The overwhelming consensus is the Cleveland Cavaliers and Golden State Warriors will be the first teams in league history to meet in three consecutive NBA Finals, playing the rubber match of their series after splitting the past two titles.
But guess what? ESPN's Basketball Power Index (BPI) says that the matchup is not likely to happen.
And almost any other statistical projection will reach a similar conclusion.
Why isn't a rematch a sure thing?
Let's look at the reasons this movie trilogy might not be as predictable as we think:
The best team doesn't always win
The NBA's best-of-seven series aren't as conducive to upsets as the one-and-done format used in the NCAA tournament and the NFL playoffs, but they still happen. Even if Golden State is truly as dominant as statistical projections suggest -- and BPI is relatively conservative here, putting the Warriors just seven games ahead of the San Antonio Spurs on average, as compared to 12 games for projections based on ESPN's real plus-minus (RPM) and 16 for FiveThirtyEight's CARMELO projections -- that alone doesn't guarantee a trip to the Finals.
Using RPM projections, if Golden State is exactly as good as projected, we'd expect the Warriors to get upset en route to the Finals about 13 percent of the time based strictly on the randomness of seven-game series, before even considering injuries or the possibility our projections are wrong. Meanwhile, Cleveland is only about 50-50 to reach the Finals based on RPM projections and the randomness of seven-game series.
The Cavaliers and Warriors might not be the best teams
What are the chances we're overestimating Cleveland and Golden State or underestimating their opposition? I've gone back to calculate RPM projections for the past seven seasons using preseason minutes estimates and player projections for those years. The average error for team forecasts over that span is 6.5 wins, though it has dropped to about six wins on average the past couple of seasons.
Over an 82-game season, six wins is a lot, especially when we're looking at the gap between two teams, because both win totals can be expected to be off by about six games on average -- for a total of 12 games.
In the Western Conference, the Warriors' 12-game cushion in RPM projections means we could be overestimating them and underestimating the Spurs by an average amount and still have them in a dead heat for first in the conference. The Eastern Conference is a different story. The Boston Celtics and Toronto Raptors are both within six games of the Cavaliers' projection, making it realistic they finish ahead of Cleveland in the regular season.
Of course, projection error isn't the only way we could be wrong about the best teams in the two conferences. It's possible another contender -- the Celtics are an obvious possibility -- swings an in-season trade that closes the gap with the favorites. That scenario is nearly impossible to model.
But it has happened at least twice in recent history, with the Lakers getting Pau Gasol in February 2008, four years after the infamous trade that landed Rasheed Wallace in Detroit and carried the Pistons to the 2004 title.
Somebody might get hurt
There's no more obvious source of uncertainty with any sports projection than the possibility of injury. While Stephen Curry was able to return from the sprained MCL he suffered in the opening round of last year's playoffs, his limited mobility was a key factor in why the Warriors weren't able to maintain their historic pace in the postseason. And Andrew Bogut's absence after suffering bone bruises in Game 5 may have help swung a series that wasn't decided until the final minute of Game 7.
LeBron James going down during the playoffs is the sort of black swan event that's almost impossible to predict with any certainty, particularly because it has never happened before -- James has suited up for all 199 playoff games his teams have played in his career.
The interesting question is whether these teams are deep enough to survive a major injury in the playoffs. If Curry is limited this time around, Golden State now has another MVP to run its offense through in Durant. And the Cavaliers reached the 2015 Finals without the injured Kevin Love and with limited contributions in the postseason from Kyrie Irving.
Cleveland could be more vulnerable to an Irving injury now after losing backup point guard Matthew Dellavedova, who played 40-plus minutes twice in the five games after Irving fractured his patella during the 2015 Finals. While James can run the offense, the only other traditional point guard on the Cavaliers' roster besides Irving is rookie second-round pick Kay Felder.
With Durant's arrival, the Warriors look better equipped to survive a key absence. An injury in the frontcourt might force them to play small more than Steve Kerr would prefer, but conveniently those lineups might be the team's best.
On the other hand: We might be underestimating the Cavs
There's one key issue with the statistical projections we've considered so far: They're designed to project the regular season, not the playoffs. Because these two teams are top-heavy with superstars -- and because both coaches are likely to be conservative with minutes during the regular season -- Cleveland and Golden State will probably be even better in the postseason than these projections indicate.
That has certainly been the case for the Cavaliers in recent seasons. They won the East by only four games during the 2015-16 regular season before going 12-2 en route to the Finals, and managed to sweep the Atlanta Hawks (who were battling their own injuries) in the 2015 Eastern Conference finals despite Atlanta having home court in that series.
Thanks to that track record, and the history of James' teams winning the East each of the past six seasons, the betting market is much higher on Cleveland's chances of reaching the Finals than statistical projections. In fact, per VegasInsider.com, the Cavaliers and Warriors have identical 1-to-3 odds of winning their respective conferences, which implies they're expected to do so about 65 percent of the time after accounting for the house edge.
That's about the same as BPI's forecast for the Warriors, who make the Finals in nearly 64 percent of simulations. However, BPI has the Cavaliers winning the East just 45 percent of the time, much lower than Las Vegas expectations. In all, BPI sees a Cavs-Warriors three-match as only 29 percent likely.
Even the implied Vegas odds suggest a third consecutive Cleveland-Golden State matchup is unlikely, at only about 42 percent to happen. So take heart, fans of other contenders in the East and West. There may still be rewrites coming before we get to the end of this season's NBA movie.