Four outlets graded the same 2026 draft class and did not come close to agreeing. Washington is the best class in the league for two of them and 21st for another. This is where the grades split, where they converge, and what our own roster archive says when we check the claims underneath them.

Team by team: NFL team reports.

1,132snaps Bobby Wagner played for Washington last season before leaving the league, the vacancy at the centre of the most divisive class in the draft2025 season including playoffs. Roster snapshot 2026-08-05, from our own archive.

Which team had the best 2026 NFL draft class?

It depends entirely on who you ask, and the spread is not narrow. Cody Carpentier of RosterWatch gave the Washington Commanders a 95 out of 100, the joint-highest mark he handed out, writing that they threw darts when they were on the clock and hit specific body types that fit their team. Ian Cummings of Pro Football Network went further and made Washington the best class in the league at A+, with Sonny Styles the top player on his board, taken seventh.

Gennaro Filice of NFL.com ranked the same class 21st and gave it a B-.

That is not a rounding difference. It is the same six players read two opposite ways, and the disagreement is not really about talent. It is about how much a team should like an empty room.

Here is the fact all three are looking at, from our own roster archive. Bobby Wagner played 1,132 snaps in the middle of Washington's defence in 2025 and does not appear on any roster in 2026. Washington signed no veteran over the top of the position. The linebackers who came back have 364 recorded snaps between them.

Carpentier and Cummings look at that and see a team that identified its single biggest hole and filled it with the best player available. Filice looks at the same emptiness and sees no margin for error. His words: he is worried the Commanders need Styles to hit the ground running on Day 1, and that is a lot to ask of a rookie.

Mike Clay of ESPN, who projects the season rather than grading the draft, sides with the optimists in the only currency that matters. He has Styles leading every rookie in the league in tackles, at 123.

We checked the other Washington claim, and it survives by one yard

Filice also flagged Antonio Williams, Washington's third-round receiver, and justified it by saying no receiver on the roster besides Terry McLaurin cleared 350 receiving yards last season. That is the kind of claim our archive can settle, so we settled it.

Receiver on the 2026 rosterPlayed 2025 forReceiving yardsTargets
Terry McLaurinCommanders58260
Van JeffersonTitans35052
Dyami BrownJaguars22737
Jaylin LaneCommanders22532
Luke McCaffreyCommanders20315
Treylon BurksCommanders13022

2025 season including playoffs, 2025-08-01 to 2026-02-08. Receivers listed at WR on the 2026-08-05 roster snapshot.

He is right, and the margin is one yard. Van Jefferson, who arrived from Tennessee, finished 2025 on exactly 350. Nobody eclipsed it.

What the claim leaves out is worse for Washington than what it says. The receiver who actually led the Commanders last season was Deebo Samuel, with 727 yards on 99 targets, and he is not on the roster either. Neither is Zach Ertz, who caught 72 targets at tight end. Washington did not merely fail to add around McLaurin. It lost its most-targeted pass catcher and its starting tight end, and then asked a rookie drafted 71st to help replace them.

Where the graders actually agree

Two answers, one at each end, and they are close to unanimous.

Cleveland is everybody's winner. Kiper gave the Browns an A. Filice ranked them first of 32 and also gave an A. Dane Brugler put them in his top three. Cummings was the low man at B+. Filice's favourite Cleveland move was not the first-rounder at all but safety Emmanuel McNeil-Warren in the second round at 58, which he called highway robbery on the value front.

Our archive explains the enthusiasm without needing a scouting opinion. Cleveland drafted Spencer Fano ninth into a tackle room that lost Cam Robinson (755 snaps), Jack Conklin (382 snaps) and Cornelius Lucas (292 snaps), none of whom appear on a 2026 roster. Kiper called Fano one of the top two tackles in the class and said the trade up for him was justified.

Jacksonville is everybody's loser. Carpentier gave the Jaguars a 63, a D-, the lowest mark in his table. Brugler ranked them last. Cummings had them at D. The reason is structural rather than a matter of taste: Jacksonville's first-round choice belonged to Cleveland, so the class opened in the second round with tight end Nate Boerkircher. San Francisco is the other agreed-upon floor, at D from Carpentier and D+ from Cummings.

The rookie three separate outlets called a steal

C.J. Allen, linebacker, Indianapolis, drafted 53rd. He turns up in three of the six pieces we read, in three different roles, and each one arrives at him from a different direction.

Kiper had him 28th on his own board and called the selection a steal, citing three-down attributes. Filice put him on a list of rookies who simply have to succeed in year one, quoting general manager Chris Ballard describing Allen as a green-dot guy from the get-go, which is to say the man who wears the helmet radio and runs the defence. James Dator of Yahoo Sports named him a Rookie of the Year candidate from outside the first round and noted he is already being written into the Colts' starting lineup.

Our archive supports the opening. Zaire Franklin played 1,112 snaps for Indianapolis in 2025 and is on the Packers roster in 2026. The linebackers Indianapolis kept are Akeem Davis-Gaither (781 snaps), Austin Ajiake (157 snaps) and Jaylon Carlies (2 snaps). Allen is not competing with an incumbent so much as replacing one.

Where the projections and the depth charts collide

Grades are opinions about April. Projections are claims about September, and two of them are currently arguing with the coaching staffs.

Fernando Mendoza. Clay projects the first overall choice for 3,052 passing yards and 14 touchdowns, and the important number in that line is the one that is not a statistic: 14 starts. Las Vegas head coach Klint Kubiak has already said that when the team goes out to practise, Kirk Cousins is the starting quarterback, that he is the guy, and that he deserves it. Cousins, for his part, said he intends to hold Mendoza off as long as he can. Clay's projection does not contradict them so much as assume the argument gets settled around the fourth week of the season.

Eli Stowers. Dator projects 700 to 800 receiving yards for the Eagles' second-round tight end in two-tight-end sets. The Liberty Line, covering camp on 2 August, reported him still buried on the depth chart behind Dallas Goedert and Grant Calcaterra. Our roster file has both of those names back, and six returning tight ends in the room in total. One of those two things has to move.

Makai Lemon. Filice's favourite Philadelphia move, drafted 20th, and the clearest example of a vacancy that is not the same thing as an opportunity. A.J. Brown took 1,028 yards on 128 targets to New England. Philadelphia then signed Marquise Brown and Dontayvion Wicks, and on 10 August Brandon Lee Gowton of Bleeding Green Nation wrote that Lemon had missed half the spring and half of training camp, had not done much on the days he did practise, and that it is fair to doubt he starts early in the season.

One claim we could not check

Filice justified Chicago drafting safety Dillon Thieneman 25th by noting the Bears lost 19 of their 23 interceptions to departures. It is a good line and we cannot confirm it.

Our archive stores interceptions thrown, not interceptions caught. Asking it for Chicago's 2025 interceptions returns exactly one player, quarterback Caleb Williams, at 12. There is no defensive interception data in our football archive at all, so that claim stays attributed to NFL.com and unverified by us.

What we can confirm is the shape of it. Chicago's secondary lost Jaquan Brisker (1,221 snaps, now with the Steelers), Kevin Byard (1,220 snaps, now with the Patriots) and Nahshon Wright (1,177 snaps, now with the Jets). The Bears then drafted Thieneman 25th and Malik Muhammad 124th into that room, and signed Coby Bryant out of Seattle. Filice's number may well be right. It is simply not ours to stand behind.

What to actually take from a draft grade

Filice said the quiet part himself: draft grades are overwhelmingly influenced by the premium choices. A class is mostly judged on three or four players, which is why the same six-player Washington class can be first and 21st in the same week. The grades worth reading are the ones that tell you what the grader is assuming, and on Washington the assumption is stated plainly on both sides. One camp assumes a clean vacancy is an advantage. The other assumes it is exposure. Both are looking at a linebacker room with 364 snaps in it.

How this was measured

Frequently asked questions

Who had the best 2026 NFL draft class?

There is no consensus. RosterWatch and Pro Football Network both made Washington the best class in the league, while NFL.com ranked the same class 21st. Cleveland is the closest thing to an agreed winner, graded A by both Mel Kiper and NFL.com and placed in the top three by Dane Brugler.

Who had the worst 2026 draft class?

Jacksonville, near-unanimously. RosterWatch gave the Jaguars the lowest mark in its table, Brugler ranked them last and Pro Football Network gave them a D. Their first-round choice belonged to Cleveland. San Francisco is the other agreed-upon bottom class.

Which 2026 rookie was the biggest steal?

C.J. Allen, the Indianapolis linebacker drafted 53rd, is the only player named by three separate outlets. Mel Kiper had him 28th on his board. Zaire Franklin, who played 1,112 snaps at the position in 2025, is now in Green Bay.

Which rookies are projected to lead the class statistically?

Mike Clay of ESPN projects Fernando Mendoza to lead rookies in passing yards, Carnell Tate in receiving yards, Jeremiyah Love in rushing yards, Sonny Styles in tackles and David Bailey in sacks. Those are ESPN's projections, not ours.

Related: NFL win totals and futures explained · NFL team reports

Every figure on this page is a historical measurement of what already happened, produced by a committed, re-runnable script over our own odds archive. Past frequencies are not forecasts and are not betting advice: Nebula Insights publishes no picks, no selections and no recommended wagers. You must be of legal gambling age in your jurisdiction.