If you play DFS or make prop bets, accurately predicting which Super Bowl team will win can improve your odds of success. This is common sense, right? But the bigger question is how it can improve your odds.

An analysis of Super Bowl teams’ propensity to run the ball vs. passing it yields actionable intelligence on betting decisions. Interestingly, the most compelling data comes from the losing teams.

Super Bowl winners these past 19 years have averaged 27.9 rush attempts in the Super Bowl vs. 28.3 in the regular season. Runner-ups averaged only 20.1 rush attempts in the Super Bowl vs. 28.5 in the regular season. In other words, on average, both teams have been similarly run-focused entering the postseason. But game scripts have altered the approach of runner-ups.

Similarly, Super Bowl winners these past 19 years have averaged 37.3 pass attempts in the Super Bowl vs. 34.1 in the regular season. Runner-ups averaged 39.0 pass attempts in the Super Bowl vs. 33.9 in the regular season. In other words, on average, both teams have been similarly pass-focused entering the postseason. But game scripts have more dramatically altered the approach of runner-ups.

These realities have led to some fascinating conclusions.

Eight of 19 times (42%), Super Bowl winners have carried the ball more in the big game than they did, on average, during the regular season. But at no time has a Super Bowl runner-up carried the ball more in the big game than they did, on average, during the regular season.

When considering 12 of the last 19 Super Bowls (63%) had their final lead change in the fourth quarter or overtime, these results are astounding. Because if these were mostly blowout victories, we might understand why the runner-ups were forced to abandon the run. For example, the 2012 Broncos were crushed by the Seahawks 43-8, resulting in only 14 carries – less than half their season average.

But how did the 2007 Patriots – coming off playoff games with 29 and 31 carries after averaging 28 during the regular season – settle for only 16 rushing attempts in a Super Bowl where they led for more than two full quarters? The 2016 Falcons led most of the way, yet rushed only 18 times – eight fewer than their season average.

To be clear, the data shows that Super Bowl winners, more often than not, also carry the ball less than during the regular season. But in six of the games when they carried it more than during the regular season, the gap was 5+ rushing attempts.

Additionally, in four of the games when they rushed more than during the regular season, the final lead change didn’t occur until the fourth quarter’s closing minutes.

11 of 19 times (58%), Super Bowl winners have thrown the ball more in the big game than they did, on average, during the regular season. But 15 of 19 times (79%), Super Bowl runner-ups have thrown it more in the big game than they did, on average, during the regular season.

Based on what we’ve already learned, this disparity makes sense. And of course, there are direct betting implications. A 79% frequency is no fluke. Super Bowl teams tend to pass more in the big game than during the regular season. When playing from behind, this tendency is heightened.

Additionally, the high frequency of narrow Super Bowl wins in recent years has translated into more aggressive fourth-quarter play – often from both teams. In easy victories, we might see leading teams run the ball repeatedly by the middle of the fourth quarter, seeking to run down the clock to minimize opponents’ remaining possessions. But close Super Bowls have produced entirely different game scripts.