When the Broncos and Seahawks advanced to the Super Bowl, we football fans were given a rare gift: a meeting of what are (most likely) the two best teams in football on the game's biggest stage.
According to Pro-Football-Reference.com's Simple Rating System (SRS), Sunday's game will be just the sixth Super Bowl since the NFL-AFL merger to pit No. 1 and No. 2 from the regular-season rankings -- and the first time the top two teams have faced each other in more than a decade.
Here's a list of every Super Bowl in that group (the team's SRS is its average point margin, adjusted up or down according to its opponents' strength):
But all of those calculations don't necessarily prove the Broncos and Seahawks are the best teams in football. It's important to keep in mind that both squads have played only 18 games apiece in 2013, with their opponents representing a mere cross-section of the whole league. The NFL season is so short -- and the role of randomness so great -- that a team's performance may not reflect its true ability.
Fortunately, the Simple Rating System also has a mechanism to quantify the uncertainty in each team's rating (including playoff results). We can use this feature to compute the probability that any given team is truly the NFL's best. It can even tell us what the odds have been since the merger that the Super Bowl featured the league's two best teams. As it turns out, Sunday's game doesn't offer the greatest confidence that the two top teams are playing, but it comes pretty close.
Let's go to the list, starting with the Super Bowls for which we could be most certain involved the two true superpowers of the NFL:
Super Bowl XIII easily ranks first here, thanks to an NFL season where Pittsburgh and Dallas dominated the league all season. The Cowboys and Rams both went 12-4 in the NFC, but Dallas outscored opponents by 11 points per game, compared to Los Angeles' plus-4.4 differential. The Cowboys also crushed the Rams 28-0 in the conference championship game.
Meanwhile, nobody in the AFC was even close to the 14-2 Steelers, who cruised through the playoffs with a 67-15 scoring margin going into the Super Bowl. Statistically, there was only a 6 percent chance that any team other than Dallas or Pittsburgh had the best true SRS. The third most likely claimant to No. 1 was Miami, checking in with a paltry 3.5 percent probability.
The 49ers were the class of the NFL in 1984, leading the league with a 15-1 record (the only loss coming to the Steelers on a missed field goal with seconds left on the clock) and logging a commanding plus-15.5 PPG differential. The Dolphins rode a record-setting Dan Marino performance to an AFC-best 14-2 mark during the regular season, then put up 38 points per game in wins over Seattle and Pittsburgh in the playoffs.
So on the eve of the Super Bowl, it would have been hard to argue that San Francisco and Miami were not legitimately the league's two best teams. (Denver and Seattle had matching 8.7 percent probabilities of being in the top two, but neither did anything to bolster its case in the playoffs.)
It's easy to explain this probability. Going into the big game, there was only a 7 percent chance that the Bears were not the NFL's best team -- the second-best pre-Super Bowl odds of any team since the merger. Those Bears are the stuff of legend, and routinely come up atop lists of the most dominant teams in NFL history.
But this Super Bowl matchup suffers in the rankings because New England was hardly a lock for No. 2, with a 41.8 percent probability.
The New York Jets, meanwhile, had a 24.5 percent shot at No. 2 after leading the AFC in SRS during the regular season. (They promptly lost by 12 at home to New England in the playoff opener.)
And even the 49ers had a 13.1 percent claim on the second slot in a down year (by their standards) after beating opponents by 9.3 PPG during the regular season.
Looking back, it's easy to forget that the Raiders were favored by 3.5 points going into this Super Bowl, and they also boasted the league's best pregame claim at No. 1 in SRS with a 59.4 percent probability. It was a year of parity in the NFL (no team won more than 12 games), but Oakland outscored a relatively difficult set of opponents by a league-best 9.1 PPG during the regular season, then beat the Jets and Titans with relative ease en route to the Super Bowl.
The Bucs were the more uncertain of the two teams, with a 25.3 percent claim to No. 1 -- though it bears mentioning that no one was especially threatening to them in the race for No. 2. This was the last time No. 1 and No. 2 met in a Super Bowl; let's hope this year's game is more competitive than the 48-21 rout that Tampa handed the Raiders.
The Packers were favored by 11 points before Super Bowl XXXII, but this was more because the game pitted a prime Brett Favre against the Ghosts of John Elway Super Bowls Past than because of any statistical evidence from the 1997 season itself. The Broncos led the league in SRS during the regular season, avenged their 1996 playoff loss to Jacksonville in the wild-card round, then gutted out a couple of tough playoff road wins to punch their ticket to San Diego, proving their readiness for the big stage.
Our model says there was an NFL-best 47.4 percent probability that Denver was the true No. 1 team before the Super Bowl, compared to a 28.9 percent mark for a Green Bay team that underperformed during the regular season (at least, relative to their otherworldly 1996 campaign). Despite the oversized pregame line, it shouldn't have been surprising that the Broncos stood toe-to-toe with the Packers throughout, eventually prevailing in one of the best Super Bowls ever.
Most of the trends leading up to Super Bowl XVIII seemed to forecast a Washington victory. The Redskins were the defending Super Bowl champs, and had only improved in the 1983 season, leading the NFL in point differential by a mile. According to SRS, there was a 90.1 percent chance that they were the league's true No. 1 team before the Super Bowl, one of the highest pre-Super Bowl probabilities since the merger.
The Raiders would have been an afterthought with their 4.2 percent pregame shot at No. 1, barely better than a 49ers team the Redskins had just dispatched in the NFC title game. Of course, that's why they're probabilities and not certainties -- Los Angeles went on to shellac Washington 38-9 behind Marcus Allen's 209 total yards and two touchdowns.
Go back to a time before the Bills suffered the first of their four consecutive Super Bowl defeats, and our method assigns them an NFL-best 71.4 percent chance that they were the league's top team in 1990. Buffalo led all teams in SRS during the regular season, then averaged 47.5 PPG in the AFC playoffs -- including a brutal 51-3 thrashing of the Raiders in the AFC title game.
The Giants were second in the league in SRS (albeit a distant second), and had a 42.6 percent probability of being one of the league's two best teams after easily disposing of the Bears and then outlasting the 49ers on the road to the NFC crown. Aside from the Super Bowl participants, the closest team to the top two was San Francisco (with a 15 percent chance of ranking either first or second), making 1990 a true two-team race.
Denver is currently favored in Vegas' estimation, but SRS says Seattle has a 51.2 percent chance of being the league's true best team (compared to Denver's 23 percent). The only other contender would be San Francisco -- which has a 40.3 percent probability of being in the top two -- but the Niners are the odd team out after falling to Seattle on the road in the NFC Championship Game.
This means that we have enjoyed all of the matchups we have wanted over the season's closing weeks, culminating in one of the best Super Bowl pairings in recent memory -- a welcome change from the trend toward more chaotic playoff outcomes that started roughly a decade ago.