https://nflanalytic.com/explainers.html New NFL Analytics explainers and season pieces: Elo ratings, weekly win probabilities and their grading, standings and playoff odds, and how football's stats are built. en-us Wed, 14 Oct 2026 00:00:00 GMT https://nflanalytic.com/explainer-epa-rankings-after-week-5-2026.html https://nflanalytic.com/explainer-epa-rankings-after-week-5-2026.html Wed, 14 Oct 2026 00:00:00 GMT NFL EPA per play rankings for all 32 offenses and defenses after week 5 of 2026, with success rates, and what a five-game EPA ranking has meant since 2002. https://nflanalytic.com/explainer-seahawks-broncos-thursday-epa-2026.html https://nflanalytic.com/explainer-seahawks-broncos-thursday-epa-2026.html Wed, 14 Oct 2026 00:00:00 GMT Seahawks at Broncos on Thursday of week 6 by EPA per play, success rate and CPOE: Seattle's 2nd-ranked defense, two offenses in the bottom ten, and the history. https://nflanalytic.com/explainer-week-6-playoff-picture-2026.html https://nflanalytic.com/explainer-week-6-playoff-picture-2026.html Tue, 13 Oct 2026 00:00:00 GMT The NFL playoff picture after week 5: AFC and NFC playoff odds for all 32 teams, and what each week 6 game can move, led by Chicago at Atlanta at 71.2 points. https://nflanalytic.com/explainer-standings-after-week-5-2026.html https://nflanalytic.com/explainer-standings-after-week-5-2026.html Tue, 13 Oct 2026 00:00:00 GMT AFC and NFC standings after week 5 of 2026 beside points scored and allowed. Since 2002, teams a win ahead of their scoring after five games played to the scoring. https://nflanalytic.com/explainer-week-5-playoff-swings-2026.html https://nflanalytic.com/explainer-week-5-playoff-swings-2026.html Mon, 12 Oct 2026 00:00:00 GMT Week 5's Sunday moved playoff odds 208 points; Chicago fell from 65.9% to 37.3%. Since 2002, week-5 falls have held and big rises have given some back. https://nflanalytic.com/explainer-week-5-sunday-graded-2026.html https://nflanalytic.com/explainer-week-5-sunday-graded-2026.html Mon, 12 Oct 2026 00:00:00 GMT Week 5's Sunday: the ledger went 8 of 11 with a tie and beat the close. Season Brier .2173 to .2243; since 2010 such leads after week 5 lasted 2 times in 4. https://nflanalytic.com/explainer-new-coaches-four-games-2026.html https://nflanalytic.com/explainer-new-coaches-four-games-2026.html Sun, 11 Oct 2026 00:00:00 GMT Four of the seven new head coaches of 2026 are 3-1, a .047 chance in August. Since 2010 fast new-coach starts kept going; since 2000 the effect halves. https://nflanalytic.com/explainer-early-bye-2026.html https://nflanalytic.com/explainer-early-bye-2026.html Sun, 11 Oct 2026 00:00:00 GMT Kansas City and Carolina now play 13 straight weeks. Since 2002 long runs without a bye have cost nothing measurable; the first game back is worth .045. https://nflanalytic.com/explainer-1-3-favourites-2026.html https://nflanalytic.com/explainer-1-3-favourites-2026.html Sat, 10 Oct 2026 00:00:00 GMT Washington and the Jets are 1-3 and favoured over 3-1 teams. Since 1999, favourites with the worse record won .625, near the line's .608; the engine said .474. https://nflanalytic.com/explainer-nfc-west-strength-2026.html https://nflanalytic.com/explainer-nfc-west-strength-2026.html Sat, 10 Oct 2026 00:00:00 GMT The NFC West has the best top three of any division at week 5 since 2002. The simulation gives it three playoff teams in 54% of seasons, and that number holds up. https://nflanalytic.com/explainer-cross-country-trips-2026.html https://nflanalytic.com/explainer-cross-country-trips-2026.html Fri, 09 Oct 2026 00:00:00 GMT Since 1999, three-hour trips have cost NFL visitors nothing the closing line or a flat home edge missed. Week 5 has three: Las Vegas, Detroit and Buffalo. https://nflanalytic.com/explainer-tb-dal-graded-2026.html https://nflanalytic.com/explainer-tb-dal-graded-2026.html Fri, 09 Oct 2026 00:00:00 GMT Tampa Bay won 24-16 at Dallas, a miss for the ledger's .6405. The close had Dallas at .8082, so every later market price scored worse than the frozen number. https://nflanalytic.com/explainer-week-5-ledger-vs-market-2026.html https://nflanalytic.com/explainer-week-5-ledger-vs-market-2026.html Thu, 08 Oct 2026 00:00:00 GMT Week 5's ledger and DraftKings agree on 13 of 15 winners. Since 2010, in 7 of 8 kinds of disagreement, the results sat nearer the market's number. https://nflanalytic.com/explainer-tb-dal-thursday-2026.html https://nflanalytic.com/explainer-tb-dal-thursday-2026.html Thu, 08 Oct 2026 00:00:00 GMT Tampa Bay is 0-4 by 20 points. The ledger has Dallas at .6405, DraftKings at .7914; since 2010, when the market was this much surer, it was the better price. https://nflanalytic.com/explainer-remaining-schedule-after-week-4-2026.html https://nflanalytic.com/explainer-remaining-schedule-after-week-4-2026.html Wed, 07 Oct 2026 00:00:00 GMT Chicago has the hardest remaining NFL schedule after week 4, Cincinnati the easiest. Since 2010, ratings have forecast remaining opponents better than records. https://nflanalytic.com/explainer-chi-gb-lambeau-2026.html https://nflanalytic.com/explainer-chi-gb-lambeau-2026.html Wed, 07 Oct 2026 00:00:00 GMT Chicago is favoured at Lambeau for the first time since 2006. Across 980 division games, a long run with the host favoured has not made the new price wrong. https://nflanalytic.com/explainer-hou-ten-winless-2026.html https://nflanalytic.com/explainer-hou-ten-winless-2026.html Tue, 06 Oct 2026 00:00:00 GMT Houston and Tennessee are both 0-4 and 217.7 rating points apart, the widest gap in a meeting of winless teams since 1999. None of 81 0-4 starts made the playoffs. https://nflanalytic.com/explainer-phi-jax-london-2026.html https://nflanalytic.com/explainer-phi-jax-london-2026.html Tue, 06 Oct 2026 00:00:00 GMT Jacksonville's London games have closed with 2.32 points of home field since 2013, more than an ordinary home game's 1.87. Twelve results cannot confirm the price. https://nflanalytic.com/explainer-board-after-four-weeks-2026.html https://nflanalytic.com/explainer-board-after-four-weeks-2026.html Mon, 05 Oct 2026 00:00:00 GMT Houston is 0-4 and 14th. In 512 team-seasons the board's moves were right-sized; what it over-trusts is August. Discounted, Houston's last 13 project to 6.50 wins. https://nflanalytic.com/explainer-september-graded-2026.html https://nflanalytic.com/explainer-september-graded-2026.html Mon, 05 Oct 2026 00:00:00 GMT The August board expected 2.44 teams at 0-4 by October 5. Five arrived, a 4.8% outcome, with Houston priced at 1.1%. On all 32 records it still beat 2-2. https://nflanalytic.com/explainer-atl-no-boom-or-bust-2026.html https://nflanalytic.com/explainer-atl-no-boom-or-bust-2026.html Sun, 04 Oct 2026 00:00:00 GMT Atlanta's start is 2026's most erratic against the spread, but in 861 team-seasons that scatter predicts nothing (r .013). Monday's .5403 for New Orleans stands. https://nflanalytic.com/explainer-week-3-graded-2026.html https://nflanalytic.com/explainer-week-3-graded-2026.html Sun, 04 Oct 2026 00:00:00 GMT Week 3: the ledger went 11 of 16 and beat the closing market, .2259 to .2541. It now leads for the season; since 2010 such leads lasted the season twice in seven. https://nflanalytic.com/explainer-denver-home-field-2026.html https://nflanalytic.com/explainer-denver-home-field-2026.html Fri, 25 Sep 2026 00:00:00 GMT Denver's home edge ranks 8th of 32 since 1999, inside the noise. Sunday's pick over the Rams is almost all home field; the market's Rams -2.5 is about the teams. https://nflanalytic.com/explainer-atl-gb-graded-2026.html https://nflanalytic.com/explainer-atl-gb-graded-2026.html Fri, 25 Sep 2026 00:00:00 GMT Atlanta won 35-14 at Green Bay, a .7139 miss. The game moved both teams 41.16 points and Atlanta's playoff odds from 8.0% to 22.8%, most of it from the re-rating. https://nflanalytic.com/explainer-order-of-a-1-1-start-2026.html https://nflanalytic.com/explainer-order-of-a-1-1-start-2026.html Thu, 24 Sep 2026 00:00:00 GMT Won then lost, or lost then won: 401 NFL 1-1 starts since 1999 finish the season level. Game three leans to the latest winner, at under two standard errors. https://nflanalytic.com/explainer-what-0-3-means-2026.html https://nflanalytic.com/explainer-what-0-3-means-2026.html Thu, 24 Sep 2026 00:00:00 GMT 129 NFL teams have started 0-3 since 1999; two made the playoffs, both Houston. What game three is worth to an 0-2 team, and 2026's expected 0-3 count. https://nflanalytic.com/explainer-how-low-is-tennessee-2026.html https://nflanalytic.com/explainer-how-low-is-tennessee-2026.html Wed, 23 Sep 2026 00:00:00 GMT Tennessee is 1314.6 after two weeks, last by 65 points. Five of 512 team-seasons since 2010 were lower; what became of every week-2 bottom, measured. https://nflanalytic.com/explainer-week-3-leverage-board-2026.html https://nflanalytic.com/explainer-week-3-leverage-board-2026.html Wed, 23 Sep 2026 00:00:00 GMT Each week-3 result priced both ways in a 20,000-season simulation: the Rams at Denver moves 68.4 points of playoff odds; interconference games split exactly in half. https://nflanalytic.com/explainer-change-its-mind.html https://nflanalytic.com/explainer-change-its-mind.html Tue, 22 Sep 2026 00:00:00 GMT New England at Jacksonville is 6.7 rating points from flipping; Arizona at San Francisco is 238.9. What a week of football can actually move, measured. https://nflanalytic.com/explainer-what-two-weeks-bought.html https://nflanalytic.com/explainer-what-two-weeks-bought.html Tue, 22 Sep 2026 00:00:00 GMT The week-3 fixtures priced on August ratings and on today's: probabilities moved 4.91 points on average and not one of the sixteen picks changed side. https://nflanalytic.com/explainer-hindsight-variable.html https://nflanalytic.com/explainer-hindsight-variable.html Mon, 21 Sep 2026 00:00:00 GMT Low-scoring NFL games are harder for this model - but only sorted after the fact. By pregame total line the effect vanishes, and that distinction is the finding. https://nflanalytic.com/explainer-honest-not-sharp.html https://nflanalytic.com/explainer-honest-not-sharp.html Mon, 21 Sep 2026 00:00:00 GMT The model went 8 of 14 on Sunday and its Brier rose to .2364. Decomposed, the calibration is intact - and the real finding is how little resolution it ever had. https://nflanalytic.com/explainer-margin-multiplier.html https://nflanalytic.com/explainer-margin-multiplier.html Sun, 20 Sep 2026 00:00:00 GMT This model's margin multiplier is its most valuable part, but its exact curve is not: once K can rescale, log, square root and linear margin all score the same. https://nflanalytic.com/explainer-engine-constants.html https://nflanalytic.com/explainer-engine-constants.html Sat, 19 Sep 2026 00:00:00 GMT Three of this model's four constants have never been tested. K = 20 is already optimal, the home-field edge is stale, and tuning them forward gains nothing. https://nflanalytic.com/explainer-three-fixes-priced-together.html https://nflanalytic.com/explainer-three-fixes-priced-together.html Sat, 19 Sep 2026 00:00:00 GMT Three defects this site measured separately, priced together on 608 team-seasons out of sample. They do not add up - the first correction does nearly all of it. https://nflanalytic.com/explainer-division-rematch.html https://nflanalytic.com/explainer-division-rematch.html Fri, 18 Sep 2026 00:00:00 GMT Every within-season NFL rematch is a division home-and-home. The same team won both in 768 of 1,340 pairs - more often than the two prices multiply out to. https://nflanalytic.com/explainer-det-buf-graded-2026.html https://nflanalytic.com/explainer-det-buf-graded-2026.html Fri, 18 Sep 2026 00:00:00 GMT Buffalo beat Detroit 41-31 and the ledger's .6612 pick hit. The margin landed on the exact rung Tuesday's preview published, so the branch grades cell for cell. https://nflanalytic.com/explainer-qb-change-blind-spot.html https://nflanalytic.com/explainer-qb-change-blind-spot.html Thu, 17 Sep 2026 00:00:00 GMT Across 656 games since 2010 a team starting a new quarterback won 36.5% of the time. The model gave it 45.8%, the closing market 38.9%. What the gap is worth. https://nflanalytic.com/explainer-game-pick-calibration-2026.html https://nflanalytic.com/explainer-game-pick-calibration-2026.html Thu, 17 Sep 2026 00:00:00 GMT The site's playoff odds needed a calibration repair. Its game picks, tested the same way across 4,363 graded games, turn out to barely need one at all. https://nflanalytic.com/explainer-det-buf-thursday-2026.html https://nflanalytic.com/explainer-det-buf-thursday-2026.html Wed, 16 Sep 2026 00:00:00 GMT Detroit at Buffalo opens week 2 with both teams 1-0. What each result does to the NFC North and AFC East races, priced before kickoff from the ledger's .6612. https://nflanalytic.com/explainer-calibrated-playoff-odds-2026.html https://nflanalytic.com/explainer-calibrated-playoff-odds-2026.html Wed, 16 Sep 2026 00:00:00 GMT Yesterday's page found this site's playoff odds too confident. One parameter fixes the levels out of sample, and costs the move-sizing that was already right. https://nflanalytic.com/explainer-playoff-odds-after-week-1-2026.html https://nflanalytic.com/explainer-playoff-odds-after-week-1-2026.html Tue, 15 Sep 2026 00:00:00 GMT Week 1 moved all 32 playoff odds by 376 points, an ordinary amount. Since 2002 those moves were the right size, but the odds themselves were too confident. https://nflanalytic.com/explainer-den-kc-graded-2026.html https://nflanalytic.com/explainer-den-kc-graded-2026.html Tue, 15 Sep 2026 00:00:00 GMT Kansas City 31, Denver 10: the model's .5805 on Denver missed, week 1 ends 10 of 16 on budget, and 41 points tie 2012's week-1 scoring record exactly. https://nflanalytic.com/explainer-0-2-starts.html https://nflanalytic.com/explainer-0-2-starts.html Mon, 14 Sep 2026 00:00:00 GMT 0-2 teams made the playoffs 11% of the time since 1999. At the same opening rating each early loss still costs 21 points of playoff odds. Week 2 of 2026, priced. https://nflanalytic.com/explainer-week-1-scoring-2026.html https://nflanalytic.com/explainer-week-1-scoring-2026.html Mon, 14 Sep 2026 00:00:00 GMT 2026's week 1 scored 50.0 points a game. Across 27 seasons a loud week 1 kept about a sixth of its surprise, and the totals moved by the same sixth. https://nflanalytic.com/explainer-sunday-graded-2026.html https://nflanalytic.com/explainer-sunday-graded-2026.html Mon, 14 Sep 2026 00:00:00 GMT The model went 9 of 13 on Sunday, exactly its break-even. Graded against Friday's pre-registered numbers: Brier, the market's grade, the splits and the top ten. https://nflanalytic.com/explainer-week-2-overreaction.html https://nflanalytic.com/explainer-week-2-overreaction.html Sun, 13 Sep 2026 00:00:00 GMT A 20-point week-1 surprise moves the week-2 line about 1.3 points. Across 422 games since 1999 that was the right size, and fading the blowout did not pay. https://nflanalytic.com/explainer-model-has-no-good-years.html https://nflanalytic.com/explainer-model-has-no-good-years.html Sat, 12 Sep 2026 00:00:00 GMT Across 16 graded seasons a week-1 record said nothing about the model's year: r .13, and seasons scatter less than coin flips. What 1-1 or a bad Sunday is worth. https://nflanalytic.com/explainer-sf-la-season-repriced.html https://nflanalytic.com/explainer-sf-la-season-repriced.html Sat, 12 Sep 2026 00:00:00 GMT SF 27, Rams 7 moved about 25 points of playoff odds each way and 1.3 wins of season. Since 2002, week-1 moves like it have been about the right size. https://nflanalytic.com/explainer-what-sunday-can-do.html https://nflanalytic.com/explainer-what-sunday-can-do.html Fri, 11 Sep 2026 00:00:00 GMT Sunday's 13 picks as one distribution: 8.08 expected correct, a losing Sunday 17.9%, what each count does to the miss budget and top ten, where the market splits.