The eighth-seeded Denver Nuggets’ upset of the first-seeded Seattle SuperSonics in the 1994 NBA playoffs is one of my fondest memories of the sport. The image of Dikembe Mutombo gripping the ball as he laid on the court with a rictus of triumphant joy on his face remains indelible in my mind 17 years later. It was the first time since the league expanded to a 16-team playoff format that an eighth seed knocked off a first seed, and it is therefore widely regarded as one of the greatest playoff upsets in NBA history.

That Nuggets team lost to the Utah Jazz in the Western Conference semifinals, but the next eighth seed to pull off a first round upset was a little more successful. That was the 1998-99 New York Knicks squad that not only beat the Heat in five games, but improbably advanced all the way to the finals before succumbing to the San Antonio Spurs. I sometimes wonder if the league liked the idea of first seeds losing in the first round and if their feelings on the subject had anything to do with why they expanded that round to a best-of-seven format for the 2003 playoffs. There has only been one eighth seed to make it out of the first round since then — the 2006-07 Golden State Warriors who shocked the 67-win Dallas Mavericks in six games.

I don’t know if I’m in the minority, but I love shocking upsets like these. Their rarity is what makes them fun — compare the NBA with the NHL, where the sixth, seventh and eighth seeds in the Eastern Conference all made it out of the first round of the 2010 Stanley Cup playoffs. But I think I’d like it if they were a little less rare, and I believe upsets like that would be more likely if the first round of the NBA playoffs still consisted of best-of-five instead of best-of-seven series.

Is there any evidence to back up my belief that shorter series lead to more upsets? I can answer that question with a definitive “maybe.” In the eight previous NBA playoffs (2003-2010) with best-of-seven first round series, 12 of the 80 series (15 percent) were won by the lower seed, and the first round “went chalk” (no upsets whatsoever) in 2004 and 2008. In the eight playoffs before that (1995-2002) with best-of-five first round series, 15 of the 80 series (18.75 percent) were captured by the lower seed, and the first round went chalk in 1997 and 2002.

OK, that’s not exactly overwhelming evidence to support my claim — especially when I include the fact that a seventh or eighth seed was only victorious twice in each of those eight-year samples. That led me to wonder how many first round sweeps there were over each of those periods and there was a slightly more significant difference there — there were 12 sweeps between 2003 and 2010 while there were 16 sweeps between 1995 and 2002.

Why would a higher likelihood of a sweep in a best-of-five series be desirable? The way I see it, by the time a team goes down 3-0 in a best-of-seven series, don’t you kind of want to see them get put out of their misery? Just another reason why the best-of-five first round is so great.

Comments (5)

  1. I get what you’re saying but the just mathematician inside me says that a win in the first round should not contribute more to a series win than a win in the final round. If the first round was best-of-five (a race to 3 might help us better picture it), then a game win would count towards 33.3% of a series win rather than 25% in a best-of-seven (race to 4) series. But I might be fine with a best-of-five, win-by-two, race to 4, whatever comes first-first round series.

  2. My opinion isn’t based on anything empirical, but I think fewer games would lead to more upsets. Basically, the larger the sample size, the more times the result would reflect the reality of the quality of team. This is why the best teams usually advance in the NBA. This isn’t the same in the NHL, even though they use the best of seven format too. The reason is that in the actual games, the average number of opportunities to score points is obviously dwarfed in comparison to basketball. So the sample size in the NBA is much larger inside the actual playing of one game. The result of a basketball game is far less random than it is in hockey.

    Bring back the best of 5! Upsets are fun.

  3. This is probably a bad time to bring this up since the Playoffs have been so fantastic. I agree that fewer Heat-Sixers games wouldn’t hurt anybody, but playoff basketball is just so good that I want more not less.

  4. Go one step further – I’ve been advocating best of 5 for the first TWO rounds myself. I’d settle for the 1st round though.

    But the highest priority in this realm is to do away with the lopsided abomination that is the 2-3-2 Finals format.

  5. don’t mean to sound cynical, but the NBA makes more money if there are more games. no way they’re moving back to a best of five especially when stern wants as much moolah as possible.

Leave a Reply

Your email address will not be published. Required fields are marked *