The Raiders Have Plenty of Room to Improve. That's the Good News.

Fantasy football loves improvement.

We are less interested in whether something is actually good than whether it is getting better.

Bad rookie becomes competent? Breakout.

Terrible offense adds a quarterback? Sleeper unit.

Team crawls out of a crater? Please distribute the fantasy points immediately.

This is how we arrive at the 2026 Las Vegas Raiders.

AutoDraftYears projects Las Vegas to improve from 14.2 points per game to 21.3, one of the largest jumps in football.

Seven extra points every Sunday sounds exciting.

The projected offense still ranks 25th.

This is the fantasy equivalent of getting your credit score from 410 to 580. Congratulations on the progress. Maybe don't finance the boat yet.

And to understand why the model remains skeptical, we need to remember just how thoroughly the Raiders broke football in 2025.

2025: Somehow, the new offense got worse

There was actual optimism here.

Pete Carroll arrived. Geno Smith followed him from Seattle. Chip Kelly returned to the NFL to run the offense. Las Vegas used the sixth overall pick on Ashton Jeanty and already had one of football's best young tight ends in Brock Bowers.

This was supposed to be the competent version of the Raiders.

Instead, Las Vegas went 3–14 and finished dead last in scoring at 14.2 points per game. It scored 241 points and was outscored by 191.

The Raiders had scored 21.8 points per game in 2021, 23.2 in 2022, 19.5 in 2023 and 18.2 in 2024.

Then came 14.2.

That isn't a slump.

That is falling down the stairs.

The Geno Smith reunion didn't fix quarterback. Smith finished with 3,025 yards, 19 touchdowns and 17 interceptions in 15 games, while taking 55 sacks.

The running game didn't rescue him.

Jeanty handled 266 carries and finished with 975 rushing yards, just 3.7 per attempt. He remained useful through volume and receiving work—55 catches, 346 yards and five receiving touchdowns—but this was not the instant transformation normally envisioned when a team drafts a running back sixth overall.

And Bowers, arguably the best thing about the offense, couldn't stay fully healthy.

He played 12 games, finishing with 64 catches for 680 yards and seven touchdowns while dealing with a knee injury that eventually landed him on injured reserve.

By November, the experiment was already coming apart.

After 11 games under Kelly, Las Vegas ranked 30th in total offense, 31st in scoring and 28th in red-zone percentage. Carroll fired him following a loss to Cleveland and handed play-calling to Greg Olson. It was the first time in Carroll's 14 NFL seasons in Seattle that he'd fired a coordinator during a season.

Then Carroll himself was gone after the season.

So when the model projects a seven-point improvement for 2026, it isn't forecasting some grand offensive awakening.

It is partly forecasting something simpler:

Surely this cannot happen again.

A dangerous sentence in Raiders history, but here we are.

What the model thinks happens next

AutoDraftYears projects:

The snap number matters almost as much as the scoring.

Las Vegas ran only 52.0 offensive plays per game last year. The model expects 59.6.

That's 7.6 additional opportunities every Sunday for Jeanty, Bowers and everyone else.

It doesn't require Klint Kubiak to build the 2007 Patriots.

The offense merely needs to stay on the field long enough for everyone watching at home to remember who has the ball.

Even with that improvement, the positional environments remain ordinary at best:

• QB: 19th
• RB: 29th
• WR: 17th
• TE: 20th

Receiver is the strength.

It ranks 17th.

Very inspiring. Like being the healthiest entrée at Taco Bell.

The historical comps aren't much more aggressive. The 30 most similar offenses from 2016–2025 averaged 2.3 fantasy starters, with a bear/base/bull range of 2/2/3.

That is really the model's Raiders forecast.

Better.

Functional.

Still not especially fertile.

The problem for fantasy managers: everyone already knows

Normally, this is where we'd hunt for a market overreaction.

Bad 2025. New coach. New quarterback situation. Massive projected rebound.

Surely somebody has been unfairly punished.

Nope.

The market ranks Las Vegas 30th in total roster value.

The model ranks it 25th.

Five spots apart, which is well inside the eight the model needs before it will call a roster mispriced.

There is no hidden team-wide discount here. Fantasy managers apparently watched the Raiders too.

And that makes the two expensive players much more interesting.

Ashton Jeanty: great player, ugly neighborhood

Jeanty has an 85/100 projected role and low risk.

That's the good part.

The Raiders' running back environment ranks 29th.

That's the part we tend to wave away while staring lovingly at talent and workload.

The model has Jeanty RB12 versus a market price of RB6, but his adjusted edge of -6 doesn't approach the -13 threshold required for an RB fade.

So I'm not avoiding him.

I am using Las Vegas as the tiebreaker when he's sitting next to another elite back in the first round.

Jeanty's problem isn't Jeanty.

It is everything happening around Jeanty.

Good chef. Questionable restaurant.

Brock Bowers is simpler

Bowers costs roughly pick 20 as TE1.

The model has him TE5.

That sounds bearish until you look at the actual edge: -4, just short of the model's TE fade threshold.

In other words:

He's excellent.

Fantasy managers know.

Please proceed to the register.

The case for Bowers is also helped by something the miserable 2025 season made obvious: when healthy, he is one of the few players this offense can reasonably build around.

The Raiders project to be extremely concentrated, with 100% of their modeled skill-position fantasy production sitting with the top two players. But concentration doesn't create more fantasy points. Across 320 team-seasons in the AutoDraftYears data, concentration had almost no relationship with the number of fantasy starters produced (r = -0.07).

It tells us who gets the pie.

It does not make the pie bigger.

And no, I am not manufacturing a sleeper

The Raiders have no model buy.

They also have no model fade.

Tre Tucker has some late-round appeal. Jalen Nailor exists. Fernando Mendoza is interesting because rookie quarterbacks are contractually required to become fantasy discussions by August.

But these players sit in the thin-market portion of the board.

Calling one a sleeper because we need a sleeper would be analytics cosplay.

Sometimes the spreadsheet looks at a team and says:

The prices are fine. Go bother someone else.

Las Vegas is one of those teams.

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.

The quarterback question can still move this

There is one important wrinkle.

The AutoDraftYears forecast currently assumes Fernando Mendoza takes the quarterback snaps.

As of August 17, the Raiders' unofficial depth chart lists Kirk Cousins first and Mendoza second, although Klint Kubiak has been giving both quarterbacks work during camp.

That matters.

Quarterback changes the scoring projection, red-zone expectation and receiving environment together. If Cousins wins the job—or Mendoza takes it and develops faster than expected—the current team projection deserves another look.

This isn't a small assumption buried three spreadsheets deep.

It's the quarterback.

Those occasionally matter.

What I'm watching in September

Forget passing yards.

Watch red-zone trips.

Las Vegas generated only 2.1 per game in 2025.

The model expects 3.0.

If they're around three through September, I'll believe the offense has made the important jump from dysfunctional to ordinary.

And ordinary would be a major victory.

If they're still hanging around two?

Then perhaps we underestimated the structural integrity of that crater.

So how should we draft the Raiders?

Don't draft the improvement.

Draft the price.

Jeanty is a terrific player whose environment makes me slightly less willing to pay RB6 prices.

Bowers is expensive because he should be expensive.

The late-round options aren't strong enough to force into sleeper status.

And the market and model basically agree about the entire roster.

That's what makes Las Vegas interesting.

This team can improve by seven points per game, generate substantially more offensive volume and become far more competent...

…and still only become the 25th-ranked scoring environment in football.

Fantasy managers love direction because direction feels like upside.

Sometimes it is.

Sometimes you're just watching someone climb out of a very deep hole.

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.