Bears Punditry Check: Everyone Was Wrong About Caleb Williams, and the Field Has Settled on 10 Wins | MissedTakes Daily

The most striking thing in our Chicago Bears file is that the loudest opinions about Caleb Williams lost in both directions.

In September 2025, John Middlekauff said he'd put money on Williams not working out — "worst case scenario, Caleb Williams is a bust". That call graded zero, and its skill score of −194 is about as far below the do-nothing baseline as a single take can fall: to count as a bust he had to finish outside the top nine quarterbacks, and his best season put him fifth. Going the other way cost just as much. Chris Brockman predicted a top-five MVP finish, Rich Eisen granted it, and Seth Rollins simply said "Caleb for MVP." Matthew Stafford won; all three scored zero.

That is the pattern across the quarterback's whole file. We've tracked 61 takes about Williams, 12 of them settled, and they average 54.6 accuracy — the worst of any position group on this roster that has a real sample behind it.

The field keeps landing a couple of wins short

Twelve published win-total calls on Chicago for 2025 averaged exactly nine wins. The Bears won 11. Brady Quinn was lowest at five; Andrew Santino had 11 back in May and drew the best score of the group (100 accuracy, +161 skill — meaning he beat the naive guess by more than the width of the whole call). Eisen's in-season 10 landed at 80.

This year the field has clustered hard. Of the 22 Bears win-total calls we've collected for 2026, nine land on exactly ten wins — Nick Wright, Dave Dameshek, Patrick Claybon, Conor Orr, Moe Moton, John Breech and Rollins among them — with Colin Cowherd and Emory Hunt taking the over on 9.5. The disagreement lives at the edges: Adam Schein has them at 12 or 13 and winning the division, while Veejay Huskey is alone in calling them a fallback team at 7-10.

Worth knowing before you weight those: on season win totals generally, Wright's record is 43.63 accuracy over 50 graded calls with a score of −10.42 — below the level of just repeating the obvious. His mock drafts (82.26 over 95 calls) are a different animal entirely.

Who has actually been right about this team

Picking the Bears has paid unusually well in our grading. Playoff-berth calls on Chicago sit at 84.6 accuracy with an average skill score of +127, and the five graded division calls went 5-for-5. Straight game picks are the soft spot: 50 graded, 66 accuracy, and a skill score of +4.08 — essentially no better than taking the home team.

Mike Moraitis is perfect across ten graded Bears calls, mostly roster and depth work. Brockman's accuracy is middling but his skill number (+87.3) is the best on the club, because he's been right about the unobvious. Pat Fitzmaurice is the only regular below the baseline, and on five graded calls that's a lean rather than a record.

The roster-call number (94.1 accuracy over 17) flatters the genre. Most of those calls were "this returning player makes the 53," which is what happens by default, so they earned nothing. Three did beat it: Moraitis on Jordan van den Berg twice and on rookie safety Dillon Thieneman. The lone miss came from Pete Martuneac, who argued undrafted rookie Skyler Thomas would be released before Week 1 despite making the initial roster. Thomas stayed.

What's on the board now

Week 1 opinion was near-unanimous: Gene Chamberlain has Chicago 34-30 at Carolina, Richard Sherman 24-17, and Danny Parkins, Jonas Knox and Kirk Herbstreit all took the Bears laying three. Kay Adams has gone furthest — Bears–Broncos in the Super Bowl.

The receiver room is where to watch. Luther Burden III has been named a breakout again this week, by Jacob Infante and by a Fox Sports column quoting an NFC scout who thinks Burden passes Rome Odunze as the No. 1 target. That exact call has already failed once: Burden's graded takes average 27.8 accuracy, dragged down by a sleeper call he answered by finishing WR49, below the startable tier. Chamberlain, meanwhile, expects the Bears to extend Williams this offseason. None of this season's calls have an answer yet.

The league day in brief

• We graded 1,233 takes over the past week, 1,165 of them roster-spot calls as clubs settled their 53s, plus 52 game picks and 15 score predictions. Another 452 new takes came in.
• The week's perfect scores were all Week 1 game picks, and the best of them took nerve: Middlekauff's 49ers-over-Rams pick in Melbourne and Tyler Sullivan's slate of five Week 1 underdogs each scored 100, and because they went against the chalk, each was worth double a routine favorite pick.
• The zeros ran the other way: projected 49ers and Lions rosters and depth charts, including Tom Jensen's 53-man projection and Grant Cohn's call on how San Francisco would deploy its defensive line in the opener.
• Lance Zierlein still leads at 39.1 on 38 graded calls, with Derek Brown and David Harrison behind him. Nathan Jahnke has the highest accuracy in the top ten at 86.72, and Daniel Jeremiah is the only one there with a deep enough body of work (278 graded) to call the number settled.

Full standings on the leaderboard.