# The 2026 Mock Draft Cycle, Graded: The Field Nailed the Top, Missed the Fallers, and Never Got Sharper

*Published 2026-08-15 — MissedTakes*

The 2026 draft is graded. Our database holds **1,413 mock-draft pick calls** settled. One row per "team X takes player Y at pick Z." And mock drafts are now the single largest category of graded takes on the site, with 2,872 graded calls across all cycles.

Two things stand out. The field was excellent at the very top and useless about the players who fell. And publishing closer to the draft did not reliably make anyone more accurate.

## What actually happened at the top

The Raiders took Indiana quarterback **Fernando Mendoza** first overall. The Jets followed with Texas Tech edge David Bailey, the Cardinals took Notre Dame running back Jeremiyah Love third, and Ohio State supplied three of the first seven picks (Carnell Tate 4, Arvell Reese 5, Sonny Styles 7).

That is more or less the draft the industry described. Grouped by player, the field's calls on the top of the board score like this:

![2026 first-rounders: how the field's calls graded](/posts/visuals/97d34780-2f5f-46dd-83b9-3bd29c14112d.png)

Mendoza's 120 graded calls average **99.1** accuracy — as close to a settled question as this database has. [Colin Cowherd](/pundits/colin-cowherd) had him at No. 1 on November 7th, 2025, months before the consensus consolidated, and was graded 100. Sonny Styles (98 calls, 92.9), David Bailey (60 calls, 92.5), Caleb Downs (77 calls, 89.2), Jeremiyah Love (82, 87.8) and Carnell Tate (65, 87.8) were all effectively pre-solved.

## The misses were about who fell, not who rose

Site-wide, our [mock-draft case study](/nfl) shows accuracy collapses as you go deeper: an average of **76.6** on players who actually went in round 1, 48.9 in round 2, 39.5 in round 3, and 26.9 for picks 176 and later. Average miss distance goes from about 10 picks in round 1 to nearly 49 late on day three.

The 2026 class produced textbook examples. Tennessee cornerback **Jermod McCoy** was mocked 21 times at an average slot of 16 and went 101st; his 28 graded calls average **2.7** accuracy. Toledo safety **Emmanuel McNeil-Warren** was mocked around 21 and went 58 (24 calls, 31.5). Georgia linebacker **C.J. Allen** was mocked around 18 and went 53 (16 calls, 32.9). Clemson cornerback Avieon Terrell (25 calls, 39.0) went 48. Guard Emmanuel Pregnon was mocked around 35 and went 88.

The purest case is Ohio State defensive tackle **Cade McDonald**, who appears 15 times in the database — placed in the first round by, among others, [Daniel Jeremiah](/pundits/daniel-jeremiah) (31st, March 17), [Gregg Rosenthal](/pundits/gregg-rosenthal) (28th, April 22), [Mike Renner](/pundits/mike-renner) (26th, April 20) and [Bucky Brooks](/pundits/bucky-brooks) (25th, February 5). He [went undrafted](/nfl/players/cade-mcdonald) and later signed with Cleveland. Every one of those 15 calls scored **0**.

Quarterbacks other than Mendoza were the other sinkhole. Drew Allar (5 graded calls) averages 34.3 after going 76th; Cade Klubnik (3 calls) averages 4.9 after going 110th; Garrett Nussmeier (5 calls) averages **0.1**. Not great. He went 249th, and Kyle Crabbs had him 16th.

## Who graded out best on mock picks

The site's published [mock-draft leaderboard](/leaderboard) pools every graded draft cycle, not just 2026, so read it as a career board rather than a 2026 scorecard:

| Pundit | MissedTakes Score | Accuracy | Graded mock calls |
|---|---|---|---|
| [Chris Brockman](/pundits/chris-brockman) | 50.00 | 75.02 | 36 |
| [John Middlekauff](/pundits/john-middlekauff) | 50.00 | 76.12 | 19 |
| [Rich Eisen](/pundits/rich-eisen) | 46.67 | 83.71 | 58 |
| [David Harrison](/pundits/david-harrison) | 43.32 | 65.62 | 120 |
| [Mike Renner](/pundits/mike-renner) | 39.47 | 73.56 | 98 |
| [Ross Jackson](/pundits/ross-jackson) | 38.67 | 61.69 | 142 |
| [Daniel Jeremiah](/pundits/daniel-jeremiah) | 31.02 | 64.86 | 226 |

All of these carry "limited" confidence except Jeremiah, whose 226 graded calls make him the only established sample in the category.

Volume matters for interpretation. Raw accuracy in this category mostly measures **how deep into the draft you were willing to go**. [Ross Jackson](/pundits/ross-jackson) (52.3 across 142 graded calls, all cycles) mocks day-two and day-three picks. He had Malachi Lawrence 148th, and Lawrence went 23rd, while a pundit who only makes 32 first-round calls is playing an easier game. Kyle Crabbs has the lowest average of any high-volume mocker (38.7 across 322 graded calls), and the reason is timing as much as depth: his only 2026-cycle mock was published a year out.

At the outlet level, across all cycles: [The Rich Eisen Show](/outlets) grades 80.8 on 233 calls, Move the Sticks 69.1 on 350, Around the NFL 69.2 on 232, and The 33rd Team 39.4 on 349. Again, the low number belongs to the shop that mocks the deepest.

## Did the calls get better as the draft neared?

This is the least-flattering part of the whole thing: Grouping every mock-draft call by the month it was published. When we do that:

![Mock-draft accuracy by month published](/posts/visuals/46a12b1a-ea3a-4315-b104-5c4c864410cc.png)

Accuracy jumped from January to February, then went **sideways** through the combine, through free agency, and through draft week. April 2026, with 916 graded calls, was no more accurate (67.6) than February (71.4).

Two individual mocks bracket the range. Kyle Crabbs published a 2026 mock on **April 29th (of 2025)**, the weekend after the previous draft, a full year out: his 28 graded picks average **36.5**, with seven of them scoring under 5 (Nussmeier 16th to 249th, Dani Dennis-Sutton 20th to 120th, and L.T. Overton 25th to 137th). Gregg Rosenthal's full first-round mock on April 22nd, days before the draft, averages **77.7** across his 30 graded picks that had an actual pick to compare against. Plus, one more, Cade McDonald at 28, who went undrafted and scored 0.

So a year of work is worth roughly forty points of accuracy at the individual-mock level. But almost all of that improvement had already happened by February. The January figure (59.8 on 153 calls) is also depressed by composition rather than ignorance. That month is heavy with Ross Jackson's deep mocks, where a 40-pick miss is routine.

A caveat: our monthly series pools everyone's calls, and the mix of players and rounds changes month to month. It is not a controlled experiment. But the specific cases point the same direction as the aggregate. The field's error on Jermod McCoy narrowed from an average mock slot of 16 to Rosenthal's 31st two days out — and he still went 101st. Mel Kiper had McCoy 12th on March 21 and C.J. Allen 20th in the same mock. Nick Wright put Cade McDonald 26th on **April 23**. The late information didn't contain the answer these pundits were looking for.

## Teams

Aggregated by team, mock-draft calls involving the Raiders grade 85.3 (116 graded, bolstered by Fernando Mendoza) and the Titans 80.2 (99), while the Saints sit at 49.9 across 185 calls and the Commanders 55.6 across 122. We'd treat that table gently: those team numbers pool multiple draft cycles, and the two lowest belong to teams with dedicated Locked On shows that mock every round, not just the first. Team-level accuracy here is a measure of coverage depth more than of any front office's unpredictability.

## The short version

- The consensus was right about the top of the 2026 draft and got graded accordingly: Mendoza at 99.1 across 120 calls.
- It was wrong in one direction: players it liked who slid. McCoy (2.7), McNeil-Warren (31.5), C.J. Allen (32.9), and 15 straight zeros on an undrafted Cade McDonald.
- Getting closer to the draft bought the field almost nothing after February.