In response to the last post, reader Matt Dalton asked for more information on Michael Ray Richardson.
Here are Richardson’s career numbers (WP48 = Wins Produced per 48 minutes).
1978-79: 2.0 Wins Produced, 0.078 WP48
1979-80: 21.4 Wins Produced, 0.336 WP48
1980-81: 19.7 Wins Produced, 0.298 WP48
1981-82: 14.0 Wins Produced, 0.221 WP48
1982-83: 6.5 Wins Produced, 0.151 WP48
1983-84: 3.2 Wins Produced, 0.121 WP48
1984-85: 14.6 Wins Produced, 0.225 WP48
1985-86: 7.0 Wins Produced, 0.210 WP48
Totals: 88.6 Wins Produced, 0.229 WP48
Richardson was suspended (for taking drugs) by the league in 1986, so his career ended when he was 30 years old. Prior to that suspension, though, Richardson was one of the better guards in the NBA. His teammates, though, were generally lousy. For his career his teammates posted a 0.076 WP48 [average is 0.100]. To put that in perspective, Kobe Bryant’s teammates have posted 0.113 WP48; and in only one season were Bryant’s teammates as bad as the average we see for Richardson.
The perceptions of a player’s value are influenced by scoring and winning. Richardson averaged only 14.8 points per game in his career and he generally played for losers. So people probably would not think he was as effective as Kobe. But the data – at least the Wins Produced data – suggests otherwise.
Update: After the 2008-09 season, Kobe’s Wins Produced stood at 149.0. His WP48 was 0.207.
- 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.
What are Kobe’s WP numbers compared to this?
Thanks for the post, prof. Sugar was one of my favorites; what Dick Vitale used to call a Renaissance Man. He could pass, score defend and rebound. Keep it up!
Posey,
I updated the post with Kobe’s career numbers.
Dr. Berri,
Richardson’s number seem to fluctuate a lot, going from a bona-fide superstar .336 WP48 with 21 WP to consecuative seasons with .151 and .121 WP48s, and most of hsi seasons falling in between. Any idea why?
Man of Steele,
I’m guessing it was the coke.
That was definitely my first thought as well.
Here’s a nice article on Michael Ray and Chauncey Billups – the best players to come out of Denver: http://www.rockymountainnews.com/news/2009/feb/12/billups-richardson-connected-in-denver-all-star/
How about a piece about another player that didn’t get enough ink, Sidney Moncrief?
here in italy, we were lucky enough to watch richardson play. he was one of the best player in 90s
he also played in bologna, in the same team in wich ginobili palyed ten years later.
Richardson made up for his horrid shooting numbers in other stats, particularly assists and steals. Bryant has much better scoring and shooting but doesn’t do as well in other measurables.
We also have to remember that Richardson and Bryant played in different eras and against different defenses as a whole. For example, Richardson played when zone defenses were banned by the league, while Bryant’s numbers in a significant chunk of his career came when zones were allowed. If Bryant got to play against man defenses all the time, his numbers would inevitably spike, and vice versa with MRR.
Peter,
I buy the zone defense argument but you have to factor in the hand check rules and the lack of physical play. Back in the day you needed to get mugged on a basketball court to get a foul call . Also WP48 is a relative and weighed statistic which wouldn’t be affected (if scoring is harder/easier for everybody it will weigh out because of wins)
Regarding zones vs. Man defenses,
I have concrete evidence that college numbers correlate directly to pro career. How come these numbers seem to stay roughly the same when players leave college? Aren’t harsher zones allowed in college than in the pros. IMO, effective players have a mindset that allows them to adapt their games regardless of situation. Basketball players aren’t anything like electronic football’s plastic pieces who are set to do a certain task by their coaches. Great basketball players have tremendous skill sets. They can and do adapt.
Arturo: The analysis has shown that, roughly speaking, the statistical contribution needed to produce a win has remained about constant (the key statistic is the league-averaged points per possession). This says nothing about how hard it is to produce a win. In other words, a player who produced 0.200 wins per 48 minutes in 1980 was twice as good as the average player in 1980, but is not necessarily twice as good as a the average player today.
I think the level of play in the NBA has improved with time. At the very least the players are more athletic, but the most telling change is the rise in field-goal percentage despite the improvements in defense.
Attributing Richardson’s inconsistency to drugs doesn’t make sense either. After a decent rookie year (for a rookie) he was legitimate superstar for two years, posting WP48s of .336 and .298. After that he had one season at .221, close to his career average, before declining to .151 and .121. After his decline, however, Richardson improved .225 and .210 in consecutive seasons before being suspended for drugs. It would seem that his performace tapered off very quickly during the initial arch of his career and then rose again before he was suspended for drugs, indicating the drugs may not have been hindering him.
Of course, it makes sense logically that the drugs were affecting him, because how in the world could you play basketball with all that stuff in your system? However, the evidence doesn’t really seem to bear out that conclusion. I guess what I’d be most interested to know is if there are other players whose numbers fluctuate this much. Perhaps a post about the 20 most inconsistent performers of all time (post ’77) would be interesting.
Lior
Isn’t the league’s average efficiency roughly the same as it was in the mid 80s through early 90s?
A lot of players in the late ’70s and early ’80s were doing drugs. I wonder how their numbers look.
Micheal Ray Richardson is also famous for this exchange:
REPORTER: What do you think is happening to the team?
RICHARDSON: The ship be sinking.
REPORTER: How far can it sink?
RICHARDSON: Sky’s the limit.
The league average efficiency should be identical each season, as the win statistics are based on the league’s teams having a cumulative record of .500, which the NBA has succeeded in maintaining every year of its existence.
If a team was stocked with players that had a .100WP48, that team would be expected to go .500. A steal may be worth more in 1980 than it is today (or vice versa) by a marginal amount, but a .200+ WP48 carries the same meaning.
Tball: the “efficiency” we are discussing is the number of points scored per possession, which can of course vary from season to season, but in fact does not vary much (it’s about 1 point per possession, give or take 10%). The relation between efficiency and wins is also roughly the same over the years. It follows that the marginal effect of a point scored, rebound, steal etc on wins has roughly remained constant.
Yes, the precise efficient of the league varies with time as the rules change (in the last 30 years the trend has been of gradual shift in favour of the defence, offset by occasional rule-changes helping the offence.
However, if players are getting better on both sides of the floor this wouldn’t affect the efficiency. If players have both better shooting skills and better defensive skills the statistical effect would be a wash.
What this says is that a WP48 of 0.200 always means the same thing: a player who is about twice as good as the average player of his time. Indeed, a WP48 always carries the same meaning — and that meaning is relative to the season in which it is calculated. Take a player who produces 0.100 in today’s NBA. Would he produce 0.100 if playing in the NBA of the early 80s? Of the 60s?
I think it’s quite likely that it is harder to produce a point, steal or rebound today than it was in 1980. This should be expected: the talent pool from which players drawn is larger, training methods and diets have improved, players are professionals from younger ages, etc.
What the data we have shows, is that in ’79-’80 Richardson was more than three times better than the average player of ’79-’80. Kobe has never been three times better than the average player of his time. In other words, Richardson was much more exceptional than Kobe.
All I’m saying is that we can’t tell if he was better on an absolute scale.
Lior, you said:
“I think it’s quite likely that it is harder to produce a point, steal or rebound today than it was in 1980. This should be expected: the talent pool from which players drawn is larger, training methods and diets have improved, players are professionals from younger ages, etc.”
in my opinion it may be easier today, because there are more teams, and the talent is “diluited”.
I wonder if this could be measured somehow
The talent is diluted? Maybe when expansion first happened. But right now, the NBA is athletic from top to bottom as it has ever been, and the defenses are more sophisticated. Now the game is also much much less physical than it used to be, and that physicality (hand checks, bumping of the cutter, grabbing, in addition to the incredibly hard “playoff fouls” that used to happen, even when guys weren’t in the playoffs) allows one to negate athletic differences.
I would say on the whole that scoring and rebounding are probably as difficult or easy as they ever were. Just for different reasons. Player length and athleticism, the change in physicality, and the change in rules all factor in. Though if teams fully utilized the zone defense rules along with the shot clock (switching to zone defenses when the offense is reset late in the shot clock, hard double teaming when the shot clock goes to 2 or 3 seconds, that sort of thing), you would probably see a temporary decrease in efficiency, before teams adapted and all started shooting earlier in the shot clock to get more efficient shots.
Hi Dave,
Thanks so much for the post. I have a feeling that the Knicks’ recent management would have traded Sugar Ray away because he didn’t score enough. But he was great player – apparently the only guy Isiah Thomas was scared of!
As a long-suffering Knicks fan, I keep praying that their executives are reading your blog – otherwise I fear Jamal Crawford is going to show up again on the roster. Is there any way you could pitch your services to Donnie Walsh.
Michael Ray Richardson played during an era when players were less selfish; therefore, his scoring would be less than NBA stars of today. In my opinion, Kobe is an extremely talented player; however, he is a ball hog (selfish) meaning he rather shoot the ball than pass it to an open teammate. Michael Ray would defintely pass the ball to an open player versus shoot a low percentage shot while he is guarded. Fans have the tendency to look at the points scored without looking at the number of shots taken during a game. Kobe takes most of his team shots during a game whether he is guarded or not. Michael Ray and Magic Johnson played the same type of game and I think Magic stated that Micahel Ray was the player that he feared the most. I do agree that Michael Ray played on less talented teams than Kobe; therefore, he could have easily been selfish and taken as many shots as Kobe; however, he was a team player. It was trully unfortunate for the NBA and Michael Ray that his career ended early due to drugs.