Fade the Field Ep. 13 spent its 38 minutes on four AFC win totals — Denver, Jacksonville,
Pittsburgh and Houston. Anchored to the live Kalshi ladder, the board comes back the opposite way
to most of this series.
On the full 18-team board the model sits
over the market more often than not, and the widest gaps are all overs on teams nobody wants.
Not here. Three of these four price above their projection.
The board
| Team | Contract | Kalshi | Ladder wins | Model wins | Gap |
|---|---|---|---|---|---|
| Jaguars | 10+ wins | 46¢ | 9.0 | 9.5 | +0.48 |
| Steelers | 9+ wins | 44¢ | 8.4 | 8.2 | −0.15 |
| Texans | 11+ wins | 47¢ | 10.0 | 9.2 | −0.76 |
| Broncos | 10+ wins | 56¢ | 9.9 | 9.0 | −0.81 |
*"Ladder wins" is the full KXNFLWINS ladder summed — every rung from "1 or more" to "17." Snapshot
August 14, 2026; the board moves daily and live prices sit on the
The cluster is the story
Every gap here is under one win. On the full board the spread runs to 2.4.
That is not a coincidence, and it is worth internalising before you go hunting. **All four of these
teams sit in the 8-to-10 win band — the part of the ladder where the market is sharpest.** Wide
model-market gaps cluster at the bottom, on teams the market has already written off, because a bad
season is a cheap story to tell and the price absorbs it. Nobody has a cheap story about a nine-win
team. So price and projection converge.
The practical read: this is a poor cluster to hunt for edge in, and a good one to check your model
against. If your projection disagrees violently with the market on a nine-win team, the market is
usually not the one that is wrong.
Denver — the biggest gap, and it is a fade
Model 9.0, ladder 9.9, 10+ trading near 56¢.
The schedule is where the projection loses its optimism. Denver's five hardest projected games are
a Week 17 trip to New England (31% — the single toughest game on any of these four schedules), the
Rams at home (42%), a trip to San Francisco (44%), Seattle at home (46%) and a trip to the Chargers
(47%). That is a brutal collection of coin flips and worse, and only one of them is a get-well game.
The model has Denver banking 2.8 expected wins across the first six weeks, then a Week 10 bye, then
2.6 across the closing five. Neither half is where a ten-win season comes from. The soft spots are
real — the Raiders twice (76% and 63%), Miami at home (69%) — but three wins against the bottom of
the schedule still needs seven from everywhere else.
Broncos win total on Kalshi → · Full Broncos model & schedule →
Houston — priced as a contender, projected as a playoff team
Model 9.2, ladder 10.0, 11+ near 47¢.
The 11+ contract is the tell. At 47¢ the market is close to a coin flip on Houston winning eleven,
and the model does not get there — it projects nine and change. The difference is the closing
stretch: 2.6 expected wins across Weeks 14–18, with road trips to Green Bay (39%) and Philadelphia
(41%) sitting in the last three weeks. Those are the two hardest games on the schedule and they
arrive exactly when a contender needs to be banking wins.
The front half is friendlier — 3.3 expected wins through Week 6, then a Week 8 bye — and the home
slate carries the softest games on the schedule: Tennessee in Week 18 at 72%, the Giants at 65%,
Cincinnati at 64% and Indianapolis at 63%. Houston can absolutely be a ten-win team. Eleven is the
number the market is paying for, and that is the one the model declines.
Texans win total on Kalshi → · Full Texans model & schedule →
Jacksonville — the only over, and a narrow one
Model 9.5, ladder 9.0, 10+ near 46¢.
Half a win of edge is not much, and the honest framing is that this is the cluster's best contract
rather than a good one in isolation. What supports it is the front of the schedule: 3.2 expected
wins through Week 6, a Week 7 bye to reset, and three home games the model likes at 74%, 70% and
69% (Tennessee, Cleveland, Washington).
What argues against sizing it up is where the difficulty sits. Four of Jacksonville's five toughest
projected games fall in the back half — at Houston in Week 15 (45%), at Chicago (46%), at Dallas
(46%), plus a Week 9 trip to Baltimore (47%). A team that banks early and then hits a wall is
exactly the profile that finishes 9-8 and misses the 10+ contract by one game.
Half a win of model edge against that shape is a small position, not a conviction one.
Jaguars win total on Kalshi → · Full Jaguars model & schedule →
Pittsburgh — the closest call on the board
Model 8.2, ladder 8.4, 9+ near 44¢. A gap of 0.15 wins is not a gap. It is agreement.
The interesting part is the shape rather than the number. Pittsburgh's road schedule is where the
season goes to die: the five toughest games are all away — New England (27%, the hardest single game
in this group), Jacksonville (34%), Philadelphia (34%), Baltimore in Week 18 (37%) and Tampa Bay
(43%). At home the model is comfortable, with Cleveland at 62%, Carolina at 61%, Cincinnati at 57%
and Atlanta at 57%.
That is a team whose record will be decided almost entirely by whether it can steal two road games
it is not supposed to win. The model says it banks 2.9 expected wins through Week 6 and only 2.4
across the final five, with that Week 18 trip to Baltimore closing it out.
No position. When the model and a sharp ladder land 0.15 apart, the contract is priced.
Steelers win total on Kalshi → · Full Steelers model & schedule →
How to work a cluster like this
Three rules that fall out of a board this tight:
1. A small gap is information, not an invitation. Four contracts inside one win of their
projection is the market telling you it has priced this tier carefully.
2. Read the shape, not just the total. Denver and Pittsburgh project within a win of each other
in completely different ways — one has a hard middle, the other a hard road slate. The schedule
shape is what decides whether a projection survives contact.
3. Size for August. Every preseason projection in circulation, ours included, runs on last
season's data plus offseason moves. Start small and revisit around Week 4, when real games have
repriced the ladder.
This episode sits inside the show's 2026 win-totals series — 18 of 32 teams on the record before
Week 1, with two independent projections each. The
full board is here, and every episode is
archived on the Fade the Field hub.
Per-game win probabilities and the 17-game schedule for each of these teams live under
NFL team hubs; the rest of the season board sits on NFL futures. To
size any of these properly, the Kelly calculator and
probability converter are free.
Analysis, not advice. Contracts settle after Week 18. Trade responsibly.