Every Desk Prediction From The Year, Graded
Thirteen days out from Indianapolis, the desk had a top-nine. Finals ran that top-nine in the same order. Here is the whole card, line by line, with the two misses called out.
"headline": "Every Desk Prediction From The Year, Graded", "dek": "Thirteen days out from Indianapolis, the desk had a top-nine. Finals ran that top-nine in the same order. Here is the whole card, line by line, with the two misses called out.", "body": "The desk started calling the finish on July 26. Thirteen days before\nIndianapolis, with thirty-six shows in the season already scored and the\nprojection engine on its fourth refit of the summer, it published a full\ntwenty-corps board. Ten thousand simulated Finals nights, refit every scored\nday after that. This is that first board versus what actually happened at\nLucas Oil on August 8.\n\nTHE TOP NINE \u2014 NINE-FOR-NINE\n\n Predicted Actual finish Delta\n 1. Bluecoats 1. Bluecoats 99.1 right\n 2. Blue Devils 2. Blue Devils 97.5 right\n 3. Carolina Crown 3. Carolina Crown 97.0 right\n 4. Boston Crusaders 4. Boston Crusaders right\n 5. Santa Clara 5. Santa Clara 95.0 right\n 6. Blue Stars 6. Blue Stars 94.1 right\n 7. Phantom Regiment 7. Phantom 92.6 right\n 8. The Cavaliers 8. The Cavaliers 92.2 right\n 9. Colts 9. Colts 90.1 right\n\nNine predictions, nine hits, in order, from thirteen days out. Gold was called\nat 74.3 percent, silver at 30.6, bronze at 30.1. Bluecoats crossed the tape\nfirst at 99.1. The desk had it.\n\nTHE BOTTOM THREE \u2014 ELEVEN-OF-TWELVE ON THE CUT-LINE\n\nWhere the board got interesting was the last three seats. The model had\nTroopers at ten, Blue Knights at eleven, Spirit at twelve, with Pacific Crest\none row down at thirteen holding a 51 percent top-twelve shot. Finals ran it\ndifferently.\n\n Predicted Actual Delta\n 10. Troopers 10. Spirit of Atlanta miss (Troopers to 11)\n 11. Blue Knights 11. Troopers correct band, wrong seat\n 12. Spirit of Atlanta 12. Pacific Crest the desk missed the crossover\n 13. Pacific Crest 13. Madison Scouts ...\n 14. Madison Scouts 14. Blue Knights Blue Knights fell out\n\nThe one clean miss: Blue Knights, predicted eleventh, finished fourteenth and\noutside the top twelve. Pacific Crest, whom the model had a shade under the\nline, finished twelfth and inside it. The desk called eleven of the twelve\nFinalists correctly \u2014 a 92 percent hit rate on cut-line placement from thirteen\ndays out.\n\nWHAT THE DESK ALSO SAID OUT LOUD\n\nTwo calls belong in this ledger, because they were made in the newsletter and\nnot just left implicit in the model.\n\n Blue Stars are the most underrated corps in DCI (July 27). Blue Stars\n finished sixth at 94.1, the exact seat the model had held for them from the\n start of the window. The desk's argument was not \"they will jump,\" it was\n \"the fan discourse has them further back than the panels do.\" The panels\n put them sixth. Correct.\n\n The twelve are set, and we think two of these are wrong (August 7,\n Semifinals). One of the two, we can name: Blue Knights, twelfth at Semis\n and out at Finals. The other side of the same call was that a corps sitting\n thirteenth or fourteenth at Semis would cross the line. Pacific Crest\n crossed the line. Correct on both, and by name on one.\n\nWHERE THE MODEL WAS QUIETLY WORKING HARDEST\n\nThe projected medians on the top nine came in tight. Bluecoats projected 99.13,\nfinished 99.1. Blue Devils projected 98.01, finished 97.5. Carolina Crown\nprojected 98.01, finished 97.0 \u2014 the biggest overshoot in the top three, half a\npoint. Boston Crusaders projected 97.38, finished 95.7 \u2014 a 1.7 miss, the worst\nin the top nine, and worth flagging because Boston's late-season trajectory was\nthe one the model consistently rated higher than the panels ended up doing. If\nthe projection engine has a systematic bias to correct in the offseason, that\nis where to look.\n\nTHE HONEST GRADE\n\nNine-for-nine on the podium and the top nine. Eleven of the twelve Finalists\nidentified from thirteen days out. One clean miss inside the top twelve, the\nright direction on the corps that would cross the line, and score projections\nthat ran within about a point of the finishing numbers for eight of nine top\ncorps. On a season the desk has been publishing daily since June, that is a\nprediction record the desk is willing to sign.\n\nThe next grading pass \u2014 every desk call from the summer, graded honestly\nagainst what happened \u2014 runs September 23. That one covers the show-by-show\ncolumn, not just the season-ending board.\n", "sources": [], "gaps": [ "Confirm the exact 'two wrong' corps named in the Aug 7 semis newsletter body once the archive is grepped end-to-end." ], "corps": [], "origination": "original", "written_at": "2026-08-21T20:09:16-04:00", "word_count": 688, "estimated_words": 800, "revision": 1, "content_type": "feature", "calendar_item_id": "seed-e3ded96be3f8624f", "calendar_planned_for": "2026-08-20", "grading_window": { "earliest_prediction_id": 22, "earliest_prediction_at": "2026-07-26T03:01:26", "finals_date": "2026-08-08", "days_out": 13, "actual_source_table": "mc_predictions", "prediction_source_table": "finals_predictions" } }