The 2025 NBA playoffs have been marked by comebacks, and nobody has mastered the art quite like the Eastern Conference champion Indiana Pacers.

The Pacers rallied from a seven-point deficit in the final 40 seconds of overtime to complete their 4-1 series win over the Milwaukee Bucks in Game 5 of the opening round, then accomplished the same feat in the final 50 seconds of their Game 2 victory over the Cleveland Cavaliers in the Eastern Conference semifinals.

All of that was an appetizer for Indiana’s most improbable Houdini act in Game 1 of the Eastern Conference finals. Down 14 points inside the four-minute mark of regulation and eight in the final minute, the Pacers used a flurry of Aaron Nesmith 3-pointers, untimely New York Knicks turnovers and Tyrese Haliburton’s high-bouncing shot at the buzzer to force overtime and eventually steal the opening game of the series.

Inspired by Indiana’s late rallies and New York winning three games it trailed by at least 20 points, the most by a team in a single playoff run in the play-by-play era (starting with the 1998 playoffs), let’s take a closer look at how comebacks have dominated the 2025 playoffs.

Are the Pacers the greatest comeback team in NBA playoff history?

The short answer, limiting it to the period where we can actually quantify comebacks, is almost certainly yes. The long answer of quantifying this title is complicated.

Nobody has done more work on win probability and comebacks than Mike Beuoy of Inpredictable.com, a terrific NBA resource. The website rates each game with a comeback score based on the likelihood of the winning team’s low point, and Indiana’s three long shot wins rank among the seven best comebacks in the playoffs since 1997.

Inpredictable also looks at the average comeback score for each of a team’s wins. (Technically, it’s the geometric mean, which weighs one outlier comeback less heavily than taking the traditional mean.) Despite that adjustment, it’s still easier to build a high comeback score with fewer ones rather than as many as the Pacers (and Knicks) have thus far. When we graph the average comeback score for each playoff team since 1997 against their wins — with this year’s teams highlighted by primary color — it’s obvious what outliers Indiana and New York are.

Working with Beuoy, we tried a few different methods to find a single comeback method that accounts for both the volume of comebacks and their improbability. The most satisfying we found was taking the product of the likelihood of each win at its lowest point — which is to say, the odds that a team would win all the games it did in the playoffs.

Looking at things this way heavily favors teams with more playoff wins, whether via comeback or not, since no game is a 100% win probability from the jump. Nonetheless, the Pacers’ 12 wins (and counting) rank second since 1997 in this group, trailing only the 2011 champion Dallas Mavericks, who had 16 wins. You can see Dallas on the chart as having the highest comeback score of any title winner.

Meanwhile, the Knicks’ 10 wins rank seventh, higher than any team before this year without reaching the Finals.

Are win probability models underestimating the chances of comebacks?

All three Indiana comebacks have seen the team emerge victorious after win probability estimates gave that a 2.1% chance of happening or less, including 0.9% against New York. It’s not quite getting struck by lightning twice (your odds of that happening once are estimated by the National Weather Service at .000065%, based on an 80-year lifespan), but it’s awfully unlikely by random chance alone.

Based on that, you can be forgiven for some skepticism about win probability. Part of the challenge is these estimates are based on historical data that may not always keep up with the fast-moving NBA. ESPN’s model, for example, was built in 2017 based on training data from the previous seven or so seasons. As comebacks grow more common due to a faster pace of play and an increased volume of 3s, a trend I wrote about with ESPN’s Baxter Holmes back in 2019, it’s possible we’re underestimating the chances to a degree.

The other issue is calibration. All models have uncertainty, but the difference between a win probability of 57% and 58% is irrelevant in most practical contexts. At the extremes, the uncertainty is magnified because a comeback from a 98% win probability is twice as likely as one from 99%. And a comeback from 99% (one in 100 odds) is 10 times more likely than one from 99.9% (one in 1,000). So even small calibration issues are important.

Is there a stat that quantifies the Pacers’ offense and defense playing off each other so well?

There could be an explanation for why this postseason in particular has seen so many comebacks, when most factors that are cited have been in place for years: the relationship between offense and defense. Offenses are typically somewhat more effective after getting a stop because it allows more opportunities for early offense and cross-matching on defense, but the benefit of getting a stop (or vice versa) can depend on a variety of factors that change from team to team and season to season.

In general, these playoffs have featured an enormous difference in efficiency depending on whether the offense starts on a defensive rebound or takes the ball out of the net after a made basket. Returning to Inpredictable.com, their data shows teams averaging a 1.17 points per possession after a defensive rebound as compared to 1.07 after a made shot or a dead ball turnover. (The average off steals, or live ball turnovers, is much higher at 1.23 points per possession.) That’s a change from the past few playoffs, when the difference by start type has been much smaller — just 0.01 points per possession better in 2022 and 2023.

As for why that might have changed, I would point to increased physicality being allowed by referees in the playoffs the past two years. Inevitably, physicality is more of an issue during half-court situations rather than transition. During the 2023 playoffs, when the whistles were tighter, teams averaged a more efficient 1.1 points per possession after a made shot or dead ball turnover.

I think that could tie into why avoiding turnovers has been crucial in this year’s playoffs. As Owen Phillips of the F5 Newsletter has been tracking, the team with fewer turnovers has gone 53-20 (.726), which would be the highest winning percentage for such teams in a single playoffs on record. Last year, teams with fewer turnovers won just 60% of the time, about the average over the past decade (62%). Turnover battle winners were barely better than .500 in 2018-19 (41-37).

It’s tougher to explain why teams are scoring so efficiently off defensive rebounds this year, though fatigue could be a factor with starters on several teams that reached the conference semifinals logging heavy minutes.

Switching our focus to Indiana specifically, the Pacers do derive more benefit than most teams from defensive rebounds. They’ve averaged 1.26 points per possession after those, per Inpredictable, third best in the NBA. Although Indiana is still third after a made shot or dead ball, their efficiency drops by an above-average .16 points per possession.

At the other end of the court, we see a similarly large split. The Pacers’ defense is 10th best after a made shot or dead ball turnover and .17 points per possession worse after a defensive rebound, dropping to 14th.

Now, what does this have to do with comebacks? The bigger the gap between stops and scores at the other end of the court, the streakier a team (or league) is likely to be because the magnitude of each possession is amplified. A stop isn’t just preventing the opponent from scoring, it’s also juicing the team’s offense, and vice versa — a virtuous cycle or a vicious one, depending on your perspective.

The streakier the game, the more likely teams are to build big leads, and the more likely opponents are to rally from them. Add it up and you’ve got the recipe for Indiana’s comebacks.

On the flip side, despite losing a fourth-quarter lead in Game 1 of their series with the Denver Nuggets and staging a comeback from down 26 at halftime against the Memphis Grizzlies, the Oklahoma City Thunder haven’t been as dependent on their offense to succeed defensively. Oklahoma City has been very good defending after a made shot or dead ball turnover (second on a per-possession basis after the Detroit Pistons) but is allowing .08 points per possession fewer than any other team on possessions that begin with defensive rebounds.