Ranking every NBA playoff series by upset potential

— -- Despite the favorite-friendly best-of-seven series the NBA adopted for the first round of the playoffs starting in 2003, each of the past 10 postseasons has featured at least one upset in the opening round.

That streak will probably continue this season. That said, there's no obvious upset pick according to my formula, which is based on the historical odds of winning a playoff series as determined by the two teams' point differential during the regular season along with the results of the head-to-head season series.

Let's take a look at the eight series as ranked by upset probability from this metric, starting with a surprising pick that makes more sense when you look closer.

1. Boston Celtics (1) vs. Chicago Bulls (8)

Could a 1-8 matchup actually have the best chance of what would be just the fifth such upset since the NBA playoffs expanded to 16 teams in 1984?

The reason my model thinks so starts with the fact that Boston's point differential (plus-2.7 points per game) was the lowest for a No. 1 seed since the 1978-79 Seattle SuperSonics (who went on to win the championship, albeit in a much more balanced league). Only the LA Clippers-Utah Jazz series is closer in terms of point differential. On top of that, the Celtics and Bulls split the season series 2-2.

The Celtics also have the least playoff experience (5,461 minutes) of any No. 1 seed since the 2004-05 Phoenix Suns. Dwyane Wade alone (6,380 minutes) has more playoff experience than Boston's entire roster, and Chicago has nearly 7,000 more minutes total. In best-of-seven series since 2000, underdogs with an experience edge between 5,000-10,000 minutes have won about 16 percent more series than my model predicted.

Upset probability: ?29 percent

2. LA Clippers (4) vs. Utah Jazz (5)

In terms of team ability, Clippers-Jazz is easily this year's closest first-round series. Home-court advantage between the two teams was decided by tiebreaker, and they posted similar point differentials (plus-4.3 for the Clippers, plus-3.9 for the Jazz) as well as identical records.

Nonetheless, the Clippers are healthy favorites. Home-court advantage is big -- even when the favorite has an edge in point differential between 0 and 0.5 points per game, like this season's Clippers, they've won best-of-seven series 63 percent of the time since 2000. And since head-to-head performance has been an important indicator, and the Clippers won the season series 3-1, which pushes their chances of winning even higher.

Upset probability: 26 percent

3. Washington Wizards (4) vs. Atlanta Hawks (5)

Given their point differential (plus-1.8 ppg) is the weakest of the teams with home-court advantage in the first round, the Wizards should be vulnerable -- except they drew the one playoff team with a worse point differential (minus-0.9 ppg) than the other 15 playoff teams as well as three (the Charlotte Hornets, Denver Nuggets and Miami Heat) that are watching from home.

One possible edge for the Hawks: They're the other underdog with an advantage in terms of playoff experience. Despite their first-round wins in 2014 and 2015, the Wizards have the fewest playoff minutes on their roster of any playoff team this season.

Upset probability: ?18 percent

4. Cleveland Cavaliers (2) vs. Indiana Pacers (7)

My model might be overestimating Indiana's chances because the Pacers' head-to-head win against the Cavaliers came with LeBron James resting in November. However, Indiana did take Cleveland to double overtime at Quicken Loans Arena earlier this month before losing an epic game.

So at least the Pacers might become the first team since the 2012 New York Knicks to win a game against James' team in the opening round of the playoffs.

Upset probability: ?16 percent

5. San Antonio Spurs (2)?vs. Memphis Grizzlies?(7)

Along with the Bulls, the Grizzlies were the other underdog who managed to split the regular-season series with the favorite. Memphis won both matchups at home and forced overtime in San Antonio last week before losing.

Still, the Grizzlies have to make up an enormous gap in performance. Teams like the Spurs, whose point differential was at least 5.0 points per game better than their opponent's (San Antonio's was a full 6.7 points per game better), are 48-3 in best-of-seven series since 2000.

Upset probability: 14 percent

6. Toronto Raptors (3) vs. Milwaukee Bucks (6)

The Bucks probably look a little friskier as an underdog than they really are by virtue of outperforming their point differential after the All-Star break. Milwaukee went 17-10 while getting outscored by 1.1 points per game.

Toronto, meanwhile, went 18-7 after the break despite not getting All-Star guard Kyle Lowry back in the lineup until last week. The Raptors, who sport the Eastern Conference's best point differential (plus-4.2 ppg), should be comfortable favorites.

Upset probability: 13 percent

7. Houston Rockets (3) vs. Oklahoma City Thunder (6)

Going by head-to-head record might underrate the Thunder, who lost the season series 3-1 to the Rockets but were close in every game, three of which were decided by three points or fewer.

At the same time, because Oklahoma City's plus-0.8 differential fell short of the team's 47-35 record, Houston had a five-point edge in differential. Teams with an advantage between four and six points per game during the regular season are 45-4 in best-of-seven series since 2000.

Upset probability: ?11 percent

8. Golden State Warriors (1)?vs. Portland Trail Blazers (8)

While the East's 1-8 series might be close, the Blazers would need a historic upset to beat the Warriors, who posted the fourth-best differential in NBA history (plus-11.6 ppg). Golden State also swept four head-to-head matchups against Portland, though all four games came before the Blazers added Jusuf Nurkic in a February trade.

Portland was far better with Nurkic, who missed the team's final seven games of the regular season with a non-displaced fracture of his right fibula and is expected to be re-evaluated prior to the start of the playoffs. In Nurkic's 19 starts, the Blazers went 14-5 with a plus-5.3 differential.

Upset probability: ?<1 percent