Mind the gap
Two numbers, not one
Every player in your league has two numbers, and almost every fight in your group chat comes from pretending there's only one.
The first number is the price. It's what your leaguemate will actually accept for Breece Hall, and it moves on highlights, podcasts, and whoever got drafted behind him in April. The second number is the value. It's what he'll actually score for you over the next three seasons, and it moves on touches, efficiency, and age. The price is set by people. The value is set by production.
Here's the part most analytics tools get wrong: the price is not noise. Your trades clear at the price, not at anyone's projection. A tool that tells you the market is dumb loses you trades. What you actually need is both numbers, measured honestly, side by side.
So I built both.
The price: VibeRank
VibeRank is the market price, measured live. It's a public ranking board built entirely from head-to-head votes: you get three players, you tap them best to worst, the board updates. Under the hood it's a rating system that carries uncertainty (a player with 4 votes is treated as less settled than one with 400), seeded by the engine so a thin board degrades to math instead of garbage, and weighted by rater trust so one motivated voter can't drag a player up a board by himself. I tested that last part personally. The math holds: conviction costs breadth, never repetition.
Eight boards cover the formats that matter (dynasty and redraft, superflex and 1QB, TE premium on or off), with 900+ players priced on each. There's a daily challenge, streaks, a public leaderboard, and 7-day movers so you can watch the market reprice a player in real time.
It's free. No account required to vote. It stays that way.
The value: the engine
The other number comes from the engine, and the engine doesn't aggregate anyone's opinion. It computes projections from production data, tuned to your league's exact scoring and roster, and recalibrates every season against what actually happened.
That last clause is the one I'd push on if I were you, because everyone claims it. So the receipts are public: the Accuracy page scores old boards against realized outcomes, per position, per year, with nothing curated. The most recent scored year, the 2024 board graded against what actually happened in 2025, put the WR board at a 0.795 rank correlation with reality across 128 wide receivers, with a mean error under 2 points per game. That's the strongest year in the engine's ten-year backtest, and the trend has been pointing that way since 2021 because the model recalibrates after every season. And when the engine misses, the page names names. Its own latest miss list includes Justin Jefferson. I left that up on purpose. A tool that only shows you its hits is selling you something other than accuracy.
The gap is the edge
Now put the two numbers next to each other and you get the only stat in fantasy that pays you directly: the spread.
When the crowd prices a player below what the math says he'll produce, that's a buy. When the price runs ahead of the value, that's a sell, and the beautiful part is your trade still clears at the inflated price. The engine isn't arguing with your leaguemates. It's telling you which of their prices to accept.
This works because the two numbers move on different fuel. Prices move on attention, and attention is fast and streaky. Values move on production, and production is slow and stubborn. The divergence between a fast number and a slow number isn't a bug in the market. It's a structural feature, it reopens every news cycle, and it's measurable per player, per format, today.
The crowd sets the price. The engine computes the value. The gap between them is your edge.
Where to start
Go vote on VibeRank. Your taps build the price side of the ledger, and you'll know where the room is on every player after twenty minutes of arguing with three-card matchups. When you want the other column, the engine's values and the per-player gap live in PureSignal.
The market is open.