James is a law student from Tacoma, Washington. He is a lifelong Seattle SuperSonics fan, but is still able to enjoy the NBA, perhaps more objectively, without his favorite team. Check out his great work at Shut Up and Jam.
One player can make all the difference in the NBA. It follows that having one elite player can help a team tremendously. In fact, a number of observers have even declared that a “superstar” is imperative in building a championship team. But how important is a “superstar”? Unfortunately, the term “superstar” in inherently subjective, and though a past post observed that only 3 teams since 1980 (5 now) have won a title without a player who posted a WP48 of at least 0.300, I decided to look deeper into the topic.
Rather than attempt to define “superstar,” I looked at players in the top five percentiles of all players in Wins Produced since 1977. The 99th percentile includes players who produced at least 19.9 wins in a particular season. For example, Michael Jordan’s ’96 season is in the 99th percentile, but his ’97 season isn’t. In a given season, we would expect to see 4 players at this mark. In 2011, Kevin Love, Dwight Howard, LeBron James, and Chris Paul were the only players in the 99th percentile. The 98th percentile includes players who produced at least 16.8 wins in a season. We would expect to see about 8 players hit this mark in a given season. The 97th percentile includes players who produced at least 14.8 wins in a season, and the 95th percentile includes players who produced 12.4 wins. We would expect to see about 12 players who were at least in the 97th percentile, and 19 in at least in the 95th. In 2011, there were 19 players in at least the 95th percentile, and exactly half of the teams in the league had at least one of these players.
Now let’s take a look at some of the numbers. The following table represents all teams that have made it to at least the conference finals since 1977, and then NBA champions since 1977 alone. The numbers indicate the percentage of these teams that have had at least one player in the 99th percentile, 98th percentile or better, and so on.
| Percentage with at least one player in the: | Top Percentile | 98th Percentile+ | 97th Percentile+ | 95th Percentile+ |
|---|---|---|---|---|
| Teams in Conference Finals | 35% | 61% | 74% | 90% |
| NBA Champions | 50% | 85% | 91% | 97% |
| Example Player (from ’11) | LeBron James | Dwyane Wade | Blake Griffin | Paul Pierce |
As you can see, a 99th percentile player isn’t essential to building a champion. Only half of the NBA champions have had one. But after that, things begin to look daunting for teams without a top player. In addition, the champions really begin to separate themselves from the rest of the conference finalists. 85% of championship teams have had a player in the 98th percentile or better, and only one team since 1977 has won a championship without a 95th percentile or better player. In fact, only 10% of conference finalists have made it without a 12.7+ win producer.
Unfortunately, the “superstar” theory seems to hold true at least to some extent. There are, of course, exceptions, and well rounded teams can beat top heavy ones. But the numbers suggest that having big producers can separate the great teams from the good ones. Moral of the story: if you have an extremely productive player, keep him! If you don’t, get one! But you probably already knew that.
-James
Wages of Wins Journal 
David Morris
August 29, 2011
I guess the Mav’s are in that 3% of champions that do not have a top 5% player?
James
August 29, 2011
Actually Tyson Chandler was in the top 5%. The only team represented in that 3% is the 1979 Sonics. But the Mavs are one of 3 champions to not have a player in the top 3% (along with ’78 Washington and ’79 Seattle).
David Morris
August 31, 2011
Is there a cut off? Are you including all players or players who got a certain amount of minutes?
James
September 1, 2011
There’s no cut off, but it’s by total wins produced. It’s very, very difficult to produce a lot of wins without playing a lot of minutes, so all of the players in the top 5% probably played at least 2000 minutes.