Picking football games is always about mitigating risk — but we need to establish a few extra ground rules every time NFL Week 1 comes around.
NFL Week 1 is in full swing, and we witnessed a great opener between the Buccaneers and the Cowboys. We always knew what we were getting from Tom Brady and the Tampa Bay offense.
We also had a pretty good idea about the potential ceiling of the Cowboys’ offense, but there was some uncertainty surrounding Dak Prescott and his ankle and shoulder injuries.
And that’s just the way it is. Uncertainty is the ultimate keyword in sports betting — but even more so in Week 1 of any NFL season.
I’m an originator. What does that mean? I come up with my number — my projected spread — for every NFL game. There goes a lot into the weekly process of betting games, but in theory, I come up with a number on a game and hope to find a decent deviation against the market number. On every game, there’s a price on which you would bet either side. That deviation between my number and the market number is called the edge.
When the edge is at a specific size, and I gather enough information, I’ll bet the game.
The primary goal (albeit not the end goal every time) for NFL handicappers is to beat the closing line. That’s a complex topic on its own; Boardroom might just attack it at some point this year. In sports betting terms, the closing line value is the difference between the implied probability of the price you bet into versus the implied probability of the closing price when a game starts.
At kickoff time, betting markets have reached their most efficient state. That state occurs after a price discovery process when the most information is available and liquidity is at its peak.
In theory, the closing line represents the “fair price,” the most accurate implied probability for the sports event. If you bet Packers -4 and they close -6, you have done your job. You generated a positive closing line value.
(A quick side note: In some situations, you don’t care about closing line value. But that’s for a different article.)
A Lack of Data
There’s a significant difference in betting week one games compared to betting in week nine. Or week thirteen. Or any other week. Even though the early part of the season has a common issue, Week 1 is an entirely different ballpark. The reason is simple: We don’t have a single in-season data point on any of the 32 NFL teams. There aren’t any actual games to take into account yet.
Instead, we enter Week 1 with our prior assumptions built through statistical modeling or any kind of subjective approach. We are betting that we have an excellent prior before adjusting our beliefs with more data and information throughout the season.
Week 1… not a lot of information… efficient markets… how does it all fit together?
You might think that bettors are having a hard time in week one and that the market must be weaker compared to later during the season, but that’s not true. Even though many bettors will (almost logically) find more edges earlier in the season, the market is usually just as efficient as in other weeks.
Over the past eleven seasons, the mean absolute error (MAE) between the closing spread at Pinnacle (one of the sharpest bookmakers in the world) and the actual game result is 10.24 points during the regular season.
Yes, it’s true. NFL spread betting markets are “off” by two scores on average, and it’s still tough to beat them at all.
In Week 1, the MAE across the eleven seasons is 9.9. Week 2 sits at 9.6. Even though we don’t have any actual data points to go on in Week 1, the market is still pretty good at this whole betting thing. The essential difference of 0.34 points between opening week and the entire season can be subject to noise, but it’s important to note that there’s not a significant weakness on display here.
Uncertainty is not a problem; it’s a game.
More or Less Uncertainty
Many roads lead to Rome. You can ask a hundred professional or semi-professional bettors about how they approach NFL betting markets to grind a profit, and everyone will give you a different response. With that in mind, for Week 1, I try to avoid the more blantant kinds of matchup uncertainty as much as possible.
We deal with uncertainty all the time on every single bet we make, but at the very start of the season, certain matchups are just more chaotic than others. It’s the bettor’s job to identify and sidestep them.
Let’s take Sunday’s Bird Bowl between the Atlanta Falcons and Philadelphia Eagles, for instance. The only thing I can say with certainty is that Philly will have a solid offensive line and that Arthur Smith should be an upgrade over last year’s coaching mess. But I have no clue about what the Eagles are actually going to look like overall.
Other major questions we don’t have good answers to regarding this matchup:
- How much will Jalen Hurts improve in year two as a passer?
- How good can Nick Sirianni be as a head coach and offensive mind?
- How does the Falcons offense shape up?
The Eagles have a significant advantage in the trenches on both sides, but is that enough to overcome the uncertainty on offense? For what it’s worth, the Falcons are favored by a field goal at home as of this writing.
When it comes to rookie quarterbacks in Week 1, I try to avoid them entirely unless there’s an excellent edge. We just don’t know what we are getting from Trevor Lawrence, Mac Jones, and Zach Wilson, so I’m not wasting any time hunting for tiny edges. The first few weeks will reveal what to think of these guys. Today, we’re not there yet.
One of the matchups that I bet this week is Sunday’s Broncos-Giants game. Aside from Denver’s QB switch to Teddy Bridgewater, there aren’t tons of changes we need to project. We have a pretty good idea about what kind of player Bridgewater is. We have a grasp of what Jason Garrett’s Giants offense will look like. It’s safe to say that Denver might have one of the better defenses in the league.
These are the kind of matchups I like to dig into in Week 1 — the uncertainties feel more manageable than a lot of other matches that feature, for instance, a debutante QB or teams debuting all-new schemes.
As the season progresses, we can incorporate a lot more data and information into our handicapping process that reduces some of the uncertainty. Until then, pick your battles. There’s value out there, but opening week is always a different beast.