ESPN’s predictive model expecting another painful year for Detroit Lions

With free agency and the draft out of the way, it’s predicting season! This is the time of year to get irrationally angry as media members take stabs at guessing how the 2019 NFL season will play out. For Detroit Lions fans, that means we’ve entered prime “DETROIT VS. EVERYBODY” territory.

While some NFL writers will just conjure up their best predictions out of a hat, others will rely on advanced metrics to spit out an equally wrong guess.

On Thursday, ESPN dropped their 2019 projections based on their own predictive model. Using their Football Power Index (FPI) projections—which, by their own admission, uses Vegas expectations as their “main component” in the preseason—they are predicting another subpar year for Detroit.

According to their model, the Lions are projected to average 7.0 wins (t-22nd). They only have a 17 percent chance at making the playoffs, with less than a one percent chance for a Super Bowl win.

Also interesting is ESPN Sports Analytics Group’s outlook for the NFC North. They give just about everyone who isn’t the Lions a realistic shot at the division crown. The Bears lead as 38 percent favorites, followed closely by the Packers (27%) and Vikings (26%). The Lions have a mere eight percent chance to finish atop the division.

Of course, this predictive model means almost nothing. It’s based on Vegas odds, which are set not to try and guess what will happen in 2010, but to try and mirror how bettors will act. So this method of predicting the season has some serious flaws. Though the model’s predictions can sometimes be more accurate than an analyst’s opinion, at this point in the year, it’s still based more on perception than anything else.

You may remember a week ago for Todd McShay’s 2020 mock draft ESPN used a different predictive model to determine draft order. According to Football Outsiders’ projected records, the Lions will be picking 23rd next year, meaning they would qualify for the postseason.

So, essentially, pick whatever prediction model you like the best, because they’re both likely going to be wrong come December.