We’re close enough to the season for win projections to have some validity. For the most part, rosters are set (the important pieces anyway). A big injury or two will obviously sway our expectations, but I thought it would be interesting to take a look at expected performance today so that we can gauge the relative importance of any injury from here on in.
I’m going to use two sources for expected wins: Bovada (a proxy for Vegas) and Football Outsiders. The reason I’ve chosen these two is because their projections are readily available (free on Bovada.lv and included if you buy the FO Almanac), and, in my opinion, represent two different viewpoints.
Basically, Bovada is a proxy for “general sentiment” while FO is a proxy for “analytical projecton”. The FO viewpoint is straightforward. Regarding Bovada, remember that gambling lines are directed at the general public. The idea, for the bookmakers, is to get as close to 50% of the bets to land on either side of the over/under line. That’s why you see gambling lines move as people place their bets.
Therefore, gambling lines are essentially a reading of the “consensus” opinion of the general public (gambling public at least) for each team.
That’s how I’m getting to over or underrated. Below are charts for each division in the NFL. Listed are the teams, their Bovada over/unders, and their FO mean win projections. Also included is a column showing the difference between the two expected values (FO – Bovada). In the difference column, RED numbers are “overrated” teams and BLACK numbers are “underrated”. The absolute value of each numbers tells you the magnitude of the difference.
Today I’m going to break it out by divisions, with minimal comments for each. Tomorrow I’ll look at the most over and underrated teams and see what the difference is telling us.
Let’s start, naturally, with the Eagles.
Within the NFC East, the Redskins stand as the most “underrated” team. FO has them nearly 2 full wins higher than Bovada. I’m not sure of the full explanation, but we can assume it has A LOT to do with RG3 and the difficult in projecting recovery from an ACL tear. The Eagles, meanwhile, are technically “underrated”, though we have to acknowledge that Bovada only deals in .5 win increments, so 7.8 is nearly the same as 7.5. I’ve said before that I currently have the Eagles at 8-9 wins, but I’ve also showed that the team has one of the highest ranges of potential performance for this season (i.e. riskiest).
The Cowboys, perhaps not surprisingly, are significantly overrated. This may reflect the optimism and size of the Cowboys fan base. A lot of “homer” bets could push the gambling line up. I don’t have data to confirm that, but my guess is the Cowboys O/U lines are frequently distorted due to that factor (as are a few other popular teams, perhaps even the Eagles).
Apparently, the entire NFC North is overrated. Minnesota shows the biggest discrepancy, 2 full wins. Detroit is a “chic” pick for surprise team this year, but FO isn’t as confident as the public.
This is perhaps the most interesting division, since FO and Bovada almost could not disagree more. FO has Carolina as the best team in the division, while Bovada (the public) has them as the worst. Conversely, the reverse is true with Atlanta. Clearly, there are some severe distortions at work here. I’ll get into it more tomorrow, but this is likely the result of the type of seasonal “luck” we’ve talked about in the past. Atlanta won 13 games last year, and has kept the bulk of its roster intact. However, the Falcons had one of the easiest schedules in 2012. The team also fell on the “lucky” side of stats like fumble recovery rate (which is likely to regress).
There are some big differences here as well, but the overall outlook for the division doesn’t change. It’s a two-horse race between the 49ers and the Seahawks. St. Louis, despite getting a lot of press as an “under-the-radar” team will probably struggle to reach mediocrity.
No big surprises here, outside of the fact that Bovada has Miami at 8 wins. That seems high to me, and FO agrees. Buffalo and the Jets are interesting because they both have potential rookie starting QBs. Typically that means a poor season, but given the terrible QB play each team has had recently, we might actually see a surprise here if either EJ Manuel or Geno Smith is legit.
Projected by FO to be the tightest division, the AFC North shows no clear favorite. Pittsburgh is a bit overrated, and might be subject to the same distortion as the Cowboys. Baltimore, though, despite winning the Super Bowl, comes in as significantly UNDER-rated.
The AFC South looks like a pretty weak division. I like Andrew Luck, but the Colts are a prime contender for regression this year. I think the FO projection for Jacksonville is high. The rest looks reasonable. Houston as a 10.5 win team seems aggressive.
Denver, barring an injury to Manning, is going to be among the best teams in the league. Still, that’s a tough over to hit (12 wins or better). I’m more bullish on Kansas City than FO is, and lean towards Bovada here at 7-8 wins. San Diego is the only team in the league for which the FO and Bovada projections agree completely.
As I said at the top, I’ll have some more detailed analysis tomorrow, I just didn’t want this post to end up at 2000+ words. For now, this is a good cheat sheet for anyone trying to get their bearings on the upcoming season.
Do you have any data on predictions at roughly this stage in previous years? It would be interesting to compare a few commentators, Vegas lines, and some really simple model (4 plus half last year’s win total, or whatever that regression looks like) with how things turned out. And it would be excellent to start accumulating this data now, for the future – perhaps it already exists somewhere for past seasons but I haven’t been able to find it.
I dont have the FO data, though somebody should (anyone who still has their previous Almanac copies). I do think I have some on the Vegas lines, but I have to find it.