Key takeaways
- The Week 2 overreactions are flying, and three big favorites earned them. The Rams lost as 8.5-point favorites to San Francisco, the Cardinals lost as 7.5-point favorites to Tennessee, and the Bills lost to New England as 8-point favorites.
- The Patriots result is the one with the most signal in it. Mike Vrabel's teams have always been annoying to play, and beating a division favorite by more than a touchdown of spread is not something a bad roster does by accident.
- The Rams loss is the one with the least. Division games between good teams are close, and an 8.5-point number in that matchup was always doing more work than it should have been.
- Arizona is the pricing question. Being laid seven and a half points is a statement about who the market thinks you are, and Week 2 was a fairly loud rebuttal.
- Survivor players: this is the warning, not the disaster. Two weeks in, the market is still guessing, and the most-picked team in any given week is a popularity contest with an entry fee.
Week 2 overreactions are a seasonal tradition, and this year the raw material is unusually good. Three teams the market had priced as comfortable winners lost outright, and every one of those numbers was big enough that a lot of people had already spent the win.
So let's do the actual work instead of the hot-take version. Some of what happened Sunday is information about how good these teams are. Some of it is the normal, boring noise you get from a fourteen-game sample spread across a league where the third-best team can beat the second-best team on any given afternoon. Sorting one from the other, in Week 2, with almost no data, is the entire job.
Here's my read, with the caveat that anyone claiming certainty two games into a season is selling something.
What actually happened: three favorites, three losses
The verified damage, in order of how loudly people yelled about it:
- Los Angeles Rams, 8.5-point favorites, lost to San Francisco. The biggest number of the three, and the one that gutted the most pools.
- Arizona Cardinals, 7.5-point favorites, lost to Tennessee. A team the market has been slowly talking itself into, laying a full touchdown, losing outright.
- Buffalo Bills, 8-point favorites, lost to New England. The one that will get replayed all week, because it's a division game with a new coach on the other sideline.
Three double-digit-implied-probability favorites in one weekend is not unheard of. Three of them at 7.5 points or larger is a bad Sunday for anyone who thinks the number is the answer instead of the question.
Week 2 overreactions worth having
Not every overreaction is wrong. Some early-season reactions are just correct information arriving before people are emotionally ready for it. Two of these qualify.
New England is the real one. Mike Vrabel took over in New England this offseason, and whatever else you want to say about his teams, they have never been fun to be favored against. A first-year staff beating a division rival that the market thought was more than a touchdown better is not a fluke result you can wave off with "they got a couple of bounces." It's a data point about coaching, effort and a defense that showed up. I'd move New England's number up and I'd stop treating them as a free win for anybody in the AFC East.
Arizona is the second one, in the opposite direction. Laying 7.5 points is the market saying you are a legitimately good football team playing a legitimately worse one. Losing that game outright doesn't prove the Cardinals are bad. It does suggest the price was ahead of the product, and prices that get ahead of products tend to stay there for a few weeks before they correct. That's a pick'em opportunity, not a reason to panic about anybody's fantasy roster.
The useful question in Week 2 is never "was this team good or bad?" It's "was the number right?" Teams regress toward what they are. Numbers regress toward what the market believes. Those correct at completely different speeds, and the gap between them is where every pool is decided.
Week 2 overreactions to throw away
The Rams game is the one where everybody's takes got the furthest out over their skis, and I think it's mostly noise.
Here's the argument. San Francisco is not a bad team, and it has never been a bad team in this matchup. Division opponents know each other's tendencies, they play close games by default, and 8.5 points in an NFC West game was always a number built on preseason narrative more than on any evidence from this season. The Rams didn't get exposed. They played a rival on a normal Sunday and lost a game that the pricing said they'd win comfortably, which is a statement about the pricing.
The Bills side of the New England result is also overcooked. A good team losing a division road-style dogfight in September is the most ordinary event in football. I'd take the signal about the Patriots and I'd take almost none of the signal about Buffalo. Those are separable, even though the same final score produced both.
A rough rule for the first month: results tell you a lot about the worse team and very little about the better one. An underdog winning proves it can play. A favorite losing one game proves almost nothing at all.
The survivor pool warning nobody wants to hear
Now the part that matters most on a pick'em site, because Week 2 was a survivor massacre and it will not be the last one.
All three of those teams were popular picks. That's what an 8.5-point spread does — it makes a game look like a free square, and free squares are exactly what survivor players are hunting in September when they think they're saving their good teams for later. The result is that a huge chunk of a pool can be standing on the same one or two games, and when one of those games goes sideways, the pool doesn't shrink. It gets clear-cut.
Three things to internalize now, in Week 2, while it's still cheap:
- A big spread is not safety, it's confidence. An 8.5-point favorite is not a 95% proposition. It is closer to a coin flip than your gut wants to believe, and you're going to make sixteen or more of these picks.
- Popularity is a real cost. If you and 30% of the field pick the same team and it loses, you didn't just lose — you lost with a crowd, which is the worst possible outcome in a contest where surviving alone is worth the most.
- Stop saving teams. The "I'll use them later" instinct assumes you'll still be alive later. Two weeks in, we've already watched three near-lock favorites lose in a single Sunday. Later is not promised.
If you're in one of our pools, this is a good week to actually look at the distribution instead of your own gut. The weekly leaderboard tells you what the field did, and knowing what everyone else picked is worth more than one extra hour of reading matchup previews.
How to price Week 3 without whiplash
The trap after a weekend like this is fading every favorite out of spite. Don't. The correct adjustment is smaller and more specific than that.
- Adjust the teams whose price moved on new information — New England up, Arizona down — and leave the rest alone.
- Discount division games in the first month. The market is still pricing them off last season's records, and last season's records were built by rosters that have since changed.
- Don't chase the number. A team that just lost as a big favorite often gets a slightly friendlier price the following week, which is a discount on a team that has not actually gotten worse.
- Keep your survivor picks boring and your spread picks brave. They're different contests. Play them differently.
Two weeks is nothing. We have no idea yet whether New England's win is the start of something or a nice afternoon, whether Arizona's number was wrong or just early, or whether the Rams' loss will look like a blip by October. What we do know is that the market has been wrong three times in one Sunday, which is a decent reminder that it can be wrong again next week — and that the people who wait for certainty in a survivor pool tend to get eliminated before certainty arrives. Get your Week 3 picks in early, then go watch what the field does.
If you're still building your roster philosophy for the year, the same logic we ran through in our late-round QB strategy piece applies here: pay for flexibility, not for reputation.