# Browns 27, Steelers 24: One Exact Margin Among Five Wrong Winner Picks

*Published 2026-10-02 — MissedTakes*

Cleveland beat Pittsburgh 27-24 on Thursday night, and the pundits who called it beforehand leaned the wrong way. We tracked eight graded winner picks on this game — five took the Steelers, three took the Browns. Cleveland winning means the majority missed, 5-to-3.

That split matters more than it sounds like it should. The Browns were the naive favorite simply by virtue of hosting the game — our baseline model, which does no more than lean slightly toward the home team, had Cleveland by well under a point. So taking Cleveland here wasn't a bold read; it was closer to the default. Taking Pittsburgh was the upset call, and it's the call that lost.

### The calls worth naming

Three of the eight picked Cleveland to win, and the differences among them tell the real story.

[John Breech](/pundits/john-breech) [picked the Browns 19-16](/opinions/2beae2d5-778a-41da-81d0-87aa79b4fc6d), good for a predicted margin of three points. The final margin was three points. That's about as clean as a score prediction gets, and it graded accordingly — 100 accuracy and a skill score of +9.07, the only positive skill mark anyone posted on this game.

[Pete Prisco](/pundits/pete-prisco) also took Cleveland, [picking it 21-20](/opinions/acf8df5f-b69a-498e-b31b-d190f627916b) — right winner, margin of one. That graded to 95.24 accuracy, but the skill number tells a different story: +0.36, barely above the baseline. Getting a near-default outcome almost exactly right still only pays off a little, because the baseline already expected something close to it.

[Mike Kadlick](/pundits/mike-kadlick) also had the winner right, writing, "I'm taking the Browns to win this one by two scores, [24–10, to move to 3–1 on the season](/opinions/0bbf96eb-4e0b-442d-ace5-dcea31e4782a)." The winner call was correct; the margin wasn't close — Cleveland won by three, not fourteen. That gap between calling the right team and overselling the blowout cost him: 73.81 accuracy but a skill score of -38.86, worse than doing nothing at all.

### The misses

Five calls took Pittsburgh, and all five lost on the winner, which is the costliest way to be wrong in our scoring since it ignores a baseline that was already pointing elsewhere. [Tyler Sullivan](/pundits/tyler-sullivan), [Jordan Dajani](/pundits/jordan-dajani), and [Walt](/pundits/walt) each landed within the same band — 47.62 accuracy, -86.79 skill — having all picked the Steelers by four points (21-17, 21-17, and 20-16 respectively). [Jared Dubin](/pundits/jared-dubin) picked [Steelers 20-14](/opinions/8d6bc2df-9fe0-4167-90e2-374b10b791e8), a six-point margin, and graded slightly worse at 42.86 accuracy and -95.5 skill.

The roughest night belonged to the Nightcap broadcast, where a co-host identified only as "Deebo" skipped the score entirely and just called the outcome: "We're going 1,000% to the land... And then I'm going to take your chain, too, after we whoop on y'all." An outright winner pick against the board's own favorite, with no hedge, scored 0 accuracy and -200 skill — the floor.

One more thread settled by this game, though it wasn't a winner pick: [Carter Bahns](/pundits/carter-bahns) had [called it "reality" back in August](/opinions/18802de8-941e-4551-b6e3-df7a752020b6) that Shedeur Sanders would be Cleveland's starting quarterback by Week 4, with Deshaun Watson "set to fail." Watson started Thursday; Sanders didn't play a snap. That call graded to zero as well.

### What this does to the record books

[Breech's](/pundits/john-breech) exact-margin call is the outlier in his own body of work, not the norm — his score predictions sit at 47.15 accuracy for his career, with a -19 average skill score, and his outright winner picks have been worse still (31.55 accuracy). One clean call doesn't erase that, but it's the best thing in his file this week by a wide margin.

[Prisco](/pundits/pete-prisco), by contrast, picked the right side again inside a season where his score predictions already run positive on average (65.85 accuracy, +6.66 skill) — a solid, unspectacular continuation rather than a correction.

[Dubin](/pundits/jared-dubin) has been a tick below average on score calls all year (48.21 accuracy, -17.88 skill), and this one — his worst skill mark of the five Pittsburgh backers — fits that pattern rather than breaking it.

### Beyond this game

A loss like this doesn't just settle one Thursday night. Our tracked season-win-total calls on the Browns have graded well so far (90% accuracy, average skill +35, though only four calls have settled), while the Steelers' playoff-berth calls have gone 100% accuracy across 29 graded picks with a modest +13.79 average skill — a record built on games like this one going Pittsburgh's way, which this one didn't. For the fuller picture on both clubs, see the [Browns](/nfl/teams/browns) and [Steelers](/nfl/teams/steelers) team pages, and see how this week's callers stack up overall on the [leaderboard](/leaderboard).