How to read this. For every corps and every caption we fit the season's trajectory, then asked one question of every show: did this room mark the corps who performed there above or below their own arc? Average that across the night's field and you get the room's signature, with lineup strength and the season's natural growth removed. One honest caveat before any of it: a residual cannot tell a panel apart from the night itself. Rain, wind, a hall's acoustics, a field that ate drill all night move every sheet at once, and no analysis from the outside can fully separate the two. Judging eight minutes of simultaneous excellence in real time is brutally hard work; nothing below is an accusation, and none of it names a judge.

The rooms were smaller than the arguments about them

The typical 2026 room moved a weighted total by about ±0.21 points against trend. That is real money in a sport that settles championships by tenths, but it is far smaller than the season's loudest arguments assumed, and most nights sat closer to zero than to the extremes. The outliers are worth naming. DCI Central Texas (2026-07-16) was the season's warmest room at +0.96 weighted points against trend, with Drum Corps at the Rose Bowl (+0.68) behind it. At the other end, Gold Showcase (2026-07-09) ran -0.90 and Cavalcade of Brass -0.90. A corps that hit one of those rooms mid-season picked up or gave back a chunk of score that had nothing to do with how it performed, and got it back at the next show, which is exactly why single-show score jumps make poor storylines.

Guard and percussion were the wild sheets; GE was the heavy one

Night to night, the most volatile sheets of 2026 were Percussion (±0.20 per room) and Color Guard (±0.19), the two captions where a single reader's taste has the most surface to grip. The steadiest sheet was Brass (±0.13). But the sheet gives General Effect double weight, so once the math is applied the biggest lever in any room was General Effect Visual: a warm GE chair moved a total more than a wild guard chair did.

Eight chairs, mostly eight opinions

The cleanest good news in the data: the panel structure mostly works. Across 44 rooms, the average correlation between two different captions' residuals was 0.08, which is to say close to nothing. Rooms did not run globally hot or cold; chairs disagreed with each other night after night, which is what eight independent sheets are supposed to do. The exception is a visual-book halo: GE Vis and Guard moved together (r = 0.50) more than any other pair. That is not a scandal; when a color guard is having a night, the effect sheet is watching the same show. But it means a guard-forward corps in front of a warm visual room collects the bonus twice.

Who lived and died by the room

Some corps got remarkably consistent reads all season. Others carried a caption the rooms could not agree on. The widest season-long spread belonged to The Academy's Color Guard (±0.57 against their own arc), with Seattle Cascades's Color Guard (±0.56) close behind. For those corps, the honest reading of any single night's caption number started with the question of who was in the box.