XFL Season Simulations: Which Team Would've Won It All?
The return of the XFL was cut short after half of the 10-game regular season was played in what has been the strangest period of sports in my lifetime.
We may have expected the Houston Roughnecks to walk through the league and win the championship, but now, we'll never know for sure.
That doesn't mean we can't wonder what if. So I simulated out the rest of the regular season and playoffs to see what was most likely to happen.
I implemented opponent adjustments to the per-play data we obtained through five weeks of the season and simulated out the rest of the schedule game by game. We know Houston was great, but would they have realistically run the table?
Projected Standings
| Team | Division | Wins | Losses | Pt. Dif |
|---|---|---|---|---|
| Houston | West | 10 | 0 | 92 |
| St. Louis | East | 8 | 2 | 50 |
| Los Angeles | West | 5 | 5 | 15 |
| New York | East | 5 | 5 | -14 |
| Dallas | West | 3 | 7 | -20 |
| DC | East | 5 | 5 | -24 |
| Tampa Bay | East | 3 | 7 | -31 |
| Seattle | West | 1 | 9 | -68 |
Yeah, kind of. Did the Roughnecks win every version of each simulated game? Of course not, but they had won at least 64% of the simulations in each of their final five matchups. They already played tough opponents, and the rest of the schedule was fairly soft: New York, DC, Los Angeles, Dallas, and Seattle. They also would've played in Los Angeles, their toughest remaining game. The average score in that matchup was a fairly tight 34-30 between two of the XFL's best offenses.
As for the St. Louis BattleHawks, they were the clear number-two team in the XFL but really weren't on pace overall with Houston in terms of adjusted point differential and efficiency. They lost by just four to Houston in Week 2, and recent hiccups against lesser teams bogged down St. Louis' adjusted metrics.
Los Angeles, DC, and New York went 5-5 in the simulated games, despite negative point differentials for the two teams out of the East. Notably, with DC and New York having the same record, it's DC that made it through, based on tie-breaking procedures. (They beat New York 27-0 in Week 2; the simulations predicted a 16-14 win for New York at home in Week 9, but it wasn't enough to overcome the tie-breaker.)
Playoffs
The would-be scenario includes the Houston Roughnecks hosting the DC Defenders in the East and the St. Louis BattleHawks hosting the Los Angeles Wildcats in the West.
Naturally, Houston's elite metrics carried them to a win at home nearly 77% of the time in my simulations against the Defenders, with an average score of 27-17.
The West battle between top-tier quarterbacks Jordan Ta'amu and Josh Johnson rated out closer with St. Louis winning 25-19 on average.
2020 XFL Championship Game
Unsurprisingly, we should have expected Houston and St. Louis to appear in the championship game. These two had dominant point differentials through five weeks, and when you adjust for opponents, the gap actually widens a bit over the rest of the league.
An added wrinkle in favor of the Roughnecks is that the game would've been hosted in their home stadium, and that tweak may have been enough to put them over the edge again.
Again, this matchup was won by Houston 28-24 in Week 2, and another close game was projected by my simulations.
Using a home-point advantage adjustment, Houston won the would-be championship game by an average score of 24-21, winning 54.3% of the simulations -- a narrow match between the XFL's two best teams in 2020.
If we assumed a neutral field, well, Houston won just 50.1% of the time. What an ending that could have been to the 2020 XFL season.
