Josh Allen Costs a Fortune. Buffalo's Other Fantasy Points Do Not.

Fantasy managers are not exactly keeping Josh Allen a secret.

He is QB1 by the market. Pick 25 overall.

We have reached the stage of the relationship where nobody needs to pretend this is casual anymore.

And then something strange happens.

The market values Buffalo as only the 21st-most valuable fantasy roster.

The AutoDraftYears model has it 9th.

Apparently fantasy managers believe Josh Allen is one of the most valuable players alive while remaining deeply suspicious of almost everyone who catches or carries the football next to him.

Which gives Buffalo a much more interesting question than Is Josh Allen good?

If the Bills keep producing one of football's best offenses, where are all the other fantasy points going?

What happened in 2025? Buffalo became a running team

Buffalo went 12–5 in 2025 and made the playoffs for the seventh consecutive season.

Nothing particularly alarming there.

The surprising part was how they did it.

The Bills finished first in the NFL in rushing yards, rushing yards per game and rushing touchdowns, piling up 2,714 yards on the ground. James Cook won the NFL rushing title with 1,621 yards, averaged 5.2 yards per carry and scored 12 rushing touchdowns.

Josh Allen was still Josh Allen.

But the offense around him increasingly looked like someone had purchased a Ferrari and decided the safest way to preserve the engine was using it primarily for Costco trips.

Allen threw for 3,668 yards and 25 touchdowns, well below the passing volume we've grown accustomed to seeing attached to Buffalo. The Bills threw on only 48.2% of their offensive snaps.

And the receiving production followed.

Khalil Shakir led Buffalo with only 719 receiving yards.

Dalton Kincaid was second with 571.

Nobody reached 800.

That's the important part of the 2025 story.

Buffalo did not become a worse offense.

It became a different offense.

Joe Brady leaned into the running game, Cook had the best season of his career, and the Bills won 12 games without requiring Allen to spend every Sunday throwing 42 passes into increasingly desperate coverage.

From a real-football perspective?

Hard to argue.

From a fantasy perspective?

Slightly inconvenient if you drafted a Buffalo receiver hoping Allen would personally escort him to 1,300 yards.

And now Buffalo has changed the equation again

The Bills did not respond to that receiver production by saying, This seems fine.

They traded a second-round pick to Chicago for DJ Moore, who joins Shakir and Kincaid in the passing game. Moore has surpassed 1,100 receiving yards four times in his career and reunites with Brady, who coached him in Carolina.

Brady himself is also moving.

After two seasons as Buffalo's offensive coordinator, he replaced Sean McDermott as head coach in January. More importantly for our purposes, Brady has said he intends to continue calling the offensive plays.

So this isn't a brand-new offensive system.

But the personnel suggests Buffalo would probably like the passing game to become more than Josh Allen occasionally checking whether anyone is open before handing the ball to Cook again.

That's exactly where the model lands.

The model expects the same scoring with more passing

Buffalo scored 28.3 points per game in 2025.

The 2026 projection:

28.3.

First in the NFL.

No rebound required.

This offense has been doing approximately this forever, or at least long enough that I have forgotten what Sunday afternoons felt like before Josh Allen scored fantasy points.

The meaningful changes are underneath the scoring projection:

That 5.5-point increase in pass rate is the entire hinge.

Because the positional environments are already excellent:

• QB: 5th
• RB: 2nd
• WR: 5th
• TE: 5th

Buffalo's worst fantasy environment is fifth.

A terrible hardship.

History says there should be more than Allen and Cook

The 30 historical offenses most similar to Buffalo's 2026 projection produced 3.6 fantasy starters on average.

Across all 320 team-seasons in the AutoDraftYears sample?

2.6.

Comparable teams produced:

The model's reasonable expectation is roughly three fantasy starters in a bad outcome and four in the base or bull case.

Which makes the market's 21st-place team valuation difficult to reconcile. The model has the same roster 9th. That is a 12-place gap, comfortably past the eight-place bar where the model stops treating a disagreement as noise and starts calling a roster underpriced.

Not because Josh Allen is underpriced.

He absolutely is not.

Because somebody besides Allen and Cook should benefit from an offense this large.

And that's where the draft board gets interesting

Market and Model are ranks within the position. Edge is the model's position rank against the market's, after removing the drift that affects every player at that price — so it reads as “the model likes him this many places more than it likes the typical player who costs this much.” A call needs the edge to clear the position's bar (QB 5, RB 15, WR 11, TE 5 places), set at two-thirds of that position's own spread.

This is my favorite kind of team disagreement.

The model is much more bullish on Buffalo than the market.

But it isn't getting there by declaring Josh Allen cheap.

Allen is market QB1.

The model ranks him QB6.

James Cook?

RB7 versus model RB6.

Perfectly reasonable.

Fantasy managers have successfully identified the two most obvious Buffalo players.

Congratulations to all involved.

The discount begins farther down.

Khalil Shakir is the actual Buffalo bet

Shakir goes around pick 132 as WR57.

The model ranks him WR22.

That's already a ridiculous-looking gap.

After adjusting for the normal amount of ranking chaos that exists around pick 132, Shakir still owns a +31 positional edge.

The WR buy threshold is 11.

He did not sneak over it.

He drove through it.

And Shakir isn't some player we're projecting from three good preseason routes and a training-camp quote involving the word explosive.

He led Buffalo in receiving in 2025.

Yes, the total was only 719 yards.

That's the point.

He led a passing game that barely passed.

Now the model expects Buffalo's pass rate to rise from 48.2% to 53.7%, while still projecting the Bills as the league's highest-scoring offense.

I don't need Shakir to become an alpha receiver.

At WR57, I need him to remain involved while the passing pie gets larger.

That's a much easier bet.

DJ Moore complicates it — but not enough

Obviously, Moore matters.

Buffalo did not spend a second-round pick on him so he could provide leadership during hydration breaks. The Bills acquired an established receiver who has topped 1,100 yards repeatedly, and he enters camp as one of Allen's most accomplished outside weapons in years.

The model likes him too.

Market: WR26.

Model: WR17.

But after adjusting for price, his edge falls to only +4.

That is the difference between liking a player and finding a price worth attacking.

Moore may lead Buffalo in receiving.

Shakir can still be the better fantasy purchase.

Those statements are allowed to coexist.

Fantasy football occasionally permits nuance, usually when nobody is looking.

Allen and Cook are exactly what premium players look like

There is nothing wrong with drafting Josh Allen.

There is also nothing wrong with James Cook.

Cook just won the rushing title. Buffalo led the league in rushing. The model ranks the Bills second at running back and Cook RB6.

That's excellent.

It's also RB7 pricing.

Allen is even simpler.

QB1 market.

QB6 model.

Pick 25.

You're buying the penthouse.

Maybe the view is incredible.

Nobody forgot to include it in the listing.

The number that decides this argument

Pass rate.

Not Buffalo's record.

Not Josh Allen's touchdown total.

Not whether Shakir catches a 52-yard touchdown in Week 2 and suddenly becomes the waiver-wire analyst on everyone's television.

Pass rate.

Buffalo threw on 48.2% of its snaps in 2025.

The model expects 53.7%.

That doesn't sound like a dramatic difference until you apply five extra percentage points of passing to one of football's highest-scoring offenses.

More Allen attempts.

More targets.

More chances for Moore, Shakir and Kincaid to coexist.

If Buffalo is still sitting below 50% through September, I become less excited about the receiving values.

If they're around 54%?

Then the model's 3.6-starter projection starts looking considerably easier to reach.

So how should we draft Buffalo?

Don't interpret a bullish Bills projection as permission to draft Josh Allen at any price.

The market already understands Allen.

It understands Cook.

And after the Moore trade, it largely understands Moore too.

The actual disagreement is farther down the board.

Khalil Shakir at WR57 is the cleanest way to bet that Buffalo's passing game gets even slightly larger.

That is what the team-level model is useful for.

Not telling us Josh Allen plays in a good offense.

I had managed to piece that together.

It's telling us that an offense projected to score the most points in football probably shouldn't have its leading 2025 receiver sitting around until pick 132.

Fantasy managers are paying handsomely for the privilege of drafting Josh Allen.

I'll take one of the guys he throws to.

Sources: Auto Draft Years FF Model workbook — Team Data By Year, Team Stat Ranking, 2026 Team Context; Auto Draft Years 2026 consensus board (16 authoritative market sources); Player outcomes 2016–2025, Pro Football Reference via the project dataset. Team environment grades, forecasts and player model ranks are proprietary. Every figure on this page is generated from the dataset at build time; none is typed in by hand.