Fantasy football reporting for the leagues we play in. Findings first, personas second.
16-team dynasty with a salary cap, contracts, IDP and coach slots. Six seasons measured; the deepest tooling here.
We joined this league last, into a room where a good share of the owners already knew each other from the commissioner’s other league. Then we went 67-16 and became the villains. Six seasons of his campaign, measured — 33.3% of all Scranton trade talk from one man who is 18.5% of the chat — and what it has actually cost us: 8.3 offers to close a trade, against 4.6 in 2020.
This piece first ran with a $107M surplus lead and a flat null that killed the owner’s thesis. Both were artefacts. A seam in the 2025 data credits contracts with 8,665 player-weeks nobody rostered, and on the clean window West Waco lead at $262.5M with us second. The null was mis-specified: hold the baseline and surplus → elite payroll runs +0.331, not −0.005. But the acquisition side is still −0.015 — the flywheel turns on retention, never on going out and landing someone new.
The commissioner writes 18.5% of this league’s group chat — and 33.3% of everything said about Scranton trades, against 8.3% for Las Vegas and 5.0% for Chicago. He is not louder when we trade; he is louder about us, for four years running. The phrase in the headline is his own.
Scranton tagged Odell Beckham Jr at $5.55M in August 2021. The commissioner said he would have paid $14.5M. Within eight days the league was debating a bylaws change. What Beckham went on to do is beside the point: the complaint was about the gap between price and perceived value on the day, and a complaint of that kind is settled at the moment of the transaction, not three seasons later.
In the league’s first season Scranton picked up free agents nobody wanted and sold them for late-round picks — twelve times, sometimes inside a week. Richard Rodgers was claimed on a Thursday and gone by the following Wednesday for a 2021 fifth, eighteen minutes after the advert went up. The trick worked because it was never about the player, and it has since almost entirely stopped.
A quarter of regular-season weeks are played with at least one idle starter, and it costs 15.58 points. That part is structural. Who pays it is not: exposure runs 5.9× from West Waco’s 0.084 idle starters a week to Oeiras’ 0.494, worth roughly seventy points of scoring a season between the ends — and the two most successful franchises in the league sit thirteenth and sixteenth on the list. Bye weeks are published in April.
‘May the best team win’ is a blessing, not a prediction — put a question mark on it and it becomes checkable. Scranton have gone 67–16 and West Waco 62–21 across six regular seasons. Between them: twelve playoff appearances and two titles — the same as two franchises that went a combined 66–100 and qualified once apiece. It is not choking, and there is no fantasy defense — nothing about them gets worse in January. But ten of their fourteen games were against franchises that put 41.7 fewer points on the board. Against teams that actually reach the bracket the league’s best have won 68.2%, not 78.4%, and that compounds to 31.7% over three rounds. Corrected 10 Aug: an earlier version claimed the edge “collapses”, which implied a defense that does not exist.
MFL prints every manager’s efficiency for free — the share of his perfect-hindsight ceiling he actually scored — and has never printed the same number for its own projections. Same rosters, same weeks, same ceiling: the owners realised 82.00% of it and the projections would have realised 81.86%. Six seasons, twenty owner careers, and not one of them separable from the chalk. Corrected 10 Aug: the coverage gate pruned 179 weeks unevenly and removed the bad ones — on the 806 weeks where both sides saw the same men the null is stronger still, p = 0.9093.
Six seasons of realized contract value put Scranton first at +$371.4M. Split against a price nobody in this league set, it comes apart into 44% price discipline, 58% selection and −2% position mix — and the bigger leg is a forecast error by construction, only the smaller one becomes cap space, and the first three seasons do not predict the last three.
Obeying MFL’s projection every week is worth −0.276 points a franchise-week. Rebuild the standings with one manager doing it and the answer depends entirely on which coverage gate you pick — six defensible ones run from −15 to +12 Victory Points. Rebuild them with all sixteen, the only version of the question worth asking, and every gate gives exactly zero, because zero is the only answer the scoring rule permits. What does move is 111 Victory Points changing hands and four of six playoff fields.
The team that acquires the most does get worse the following year — r = −0.246 across 80 franchise-season pairs. Hold this season's record fixed and it goes to −0.041. And the confound runs backwards from the warning: here the acquisitive team is the losing team, so the raw signal is losers persisting, not winners regressing.
Six seasons of in-season pickups, graded on whether the starts were any good rather than how often they happened. At QB, RB and WR a street pickup delivers more actively-bad starts than good ones.
Matched against lots the outside market ranked the same, a contested lot returns about eleven more season points — a quarter of the ordinary spread between two equally ranked players, at 4.58x the price. Matched on money instead of rank, the edge vanishes entirely.
Re-starting last week’s lineup every week would cost 20 points a week — and obeying the projection would be worth nothing at all. Both are true. The gap is dead slots, not better picks: the stale lineup starts 3.48 players a week who score zero against the manager’s 1.58.
The IDP heuristic says start the underdog’s defenders, because a losing defence is on the field longer. The field time is real — 1.75 fewer snaps and 0.106 fewer solo tackles per 12 points of spread the other way. It is also worth less than the sacks the favourite’s pass rushers collect, and the whole edge dies at corner and safety.