Are the Vikings the NFL's most tragic playoff team?

Coach Mike Zimmer says there is "no damn curse" against the Vikings, but devoted fans know this franchise has a long track record of losing at crunch time — even when they are favored to win.

It's understandable, then, why there are mixed emotions — a cocktail of enthusiasm and dread — as the Vikings enter Sunday's NFC divisional playoff against the New Orleans Saints as a postseason favorite for the first time since 2008 (and yes, they lost that game).

From the dynamic 1998 team that lost to Atlanta in the NFC Championship to their four upset losses in the Super Bowl, heartbreak is part of the Vikings' story.

But have they suffered more than any other team? We wondered: Are the Vikings truly as tragic they seem?

First, the bad news.

Since the first Super Bowl in the 1966 season, the Vikings have lost 48 percent of postseason games they were favored to win, according to the Star Tribune's analysis of Elo ratings and game outcomes compiled by FiveThirtyEight.com. (read about our analysis below).

That's the fourth-highest rate among the 30 NFL teams that have had at least one unexpected playoff loss. The worst rates across the NFL belong to the Detroit Lions and the Kansas City Chiefs, who lost unexpectedly in last weekend's wild-card round. Both have lost 67 percent of playoff games they were expected to win.

Among remaining playoff teams, though, the Vikings have lost the greatest percentage of playoff games that Elo ratings predicted they would win.

The Arizona Cardinals and Houston Texans are the two teams that have not lost unexpectedly in the postseason. Both won the only playoff game that Elo ratings favored each of them to win since 1966.

Set aside percentages, though, and this analysis shows the Vikings have had 11 disappointments out of 23 games. No other team has had that many heartbreaking losses.

However, it's important to note that most of those games were nearly 20 or more years ago.

That loss to the Saints in the 2009 NFC Championship Game everyone talks about? The Elo ratings had the Vikings as underdogs, if that's any consolation.

In fact, the Vikings came into their last four playoff appearances as Elo-rating underdogs — losing the most recent three and beating Dallas in a 2009 divisional game where the rating only slightly favored the Cowboys.

The last time the Vikings were favored by Elo to win a playoff game was a 2008 season wild-card match against the Eagles. The Vikings had a slim 55 percent probability of winning but lost 26-14.

Prior to that, you have to go back to the 2000 season, when they were favored by Elo to beat the Saints in the divisional round. They won that one, only to be crushed 41-0 by the Giants in the NFC Championship Game. The Elo ratings gave the Giants a slight edge, but many (especially the fans) thought the Vikings' high-powered offense would prevail.

You might have forgotten that the Vikings also have pulled off the occasional upset, winning about 30 percent of playoff games that Elo ratings didn't think they would.

The regular season has seen them pull off upsets, too — including closing the 2012 season on a memorable winning streak to sneak into the playoffs (where they promptly lost to the Packers).

How do the other playoff teams stack up?

Elo ratings are giving the Vikings a 63 percent odds of beating the Saints this weekend.

In past postseason play, the Saints have unexpectedly won about one-third of games that Elo ratings expected they would lose. On the flip side, the Saints have unexpectedly lost 44 percent of playoffs games they were favored to win, slightly less than the Vikings.

That's four out of nine games. They split the difference against Minnesota — the Vikings upset the Saints in a 1987 wild-card game, but then the Saints beat the Vikings (as expected) in the 2009 season conference championship.

Since 1966, the Eagles have lost about 44 percent (7 of 16) of the playoff games they were expected to win. They haven't been favored to win a playoff game since 2008, which they lost. The odds in their last three playoff games have been either too close or have favored the other team — and they lost all three.

The Elo ratings predict the Eagles will beat the Atlanta Falcons this weekend. But since the ratings don't account for things like the Eagles' starting quarterback, Carson Wentz, being sidelined by an injury, the ratings might not be terribly reliable for this game.

Among the remaining NFC teams, the Falcons have lost the fewest games unexpectedly — just two out of seven playoff appearances. But they also haven't scored many upset wins (well, except for that one game and a couple others before that).

In the AFC, the Patriots have only lost 6 of 28 games they were favored to win, one of the lowest rates in the league. The Pittsburgh Steelers have the best track record, losing only 6 of 33 games they were expected to win, including wild-card games in 2012 and 2015.

Stepping back and looking at all the Vikings' games since 1966, you can see that unexpected losses have occurred every year. But so do the underdog wins.

Some of this might be due to how the Elo ratings work, favoring teams that have won games in previous weeks, but not taking into account factors like recent injuries or home-field advantage. Some of it might be due to the high level of parity in the league. It's impossible to say.

Under Zimmer's watch, the Vikings have done considerably better at not losing games they were favored to win — both regular and postseason.

During this year's regular season, the Vikings only unexpectedly lost 2 out of 11 games where Elo ratings predicted they would win. That's 18 percent. They also upset four opponents — Green Bay (in October), Washington, Detroit and Atlanta.

However, this weekend's game will be the first postseason game under his leadership where the team is favored.

To do this analysis, we needed a way to identify which team was expected to win each game.

We considered using point spreads or betting odds, but getting that data going back to the 1960s didn't pan out. Instead, we used the Elo rating system, a method that was originally devised for comparing chess players but has since been expanded to other competitive activities, including football. FiveThirtyEight.com calculated Elo ratings for all NFL games going back to the 1920s and publishes that data.

A team's Elo rating going into a game depends on the outcome of previous games played, and the difference between two teams' ratings can be used to calculate odds of winning. A random check of FiveThirtyEight.com's data shows their Elo ratings line up consistently with betting odds that are published on other websites.

The Star Tribune used the Elo rating probabilities to identify which team was expected to win a game, then determined what percentage of those expected wins resulted in losses. Games where the probabilities were less than five percentage points apart were excluded.

Teams are listed by current name, but data for those that have moved and changed names encompass all years.