Dennis Johnson in the Hall of Fame?

ESPN.com is reporting that Dennis Johnson has been elected to the Hall of Fame.   This selection — as discussed below — seems to be somewhat questionable.

Let’s begin with a model of Hall of Fame voting.  Justin Kubatko – at Basketball-Reference.com – reports a model of Hall of Fame voting that incorporates the following factors: height, points per game, rebounds per game, assists per game, All-Star game selections, MVP award shares, NBA championships won, and whether or not the player’s career ended before 1960.  The model ignores per-minute performance, shooting efficiency, turnovers, steals, blocked shots, and personal fouls.  This is not surprising since most player evaluations in the NBA focus on the per- game factors Kubatko included (i.e. don’t value per-minute performance highly) and ignore such issues as shooting efficiency and turnovers.  In sum, I think Kubatko’s model captures much of what goes in to the voting for the Hall of Fame.

That being said, according to this model, the probability that Johnson would be elected was only 0.498.  So Johnson – according to the limited list of factors generally noted in evaluations of NBA players — wasn’t a particularly strong candidate. 

If we shift our focus to Wins Produced (which I am sure would not be the best predictor of Hall of Fame voting) we see that Johnson did have some good seasons.  But he would not rank among the most productive players who ever played in the NBA.

  • Seattle [1977-78]: 7.0 Wins Produced, 0.153 WP48
  • Seattle [1978-79]: 8.4 Wins Produced, 0.149 WP48
  • Seattle [1979-80]: 6.9 Wins Produced, 0.112 WP48
  • Phoenix [1980-81]: 7.7 Wins Produced, 0.141 WP48
  • Phoenix [1981-82]: 9.8 Wins Produced, 0.161 WP48
  • Phoenix [1982-83]: 7.6 Wins Produced, 0.143 WP48
  • Boston [1983-84]: 4.7 Wins Produced, 0.085 WP48
  • Boston [1984-85]: 3.3 Wins Produced, 0.053 WP48
  • Boston [1985-86]: 2.0 Wins Produced, 0.036 WP48
  • Boston [1986-87]: 1.8 Wins Produced, 0.029 WP48
  • Boston [1987-88]: 2.0 Wins Produced, 0.036 WP48
  • Boston [1988-89]: 0.6 Wins Produced, 0.013 WP48
  • Boston [1989-90]: 2.2 Wins Produced, 0.052 WP48
  • Career [1977-1990]: 64.1 Wins Produced, 0.090 WP48

The data needed to calculate Wins Produced only goes back to 1977-78.  So we are missing what Johnson did his rookie season. We should note, though, that Johnson’s rookie season was the only time he did not play on a winning team. 

Johnson’s ability to find winning teammates was never more true than during his time in Boston.  An average player posts a WP48 of 0.100.  As one can see, Johnson was below average as a Celtic.  One suspects, though, that seeing Johnson celebrate at title as a Celtic in  1984 and 1986 led people to believe that he was an above average player.  Certainly Johnson was above average in 1979 when he won a title with Seattle.  But Johnson didn’t join the Celtics until he was 29 years of age.  And at that point, he was past his prime (and not an above average player).

One probably does not need Wins Produced to know that Johnson was not one of the all-time great NBA players.  And one also suspects that Johnson’s “skill” at finding great teammates drove this selection.  To illustrate, consider the top three producers of wins on the teams that Johnson won a title with during his career. 

The 1978-79 Seattle SuperSonics

  • Jack Sikma: 10.6 Wins Produced, 0.173 WP48
  • Gus Williams: 9.8 Wins Produced, 0.208 WP48
  • Dennis Johnson: 8.4 Wins Produced, 0.149 WP48

The 1983-84 Boston Celtics

  • Larry Bird: 24.2 Wins Produced, 0.383 WP48
  • Robert Parish: 13.1 Wins Produced, 0.220 WP48
  • Kevin McHale: 9.9 Wins Produced, 0.184 WP48
  • Dennis Johnson ranked fourth on this team in Wins Produced

The 1985-86 Boston Celtics

  • Larry Bird: 26.2 Wins Produced, 0.404 WP48
  • Robert Parish: 11.7 Wins Produced, 0.219 WP48
  • Kevin McHale: 11.6 Wins Produced, 0.232 WP48
  • Dennis Johnson ranked sixth on this team in Wins Produced

Looking at these brief lists it’s clear that Dennis Johnson was a significant contributor to the title team in Seattle.  But in Boston – as noted above – Johnson was not the primary producer of wins.

One suspects that had Johnson spent his career with less successful teams that his selection to the Hall of Fame wouldn’t have happened.   In other words, this selection suggests that voters are simply not able to separate a player from his teammates.

Or to put it another way… Dennis Johnson produced 64.1 from 1977-78 to 1989-90 and he is in the Hall of Fame.  Artis Gilmore – as we note in Stumbling on Wins – produced 130.3 wins from 1977-78 to 1987-88 (and likely many more wins in the ABA).  And Gilmore – who never won an NBA title — is still not in the Hall of Fame. 

- 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

Wins Produced vs. Win Score

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.

Comments are closed.