The first trade offers of the season land a few hours after the first box score. They are almost always priced on that box score.
You do not have to guess whether an offer is fair. Dynasty is a market, and the market leaves records. This post is a reading guide: what real trades look like at scale, how lopsided a normal deal is, what a package really costs, how your format changes the price, and what one week can and cannot move.
Every number below comes from the SharperSunday trade corpus and the Edge value model, with the date it was measured. The sources are listed at the end.
Start with the market, not the group chat
Your league is twelve opinions. The corpus behind Edge is 1,513,193 Sleeper leagues and 8,544,114 trades as of September 7, and it grows every four hours.
The Edge model retrains every night on the deduplicated, format-tagged subset of that corpus, which was 7,853,765 trades in the table published September 8. Every value it produces is priced for a specific format: team count, superflex or 1QB, PPR, tight end premium and passing touchdown points.
That matters in Week 1 because your leaguemates are pricing on narrative and the market is pricing on what people actually paid. When the two disagree, the market is usually the better anchor.
How lopsided is “normal”?
Managers argue about fairness as if every deal should be dead even. Real trades are not.
Take clean two-team dynasty trades, both sides worth a meaningful amount, and measure the gap between the two sides as a share of the larger side. Among trades with a gap under 60% (beyond that it is a fleecing, not price discovery), the distribution looks like this in the table fit on September 8:
| Share of clean two-team trades | Gap between the sides is under |
|---|---|
| 70% | 30% |
| 90% | 46% |
That is 334,841 trades in a twelve-month window, superflex PPR, measured against current Edge values.
The read for Week 1: a 15% gap is inside the range the market treats as ordinary. A 40% gap is in the top fifth of lopsided deals. If someone tells you a deal with a 10% gap is a robbery, the market disagrees.
Packages cost more than their parts
The best player wins the two-for-one. Everyone knows that. The question is how much extra the package side has to carry for the market to call it fair.
We measured it in July on 303,693 clean consolidation trades, one player on one side and two or three on the other:
| Package shape | Premium the market pays |
|---|---|
| 1-for-2 | 12.1% |
| 1-for-3 | 22.9% |
| 1-for-4 or more | 38.8% |
The premium scales steeply with how many pieces you give up. Two throw-ins cost you about an eighth over the stud’s value. Four cost you nearly forty percent.
It also depends on your league’s shape, and every difference below is well outside its error bar:
| League factor | Higher premium | Lower premium |
|---|---|---|
| Roster format | 1QB (17.4%) | Superflex (14.1%) |
| League size | 10-team (18.3%) | 14-team or larger (12.1%) |
| Starting lineup | Shallow, 8 or fewer starters (16.6%) | Deep, 11 or more (13.9%) |
The mechanism is replaceability. In a 10-team league the waiver wire is deep, so the depth you shed is easy to replace and the stud commands more. In a 14-team league every startable body is scarce, so consolidation costs less.
One honest caveat: the overall level moves with how far back you look. A six-month window reads higher, around 19 to 26%, and a twelve-month window reads 13 to 16%. The nightly fit on September 8, twelve months, superflex PPR, 143,181 trades, landed at 12.5%. The relative structure, more pieces and smaller leagues meaning a bigger premium, holds in every window.
Your format is the price
A player does not have one value. He has a value in your league.
The clearest case is quarterback. Take the twelve most valuable quarterbacks in a 12-team superflex PPR league and reprice them for the same league as 1QB. In the September 8 table, the median quarterback loses 41% of his value, with the range running from 32% to 45%.
Tight end premium runs the other way. The twelve most valuable tight ends gain a median 12% at half a point per reception and 25% at a full point, in the same 12-team superflex PPR setting.
So a Week 1 offer that quotes a “consensus” quarterback price at you in a 1QB league is quoting a superflex number. The market pays roughly three-fifths of it. If you are new to either format, superflex and TE premium are short reads.
Picks are priced on a curve
Dynasty picks are the currency of September. They are also the asset most managers price by feel.
The Edge pick table prices picks by round and by how many years out the draft is. Indexing each round to its own value in the draft year makes the curve easy to see:
| Pick | Draft year | 1 year out | 2 years out | 3 years out |
|---|---|---|---|---|
| 1st round | 100 | 54 | 37 | 26 |
| 2nd round | 100 | 71 | 54 | 19 |
| 3rd round | 100 | 75 | 61 | 24 |
Two things fall out of that table. A first loses roughly 46% of its draft-year value one year out and about a third more the year after. And seconds hold their value better one year out than firsts do, at 71 cents on the dollar against 54.
Right now, in September 2026, the 2027 class is “1 year out.” A 2027 first is worth about half of what a first in its draft year is priced at. That discount is why in-season pick buying is a strategy at all, and why a contender offering you a 2028 first for a starter is paying in the cheapest currency on the board.
Treat the index as market pricing at one moment, not a forecast. It says what a pick is worth relative to the same pick in its draft year today; it does not promise your 2027 first will grow into that number.
What one week actually moves
Here is the part most Week 1 offers get wrong.
Edge values come from trades, and one game does not produce many trades. The model retrains nightly, and the trainer refuses to publish a table where any top-200 player moved more than 20% overnight. In the six days before kickoff, zero of the top 200 players moved that much. The board was stable going into Week 1.
That does not mean values will not move. It means they move when managers actually pay a new price, not when a player has one big or bad Sunday. A Week 1 offer that prices a player as if his value already jumped is asking you to pay for a move the market has not made.
The practical rule: if the argument for the offer is a box score, it is priced on narrative. If the argument is “here is what he went for in three leagues this week,” it is priced on the market.
Reading your own league in five minutes
You can check a Week 1 offer against the market without a spreadsheet.
- Open the Dynasty Trade Database. It shows up to 500 real trades from the past seven days, tagged by format, with Edge values on every asset. Look for the same shape of deal you were offered.
- Run the offer through the Dynasty Trade Calculator with your league’s settings, not the default ones. The gap it reports is the one to compare against the 30% and 46% lines above.
- Check the other manager’s win window. A contender and a rebuilder should not price the same pick the same way, and the offer usually tells you which one you are talking to.
Then ask the co-manager. When you connect your Sleeper league, it already knows your roster, your format, every team’s win window, and where each asset sits on the Edge board. Paste the offer and ask what the market says about it for your league, not a generic one.
Start free and check your first offer
Data note: corpus and league counts are the /api/corpusCount bake of 2026-09-07T23:21Z as recorded in docs/audit/DATA_INTEGRITY_2026-09-08.md. The trainer subset (7,853,765 trades), the fair-gap bands (70th percentile 30.5%, 90th percentile 46.1%, n = 334,841), the 12.5% nightly consolidation premium (n = 143,181), the quarterback and tight end format sensitivities, and the pick index were read from the published Edge table values/value_tables.json (schema 160-continuous-v1, generated 2026-09-08T04:52Z) using the same feature-vector formula the app serves; positions for the format calculation come from the Sleeper player database. The consolidation premium by package size and league shape is the July 2026 Band Calibration Study (docs/BAND_CALIBRATION_STUDY_2026-07.md, run 2026-07-10 on 7,620,882 trades, 303,693 clean fit rows) and its sensitivity grid. The pre-kickoff stability figure (0 of the top 200 players moved more than 20% between the 2026-09-01 and 2026-09-07 tables) and the trainer’s drift guard are from the same data-integrity audit. All numbers are observations of market behavior, not guarantees.