What was the worst decision in the history of the NBA draft? For many NBA fans, the Portland Trail Blazers selection of Sam Bowie in 1984 – with Michael Jordan still on the board – must rank towards the top of any list of bad draft day decisions.
Bowie played for 10 seasons in the NBA. He was never named to an All-NBA team, never played in an All-Star game, and he never played for a team that won the NBA title. In fact, Bowie was never even a full-time starting player. In every season he played he came off the bench in some of the games he played.
In contrast, Jordan… well, Jordan’s list of accolades is familiar to every NBA fan. When we consider the entire careers of these two players, it seems pretty obvious that someone in Portland in 1984 just blew this decision.
Of course at this point – with the complete story of each player’s career known — our perspective might suffer from “hindsight bias.”
Robert Shiller – author of Irrational Exuberance – defines hindsight bias as “…a tendency to think that one would have known actual events were coming before they happened, had one been present then or had reason to pay attention. Hindsight bias encourages a view of the world as more predictable than it really is.”
At this point in history, we know that Jordan became Jordan and Bowie… well, didn’t. So it’s natural to think that we knew this all along. And thus, decision-makers in Portland – who clearly didn’t know this in 1984 – must not be too smart.
Just a Bit of Bias
But let’s go back and re-visit this decision and eliminate most of the hindsight bias. I say most, because I don’t want to go all the way back to the decision in 1984. No, I want to re-visit how this decision looked in 1985, or after each player’s rookie season. If we only consider each player’s first season, does the Portland decision still look so bad?
To answer this question, I evaluated every single player from the 1984-85 season (according to the Wages of Wins metrics). I then ranked the rookies in terms of Wins Produced. Here are the top ten rookies from the 1984-85 campaign:
1. Michael Jordan: 23.3 Wins Produced, 0.355 WP48, drafted #3
2. Hakeem Olajuwon: 15.0 Wins Produced, 0.247 WP48, drafted #1
3. Charles Barkley: 13.2 Wins Produced, 0.270 WP48, drafted #5
4. Sam Bowie: 10.1 Wins Produced, 0.218 WP48, drafted #2
5. Otis Thorpe: 6.8 Wins Produced, 0.170 WP48, drafted #9
6. Sam Perkins: 6.6 Wins Produced, 0.136 WP48, drafted #4
7. Alvin Robertson: 6.0 Wins Produced, 0.170 WP48, drafted #7
8. Michael Cage: 3.9 Wins Produced, 0.110 WP48, drafted #14
9. John Stockton: 3.2 Wins Produced, 0.101 WP48, drafted #16
10. Tim McCormick: 2.7 Wins Produced, 0.081 WP48, drafted #12
When we look at this list we are first struck by how amazing the draft was in 1984. A number of truly great players entered the league that year. And this list doesn’t even include Kevin Willis or Jerome Kersey (who ranked 13th and 15th in Wins Produced as rookies).
Although the plethora of talent is interesting, let’s focus on where Bowie ranks in this very talented class. Remember – as I noted earlier — Bowie doesn’t have a stellar reputation today. But if we look back on his production from just his rookie season, our perception of Bowie changes. An average player posts a WP48 [Wins Produced per 48 minutes] of 0.100. Bowie’s mark was twice as good as the average, and not terribly far behind the per-minute production offered by Hakeem Olajuwon.
Yes, Jordan was already amazing as a rookie, posting a WP48 that was more than three times the average. So MJ was better than Bowie. Bowie, though, was clearly a very good player as a rookie. And when we consider Clyde Drexler’s performance in 1984-85 – 14.5 Wins Produced and 0.272 WP48 – we can see why Portland looked past MJ (yes, we are still employing just a bit of hindsight bias when we look at Drexler in 84-85).
Yes, I know. Portland would probably been better off with both MJ and Drexler. Nevertheless, I think it’s important to note that our perception of the choice to take Bowie is colored by what happened after his rookie season. After his first season Bowie’s career was wrecked by injury. Consequently our view of Bowie’s skills has been shaped by the many poor performances he offered later on. I think, though, that The Bowie choice should be evaluated in terms of how he played before he was hurt. And that performance was very good.
The Greatness of the rookie MJ
So we see that Bowie was very good, but of course, never as good as Jordan. It’s important to note, though, that the rookie MJ was truly amazing. How amazing? Here are the top five players in Wins Produced from 1984-85:
1. Larry Bird: 27.6 Wins Produced, 0.419 WP48
2. Magic Johnson: 24.3 Wins Produced, 0.420 WP48
3. Michael Jordan: 23.3 Wins Produced, 0.355 WP48
4. Isiah Thomas: 17.8 Wins Produced, 0.277 WP48
5. Moses Malone: 16.2 Wins Produced, 0.263 WP48
Yes, Jordan was more productive – as a rookie – then every single player not named Bird or Magic. He even offered more in his first season than Isiah (and I think this was Isiah’s best season). So Jordan started his career at the top of the NBA, and it wasn’t until his second comeback with the Wizards the he fell from this perch.
So it’s certainly the case that even if we consider Bowie’s stellar first season, he still falls short of Jordan. But is the choice of Bowie truly indefensible? When we consider the fact that Bowie was an outstanding big man before he got hurt, and furthermore, the fact Portland already had Drexler (who ranked 13th in Wins Produced in 1984-85), it’s easy to understand why Portland made this choice.
And I think, a year after the fact, the choice could still be defended. Before injuries struck him down, Bowie was a very good big man. Unfortunately, we didn’t see the “good Bowie” for very long. And consequently, the people who chose Bowie over Jordan are remembered – in hindsight – as not being too bright. That memory, though, appears to be just a little bit biased.
The WoW Journal Comments Policy
Our research on the NBA was summarized HERE.
Wins Produced, Win Score, and PAWSmin are also discussed in the following posts:
Finally, A Guide to Evaluating Models contains useful hints on how to interpret and evaluate statistical models.