OKC Thunder: Data shows five-game playoff series is better for upsets
A couple of days ago, we ran a story on potential NBA playoff modifications. If the playoffs are condensed, perhaps to shorter first-round series, would it benefit a team like the Thunder, which potentially could be a first-round underdog? You can read that story here.
It’s a good question. When the NBA season was suspended, the Thunder was in fifth place in the Western Conference, a game behind the fourth-place Jazz, so a 4-5 OKC-Utah matchup would have no real underdog. That’s a virtual tossup series.
Even if the Thunder was the sixth seed, it wouldn’t likely be a big underdog if the opponent was Denver. OKC is 2-1/2 games behind the Nuggets. But if the Thunder should fall to seventh, it would be a sizable underdog against likely the Clippers – OKC is four games behind Los Angeles.
The theory goes that underdogs have a better chance in a shortened series. And that makes intuitive sense, doesn’t it? The NCAA Tournament is full of upsets, and it’s a batch of one-game series. If Duke plays Mercer in the best-of-seven, Mercer’s not advancing. Maryland-Baltimore County is not beating Virginia four times.
The chances of Bucknell beating Kansas twice in a three-game series are remote. The chances of Bucknell beating Kansas four times in a seven-game series are astronomical. So we sort of get it.
The chances of the Thunder beating the Clippers in a best-of-three? Long. Best-of-five? Longer. Best-of-seven? Longest.
But it got me to thinking. Were NBA playoff upsets more prevalent when playoff series were shorter? The NBA began going to a best-of-seven format in the first round. That’s 17 seasons worth of data.
So I looked at the last 17 playoffs and the 17 playoffs before that, which were conducted with a first round best-of-five format.
Rather than go off seeding, I went off record. I found all the playoff series that matched opponents with at least a five-game differential in the standings. If the 52-win Nuggets beat the 54-win Jazz in the first round, is that really a big upset? No. An upset, sure, because Utah would have homecourt advantage. But not a big upset. I thought five wins was a large enough difference to matter, and most of the matchups in the study featured a far greater difference than five victories.
So here’s what I found. From the 1986 playoffs through the 2002 playoffs, the NBA staged 108 series that matched teams with at least a five-game difference in wins. Of those 108, 16 resulted in upsets. That’s 14.8 percent.
From the 2003 playoffs through now, the NBA staged 97 series with at least a five-game difference in wins. Of those 97, eight resulted in upsets. That’s an 8.2 percent chance.
So it seems clear that underdogs had a better chance with a shorter series.
Next, I looked at the upsets themselves. Was there a difference in the quality of the upset? Were there more major upsets in the five-game series compared to the seven-game series?
Under the five-game format, six of the 16 upsets were by teams that had at least 10 fewer wins.
Notable among that group: the 1998 Knickerbockers went 43-39 but upset the 55-27 Heat; the 1995 Rockets went 47-35 but upset the 60-22 Jazz and went on to their second straight NBA title; the 1994 Nuggets went 42-40 but upset the 63-19 SuperSonics (Dikembe Mutombo!); and the 1991 Warriors went 44-38 but upset the 55-27 Spurs.
In the eight upsets during the seven-game format of first-round series, four were by teams that had at least 10 fewer wins than their opponents.
Notable among that group: the 2007 Warriors, who went 42-40 and beat the 67-15 Mavericks; the 2011 Grizzlies, who went 46-36 and upset the 61-21 Spurs; and the 2012 76ers, who went 35-31 in that lockout-shortened season but upset the 50-16 Bulls, thanks in part to the Derrick Rose knee injury.
Those are some massive upsets in the seven-game era. So taking down the mighty remains possible. But it appears that the five-game series gives the underdog a better chance.