all-time great, by virtue of the fact that high efficiency scoring tends to directly impact wins.
But if we wish to predict how much a given player is expected to positively impact his team on any given night, we must take all of his performances into account, good and bad. Kobe has had some fantastic games and some dreadful games throughout his career, both in the regular season and in the playoffs. The 81 point game is just noise, in the same sense that his rare 5-point, 25% eFG is. Why should MJ and Drexler lose credit for not lucking out in this way? You’re essentially arguing that a strong aspect of Kobe’s legacy is a single set of 48 minutes in which he was spectacular, but throughout the other thousands of minutes in his career, he paled in comparison to Jordan and Drexler for reasons stated already.
And hey, I’m a Lakers fan too. I’d love it if Kobe raised his game to Jordanesque levels, but the empirical evidence is clear–Kobe is less accurate from the floor, an inferior rebounder, and just flat out inferior to Jordan (but still very good).
Ethan
March 22, 2010
Kobe’s 81 point game is irrelevant to determining his greatness vis a vis MJ or Clyde.
To wit, when have you ever heard MJ is the GOAT because he scored 69 points in a single game or because he scored 32, 292 career points?
Bear in mind, other players have ALREADY scored more than Jordan over the course of their careers. Other players have ALREADY won more than Jordan over the course of their careers.
Yet, nobody has put together the sublime combination of insanely good stats, otherwordly accomplishments, mvps, finals mvps, records, rings, and impact on the game.
Kobe can never be GOAT because he simply isn’t good enough. He can never have a 6 for 6 record in the finals. He can perhaps one day have 70 nba records, like Mike, but this is highly doubtful. He is simply less efficient, and takes far longer to accomplish equal tasks. This is already reflected in the metrics, and there is nothing he can do about it.
khandor
March 22, 2010
brgulker,
1. If “individual production numbers” are indeed directly linked to predicting future team win totals then please go right ahead and state the existing “correlation” which is identified in the Wages of Wins, the Wages of Wins Journal, WP48, or Stumbling on Wins, etc..
As best as I can see, this “correlation” is not listed anywhere on this blog.
2. Although you might think that you did, in fact, you did not answer the question which I asked.
3. I invited others to provide their own answer to the question which I asked, not provide what they happen to think David’s answer might entail, according to the metric that he’s developed.
4. I did not say that Bill Russell’s status as an all-time great is in question by any regular contributor here. What I said is …
——————
Please explain the reason The Great Bill Russell is not unanimously recognized as the “best” basketball player of all time? … since THE FACT is:
1. The teams he played on have the highest regular season win percentage, and;
2. Won the most NBA Titles;
in league history.
[i.e. when comparing the ability of one Centrifugal Force player with another]
——————
which identifies Bill Russell as THE Single Greatest basketball player of All-Time, according to any stats-based metric associated with the INDISPUTABLE data concerning Wins, Losses and Championships.
5. As I’ve said several times before, there is actually a great deal that I agree with David about, when it comes to understanding how the game of basketball actually works and which individual players are, in fact, more “productive” … according to their game stats … than others.
Phil
March 23, 2010
Christian,
My quantitative, in-no-way subjective analysis has determined that Carl Landry is the toughest player in the NBA.
brgulker
March 23, 2010
It is, actually, but it’s spelled out in great detail in the book. You should read it. You would enjoy it (or at least I would think you would given what I know of you).
If I find Berri’s metrics convincing, then my answer will obviously include some of Berri’s metric. That’s fair, right?
Alright, I’ll try this again. I think cross-generation comparisons are fun to think about but are ultimately very problematic for a variety of reasons. That’s personal opinion.
But, IMO, the definitive answer to your question is this, I think: people who frequent this site tend to argue from evidence to conclusion. In other words, we would be reluctant to make claims about the “greatest player ever” without having studied all the available evidence first.
Unfortunately, the NBA didn’t start tracking the type of statistics that Win Score/Wins Produced/WP48/PAWS48 require until 1977. As a result, it’s not possible to rate Bill Russell by any of these metrics. Thus, it’s not possible to compare players before that date with players after that date by these metrics (all of this is in the book and all over the blog).
So on this particular blog, you won’t find people clamoring about Bill Russell because we can’t – in the literal sense of the word, we cannot – argue from a position of strong evidence to a well-supported conclusion.
Brief personal aside: I never saw Bill Russell play, so the only knowledge I have about him is 2nd hand, which would make any personal opinion shaky, at best, because I don’t even have any subjective knowledge about him.
So was he the best player ever? Maybe he was. There seems to be good reason to argue that he was, and I don’t think anyone here would disagree with that, based on what we do know about Bill Russell. But definitively, beyond all shadow of a doubt? I don’t think a lot of WoW readers would make that argument with a high level of certainty.
khandor
March 23, 2010
brgulker,
1. Where is the actual “correlation” number listed on this site? [e.g. +0.857]
IMO, it does little good to make constant reference to a supposed level of “correlation” without identifying clearly what that number actually is.
2. re: “You would enjoy it (or at least I would think you would given what I know of you).”
No doubt.
Given the immediacy available through this blog, however …
[by way of analogy]
Why on earth would anyone choose to read about the life and teachings of Jesus Christ, Mohammed, Buddha, Confuscious, etc., when the opportunity exists to speak directly with him on-line? :-)
Especially, if one already have a solid connection with him and likes/agrees with a good portion of his methodology, but for a few distinct areas where it seems as though he has gone slightly awry and fallen prey to using terminology that is not properly associated with someone in his field of specialty, in the first place … e.g. confusing the terms [i] “the best” X with [ii] “the most productive” X, according to specific metric Y.
3. No, not fair, actually.
Unless you are David, himself, you are compelled to provide your own perspective on the matter, in question … if, at least, what you are trying to do is sincerely answer the question which I asked originally.
As far as I can see, David is here and is perfectly capable of speaking for himself, if he deems it to be appropriate.
When you speak on-line, it should be to represent what you happen to think.
Once you do that, then, you are certainly free to use whatever evidence you deem appropriate to illustrate the correctness of your perspective, if you feel it necessary to do so.
[If you choose not, however, that’s okay, too … because this specific environment happens not to be a PhD thesis/dissertation that you are called to defend in a rigorous academic-only manner. Afterall, if it was, in fact, just that and nothing else, then David, himself, would not be free to make many of the sweeping observations/generalizations which he makes on occasion when he writes what he writes on this blog, would he?]
It would be a different situation entirely if David wasn’t actually here to speak for himself.
4. If you are not someone who thinks that cross-generalisations are appropriate … then, unfortunately for you [and me], you should not be the one who is attempting to provide an answer to the specific question[s] which I asked, in reference to the methodology [i.e. quantitative vs qualitative] involved with making that type of comparison, in the first place … i.e. Jordan to Bryant, Jordan to X, Drexler to Bryant, Bryant to Y, Russell to Jordan, Russell to Z, Artist I to Artist II, etc.
As well, please keep in mind that:
– I am not the one who wants to find a definitive … i.e. quantitative … answer to the specific question of, “Who was/is the ‘better’ off guard, Clyde Drexler or Kobe Bryant?”
– I have already acknowledged, above, that Drexler is, in fact, the one who has amassed the “higher” individual production numbers … at least, to this point … over the course of his career
5. If you are someone who asserts the following:
i. Rebounds, in general, are under-valued by those who do not really understand how the game of basketball actually works;
ii. A myriad of specific defensive skills … many of which to this point are observable but still largely unrecorded by game statisticians … in general, are under-valued by those who do not really understand how the game of basketball actually works;
iii. A myriad of specific offensive skills … many of which to this point are observable but still largely unrecorded by game statisticians … in general, are under-valued – except, of course, for Points Scored – by those who do not really understand how the game of basketball actually works;
iv. Which team actually wins a specific basketball game/series/championship is VERY important … and should actually form the basis for making decisions concerning the relative worth of individual players;
v. A strong positive correlation exists between i., ii., iii. and iv.;
then …
it should also be the case that, based upon examination of the available data/evidence … Bill Russell is recognized as both:
I. “the most productive”;
and,
II. “the best”;
basketball player of all-time, since he amassed solid-to-terrific individual production numbers and HIS teams actually won more games/series/championships than anybody else’s in the history of the game.
[PLEASE NOTE: I am only using the case of Bill Russell to illustrate the point that actual WINS AMASSED should be a focal point for any legitimate discussion when comparing the relative worth of Player I to Player II, etc., in a team sport environment.]
brgulker
March 23, 2010
That’s ridiculous. You just said that if I have an opinion that’s partially informed by the work of Dr. Berri, I’m not allowed to offer my opinion, because it’s somehow less mine or original or something.
Christian
March 23, 2010
Phil,
Carl is the 2nd toughest. :)
khandor
March 24, 2010
brgulker,
Your specific question to me wasn’t whether, or not, you should be “allowed” to offer your opinion at all on the topic I outlined here but if it would be considered “fair” for you to do so, with reference to what you think David’s answer would be, based on the metric which he’s developed, in this particular environment.
IMO …
While you certainly have the right to do that, it’s a mistake in judgment to think it “fair” to do so.
What would be “fair” is that:
i. You let David answer questions which are directed at him, exclusively;
ii. You answer questions which are directed at David and others without making specific reference to what you may think of David’s work, while providing your own thoughts about the topic;
iii. You answer questions which are directed at you, exclusively, by providing your own thoughts about the topic + whatever evidence you feel is needed to support/illustrate your perspective.
brian
March 24, 2010
khandor, if Dr. Berri actually wanted to reply to you, I’m sure he would.
re ii., I’m not sure what your problem is. It’s like saying we can’t believe in Newton’s laws just because we didn’t come up with them.
khandor
March 24, 2010
brian,
1. re: “… if Dr. Berri actually wanted to reply to you, I’m sure he would.”
I agree, completely.
From my end, there’s no problem, whatsoever, if David chooses not to reply to any question which I ask.
It is a slightly different matter, however, if/when David chooses not to reply and, then, others think it “fair” that they can step in for him to provide what they think he might have to say about the original question, or provide their interpretation of HIS metric.
2. Hopefully you can see the difference between:
A. David being present in this environment and being able to speak for himself, as the living owner/author of this blog;
vs.
B. Sir Isaac Newton being dead for quite some time now.
[Aside: If Isaac was alive today, and the author of a lively and informative blog … Do you really think it would be the best use of my time to get the impressions of others concerning how HE might choose to answer my questions about the Laws which HE derived? … as opposed to just having HIM reply directly to me, or not.]