| # | Player | Pos | Team | Proj ▾ | Projected stats |
|---|
Source Disagreement
For each player, the spread (max − min) between projection sources. Bigger bars = sources disagree more. Useful for spotting draft values where one source is bullish vs. consensus.
Position Tiers — Cliffs & Dropoffs
Projected PPR for each ranked player at the position. Watch for the cliffs — large vertical drops between consecutive players are tier breaks.
Position Distributions
Box plot of projected PPR for each position. The taller the box, the wider the spread of viable starters.
Tier Map by Position
Every projected player as a dot, colored by auto-detected tier (consecutive PPR drops larger than 1.0 standard deviation mark tier breaks). Hover any dot for player + projection.
Season-long market lines, FanDuel vs Kalshi.
FanDuel posts a single over/under line; Kalshi posts a probability
ladder, from which we interpolate the 50% strike — the closest
equivalent to a book's line. The last column is how far Kalshi sits
from FanDuel: green means Kalshi is higher, red means lower. Both price
the same events, so a large gap means they genuinely disagree.
A * marks a thin Kalshi market (fewer than two tight-spread strikes) — treat those loosely. FanDuel posts no receptions or receiving TD markets, so those rows are Kalshi-only and have no comparison.
A * marks a thin Kalshi market (fewer than two tight-spread strikes) — treat those loosely. FanDuel posts no receptions or receiving TD markets, so those rows are Kalshi-only and have no comparison.
| Player | Pos | Stat | FanDuel | Odds | Kalshi | Kalshi vs FD ▾ |
|---|
Fantasy points implied by the betting markets.
Each player's market lines converted to PPR / half-PPR / standard using
the same scoring as the Rankings tab. Lines come from FanDuel where
posted, then Bovada, otherwise Kalshi's ladder. Bovada is the only
book here quoting receiving TDs; Kalshi is the only source
for receptions.
A player is scored only when every stat their position needs is available. Anyone missing one shows NaN — the markets simply don't price that stat for them, and a partial sum would understate them badly. Values marked ~ are modelled from a Kalshi ladder rather than posted: either a lognormal fit (ladders that never cross 50%) or an assumed-sigma fit (rungs tied at one probability). Hover any stat for its source.
Measured against the projection consensus, the plain fitted values miss by ~34% at the median and collapse toward zero on thin books — Puka Nacua priced at 1.8 receiving TDs, Michael Penix at 3.8 passing yards. They are replaced with projections by default; untick to see the raw fits. Assumed-sigma is far better (~15%) and is the only thing giving most receivers a receptions line, so it is kept unless you ask otherwise.
UD ADP is the current Underdog board — click the column to sort by draft position and read the market's points against where a player is actually going. Undrafted players show —.
No book anywhere posts a season receptions market, and books skip receiving props entirely for most committee RBs — so those players can never score on market data alone. Tick Fill unpriced stats with projections to substitute the FantasyPros/Clay consensus for just those gaps. Filled cells are dashed and the total is flagged P; hover either to see which stats came from projections rather than a market.
A player is scored only when every stat their position needs is available. Anyone missing one shows NaN — the markets simply don't price that stat for them, and a partial sum would understate them badly. Values marked ~ are modelled from a Kalshi ladder rather than posted: either a lognormal fit (ladders that never cross 50%) or an assumed-sigma fit (rungs tied at one probability). Hover any stat for its source.
Measured against the projection consensus, the plain fitted values miss by ~34% at the median and collapse toward zero on thin books — Puka Nacua priced at 1.8 receiving TDs, Michael Penix at 3.8 passing yards. They are replaced with projections by default; untick to see the raw fits. Assumed-sigma is far better (~15%) and is the only thing giving most receivers a receptions line, so it is kept unless you ask otherwise.
UD ADP is the current Underdog board — click the column to sort by draft position and read the market's points against where a player is actually going. Undrafted players show —.
No book anywhere posts a season receptions market, and books skip receiving props entirely for most committee RBs — so those players can never score on market data alone. Tick Fill unpriced stats with projections to substitute the FantasyPros/Clay consensus for just those gaps. Filled cells are dashed and the total is flagged P; hover either to see which stats came from projections rather than a market.
| # | Player | Pos | UD ADP | Proj ▾ | Market-implied stats |
|---|
Fantasy points implied by Week 1’s game lines.
Sportsbook consensus (DraftKings, FanDuel, Bovada, BetRivers,
BetOnline via The Odds API) merged with Kalshi's per-game ladders,
converted with the same scoring as the other tabs. Where both price a
stat the book consensus wins — five books beat one venue —
and hovering a chip lists every book by name with its own number and
price. A chip with a dotted underline is one the books disagree on.
Kalshi still supplies touchdowns, which none of those books post as a
per-game count. Unlike the season-long board, these ladders are deep
(9–10 strikes is typical, versus 1–4 for season markets),
so nearly every line is a real interpolated 50% crossing rather than a
model.
Kalshi deletes settled markets, so this is always the upcoming slate — a week already played cannot be fetched back. The week number is resolved from ESPN's schedule rather than guessed from the date, since the season opens midweek. Touchdowns come from Kalshi's anytime-TD ladder as an expected count (the sum of P(X ≥ k)), with DraftKings' anytime-TD price filling the 198 players Kalshi does not cover — 429 priced against Kalshi's 236. Kalshi's ladder is preferred where it exists because DK quotes P(scores 1+), which reads ~0.3 low for goal-line backs who can score twice. Neither book posts a split rushing- vs receiving-TD game market, so that one number carries all non-passing scoring.
A player is ranked on whatever stats are priced for them; unlike Market Points there is no completeness gate, because a per-game line set is inherently partial (nobody posts rushing TDs for a WR). Treat cross-position totals loosely for that reason — QBs rank high partly because passing yards are always priced.
Kalshi deletes settled markets, so this is always the upcoming slate — a week already played cannot be fetched back. The week number is resolved from ESPN's schedule rather than guessed from the date, since the season opens midweek. Touchdowns come from Kalshi's anytime-TD ladder as an expected count (the sum of P(X ≥ k)), with DraftKings' anytime-TD price filling the 198 players Kalshi does not cover — 429 priced against Kalshi's 236. Kalshi's ladder is preferred where it exists because DK quotes P(scores 1+), which reads ~0.3 low for goal-line backs who can score twice. Neither book posts a split rushing- vs receiving-TD game market, so that one number carries all non-passing scoring.
A player is ranked on whatever stats are priced for them; unlike Market Points there is no completeness gate, because a per-game line set is inherently partial (nobody posts rushing TDs for a WR). Treat cross-position totals loosely for that reason — QBs rank high partly because passing yards are always priced.
| # | Player | Pos | Game | Proj ▾ | Market-implied stats |
|---|
Market lines averaged across the first four weeks.
One week of props says as much about the opponent as about the player
— a soft matchup in week 1 and a shutdown corner in week 2 pull
the same player in opposite directions. Averaging four weeks gives a
market-implied baseline: what the books expect in a typical
game.
Each week is first collapsed to a consensus across the books quoting it, then the weeks are averaged equally — so a week priced by five books does not outweigh one priced by two. Hover any stat to see the week-by-week lines behind the average.
Only The Odds API publishes future weeks — Kalshi lists the upcoming slate only, so touchdowns (which no book posts as a per-game count) are absent here and these totals are therefore lower than the Week 1 tab’s.
Each week is first collapsed to a consensus across the books quoting it, then the weeks are averaged equally — so a week priced by five books does not outweigh one priced by two. Hover any stat to see the week-by-week lines behind the average.
Only The Odds API publishes future weeks — Kalshi lists the upcoming slate only, so touchdowns (which no book posts as a per-game count) are absent here and these totals are therefore lower than the Week 1 tab’s.
| # | Player | Pos | Wks | Avg Proj ▾ | Averaged market lines |
|---|
Paste your roster, get the optimal lineup.
One player per line — names are matched loosely, so
“ceedee lamb” or “D. Lamb” both work. Every
player is scored on this week’s market lines, then the best
legal lineup is chosen for
QB / RB / RB / WR / WR / TE / FLEX / FLEX in
half-PPR. FLEX takes RB, WR or TE.
Price from these books re-derives every line from only the sportsbooks you tick — score off DraftKings alone, or drop one you think is soft.
Narrowing to a single book drops any stat that book does not post, so totals fall: Drake Maye reads 9.7 on the full consensus but 2.8 on DraftKings alone, because DK prices his rushing yards and not his passing. That is the filter working, not a missing line — use it to ask “what does this book think”, and leave all books ticked for the best projection. Useful when you distrust one number — a touchdown price is far noisier than a receptions line — or to ask a narrower question, like who starts on receiving volume alone. Excluded lines still show, struck through.
A player whose game has kicked off is locked: his actual points are shown, his slot is marked spent, and the optimizer fills only what is still changeable. The total splits into what is banked and what is still to play.
Every row shows the market lines behind its projection — hover any chip to see each book by name with its own number, and a dotted underline marks the ones the books disagree on.
Two players alone works too: add both and the tab tells you which to start and by how much — or use the H2H tab for a stat-by-stat breakdown. Anything the market hasn’t priced is listed separately rather than silently scored as zero — an unpriced player is a market signal, not a projection of nothing.
Price from these books re-derives every line from only the sportsbooks you tick — score off DraftKings alone, or drop one you think is soft.
Narrowing to a single book drops any stat that book does not post, so totals fall: Drake Maye reads 9.7 on the full consensus but 2.8 on DraftKings alone, because DK prices his rushing yards and not his passing. That is the filter working, not a missing line — use it to ask “what does this book think”, and leave all books ticked for the best projection. Useful when you distrust one number — a touchdown price is far noisier than a receptions line — or to ask a narrower question, like who starts on receiving volume alone. Excluded lines still show, struck through.
A player whose game has kicked off is locked: his actual points are shown, his slot is marked spent, and the optimizer fills only what is still changeable. The total splits into what is banked and what is still to play.
Every row shows the market lines behind its projection — hover any chip to see each book by name with its own number, and a dotted underline marks the ones the books disagree on.
Two players alone works too: add both and the tab tells you which to start and by how much — or use the H2H tab for a stat-by-stat breakdown. Anything the market hasn’t priced is listed separately rather than silently scored as zero — an unpriced player is a market signal, not a projection of nothing.
Paste a roster and hit Optimize.
Two players, one call.
Pick any two and see the market’s case for each — not just
the projected total, but which stat the edge actually comes from.
Half-PPR, this week’s lines. Start typing to search.
A gap under half a point is inside the noise and the tab says so rather than manufacturing confidence.
Any stat the books do not agree on gets its own row listing every book's line — green above consensus, red below. A projection six books agree on is a firmer number than one they spread ten yards across, and only the book-level view shows that.
A gap under half a point is inside the noise and the tab says so rather than manufacturing confidence.
Any stat the books do not agree on gets its own row listing every book's line — green above consensus, red below. A projection six books agree on is a firmer number than one they spread ten yards across, and only the book-level view shows that.
vs
Pick two players to compare.
Enter your Sleeper username and get this week’s
start/sit calls.
Your leagues, rosters and scoring are read live from Sleeper —
no password, nothing stored on a server. Every rostered player is
scored on this week’s betting lines, then the best legal lineup
is built for each league’s own slots and scoring.
The useful part is the difference from what you already have set: players the optimizer would start who are currently benched are marked SWAP IN, and starters it would bench are marked SITTING. A player whose game has kicked off is locked: his actual points are shown, his slot is marked spent, and the optimizer fills only what is still changeable. The total splits into what is banked and what is still to play.
Every row shows the market lines behind its projection — hover any chip for each book by name with its own number. Kickers and defenses are listed but not projected here — no book prices them as players. The lineup page scores them off the game line instead. Players whose game has already kicked off are listed separately: the books stop quoting them, so the lineup above covers only who is left to play.
Best available reads every roster in the league, so anyone left over is a free agent. Those are ranked by the upgrade they would give you, not by raw points — an unrostered QB tops any points list while being worthless to a team that already starts one.
The useful part is the difference from what you already have set: players the optimizer would start who are currently benched are marked SWAP IN, and starters it would bench are marked SITTING. A player whose game has kicked off is locked: his actual points are shown, his slot is marked spent, and the optimizer fills only what is still changeable. The total splits into what is banked and what is still to play.
Every row shows the market lines behind its projection — hover any chip for each book by name with its own number. Kickers and defenses are listed but not projected here — no book prices them as players. The lineup page scores them off the game line instead. Players whose game has already kicked off are listed separately: the books stop quoting them, so the lineup above covers only who is left to play.
Best available reads every roster in the league, so anyone left over is a free agent. Those are ranked by the upgrade they would give you, not by raw points — an unrostered QB tops any points list while being worthless to a team that already starts one.
Price from these books
all
Enter a Sleeper username to load your leagues.