The FAAB Market Report

1We said Wright was the better buy. Gordon has the job.

Three weeks ago, after De'Von Achane's injury, our leagues had to pick his replacement. Ollie Gordon went for a median 35% of budget. Jaylen Wright went for 9.5%. We said Gordon had the job, but Wright was the better bet for the price. So far, we have been wrong.

Miami has played two games since then. Week 6 was its bye.

• Gordon: 27 carries for 188 yards and 2 touchdowns. He played 56% of the snaps in Week 4 and 65% in Week 5.
• Wright: 11 carries for 18 yards. He played 31% of the snaps in Week 4 and 14% in Week 5.
• In half-PPR points: Gordon 31.8, Wright 1.8.

Our study counted the market as badly wrong when a cheaper back outscored its favorite by 15 points over three games. Gordon is 30 points ahead of Wright after two. By that same test, our call has missed badly.

What we got wrong. We knew Gordon's 84% came while Wright was out. We read it as an opening, not a promotion, and expected Wright to take back last year's role once healthy. He did not. In the first game they both played after the injury, Gordon took 56% of the snaps to Wright's 31%, and the gap grew the next week.

The case in our history that looked most like Miami pointed the same way. In Arizona in 2023, the favorite took the snaps while the better-drafted back sat out, and the favorite outscored him 20 to 8. We set that case aside for Miami's usage last year. It was the better guide.

What we take from it. When a backup takes over because the other backup is hurt, the first game they both play after it tells you more than last year does. Miami joins the study after its third game, in Week 7.

2The five most contested players

Ranked by how many leagues saw two or more managers bid on the same player. That is a different list from the most-claimed players: half of week-one bids are for nothing, and a free add is not a contested claim. Team defenses are left out here; they are in the full table below.

3Did our number work

Before you bid, the board gives you a price at each of three rungs. P60 is built to win 60% of the players somebody else is also bidding on; P75 and P90 are the same idea at higher confidence.

What we bid, against what it took

Only claims where two or more managers bid. A bid counts as a win when it would have beaten the manager who actually won, by a dollar.

Our P60 bid ran between 11% and 20% of budget across these rows. What players actually took to win ran from 3% to 28%. This table groups players by what they ended up costing, which nobody knows until waivers run. A player lands in the top row because someone bid big, so no bid made in advance wins most of those. It shows where the money went, not whether our number was right. Grouped by what our P60 said instead, it won 52% when it said under 8% of budget and 69% when it said 22% or more, against a 60% promise.

When our P60 did win a contested claim, it overpaid a median 7.0% of budget over what it took. Our target is a max of 5%: a small overpay beats losing the player, but it should not be more than that.

Did we predict which claims would be contested?

Every card leads with one number: the minimum bid when we expect nobody else to bid, or P60 when we expect other managers to bid. This is how often that call was right. It is not part of the grade yet.

All three of our bids

P90 was shown on far fewer players than P60, so it has fewer claims to score. Grey cells are too thin to count toward the grade.

4New method against old

Every card now leads with a player-specific bid. It adjusts your league's price for how widely a player is rostered across the country, how often he is being added, and player sentiment. The previous method, the same price for every player at a position, is still shown beneath it. Both are graded here on the same contested claims, with the same rules.

5We changed how we grade. Every week now uses the same rule.

From this week, a claim counts toward our grade only when the winner paid at least $1 and another team also bid at least $1. A $0 bid never competes, because $1 beats it. Until now we counted those claims, and many of them were easy wins for our bids. Our bids have been priced on this rule since mid-September. Now the grade is too.

We re-graded every past week the same way, and each re-graded page says so. The rule cut both ways. It raised our miss in Weeks 2, 3 and 5, and lowered it in Weeks 4 and 6. Every letter already printed on a week's page held. Week 2, which Week 3's page called a D, is now an F. Weeks 3 and 4 gain key-player grades: a C, and a D.

It also helped us this week. Under the old rule, this week would have been a B, 5.5 points missed. Under the new one it is an A: our bids missed their promise by 3.0 points, our smallest miss yet. Last week, under the same rule, it was 5.6.

The new rule also shows our weakest spot. In every week so far, our cheapest bids fell short. When our P60 said a player would go for under 8% of budget, it won as little as 8% of the time (Week 2) and at best 52% (this week), against a 60% promise.

Key players got a B. Omar Cooper was the only player who qualified, so the grade rests on one player. In 198 leagues, two or more managers bid on him. His price depended a lot on the league. In a quarter of the leagues that won him, the winner paid 10% of budget or less. In another quarter, the winner paid 25% or more. The middle league paid 17%.

Our P60 said 21% and won him 55% of the time, 5 points short of its promise. Of the 181 leagues we could score, the winner paid 25% or more in 62. Our bid won 6 of them.

More players drew a second bid than usual, counting $0 bids: 32% of claims, against 21% in a typical week like this one. Quarterbacks rose the most. 30% of quarterback claims had two or more bids, against 12% in a typical week.

Running backs were cheap again. The typical winning running back bid was 6% of budget, against about 10.7% in a typical week. Receivers cost more than usual: 10%, against about 7%.

6What we are still working on

Fixing our cheapest bids is our next pricing job. This week they won 52% of contested claims, against a 60% promise.

Our high bids still run a little high. When P60 said 22% or more, it won 69%. Across every contested claim it won 64%, on target. When it won, it overpaid a median 7.0% of budget. Our target is a max of 5%.

We are calling fewer claims contested, and that cut both ways. Each player on our board leads with one bid: the minimum when we expect no other bidders, or P60 when we expect some. On claims nobody else bid on, the board led with P60 36% of the time, down from 49% last week. The minimum would have won all 819 of those. But on claims that did draw another bidder, it led with P60 70% of the time, down from 79%. So 30% of real contests led with the minimum and lost, up from 21%. This call is not part of the grade yet, and we are still tuning it.

Last week's report now counts the late private leagues and uses the new rule: 434 leagues, 906 contested claims, still a B (5.6).

The Shipley follow-up comes with the Week 8 report.

7Who got added, and what they cost

Measured across 417 FAABFAX member leagues whose waiver results we have for this run. Private leagues reach us only when a member refreshes from their own browser, usually about a week later. We re-score this page the following Wednesday to include them. Aggregates only. No league, team or manager is named anywhere on this page, and a player appears only when at least 5 leagues bid on them.

Prices are shares of budget, not dollars. Leagues run different budgets. A $50 bid in a $100 league and a $500 bid in a $1,000 league are the same bid, and both show here as 50%. That is the only way to put every league in one table.

Two different prices, and the gap between them matters. "A typical bid" is the middle of every offer made on that player, winners and losers together. "The winning bid" is what the manager who won actually paid, so it sits at the top of that range by definition. When the two are far apart, a few managers wanted that player badly and the rest did not.

"Won by" is the winning bid minus the next-highest bid. It needs the losing bids to exist at all, and it tells you something the winning bid alone cannot: whether the winner had to stretch. A small gap means everyone agreed on the price. A big gap means the winner could have paid less and still won.

8How this week compares to a typical week

Each number is this week against a typical week like it: the same week in 2024 and 2025, from public leagues, weighted to match our mix of superflex and one-QB leagues.

"Two or more bidders" is how often a player drew a real auction instead of being picked up for nothing. When that share rises, people are panicking. "Won by" is how far above the next-best bid the winner landed. A small number means everyone agreed on the price. A big one means the winner could have paid less and still won. "Paid double" is how often the winner bid at least twice the next-highest bid.

9Quarterback prices by league type

In a superflex league a quarterback can also fill the flex spot, so teams need more of them. We keep those leagues apart from one-QB leagues so a quarterback price is never a blend of the two.

The week's most-claimed quarterbacks

10What we learned

More claims were contested than usual. 32% of claims had two or more bidders, against 21% in a typical week like this one (+50%).

Winning bids ran lower than usual: a median of 5.0% of budget against 7.3%.

QB drew more competition than usual: 30% of QB claims had two or more bidders, against 12% in a typical week (+146%). The winning QB bid was a median 4.5% of budget, against 6.0%.

Our P60 bid swung more than prices did. When it said under 8% of budget it won 52% of contested claims, and those players took a median 5.1%. When it said 22% or more it won 69%, and those took a median 20.5%. Too low at one end, too high at the other.

Our card led with the contested price on 819 claims that ended up drawing a single bidder. The minimum bid would have won every one, so following our number there cost a median 14.0% of budget for nothing.