ELI5 · FloatNet
Why price is not enough
Five pictures. Why a contract needs a signed number that is not the spot price — and why that number should float free of any one exchange.
1. Why is price not enough?
Price tells you what the thing costs right now. An apple is $2. Bitcoin is some dollars. That is useful. It is also only one fact.
A volatility contract is not a bet on the apple. It is a bet on how jumpy the apple’s price will be. Jumpiness lives in the option surface: many little bets at many prices (cheap apples, expensive apples, crashes, quiet weeks). You cannot look at today’s $2 and decide who won a bet about jumpiness. The scale and the bet are different objects.
2. Why settle a number that is not price?
When a game ends, you pay from the scoreboard, not from a photo of the ball. Price is the ball in the air — where the asset is at this second. Settlement is the agreed score at the end of the contract.
For a vol contract, the scoreboard has to be a number built from the option surface (or from a published formula on that surface). Two desks wrote “we pay the difference versus THE NUMBER.” On expiry morning they must be able to point at the same print. If that print is just “the last trade in the coin,” they settled a price bet, not a vol bet. They wrote the wrong contract.
3. The underlier, and the auction
An underlier is what the wrapper is written on. A note on bitcoin expires against bitcoin’s price. A note, a future, or a prediction on jumpiness, on the wing, on the tail, needs a different underlier: a named slice of the option surface, plus a small book of listed options that replicates that slice. That book is the strip — the same idea as a variance strip. The slice is the line. Together they are the underlier the wrapper can expire against.
On expiry morning those options have to print. An auction is how the market prints them in one go: bids and offers meet, a single opening price comes out for each strike in the strip, and everyone can see it. Anyone who thinks the opening is wrong can trade in it. The published recipe then reads those opening prints — each strike with a published weight — and turns them into one number: tail mass, term, shortfall, whichever line the wrapper named.
Named firms run that recipe on those prints and sign. The registry holds the result. The wrapper pays against it. The next token, the next note, the next listed future can name the same underlier and go to the same auction.
4. A shared scoreboard
Someone has to write the scoreboard. Today that is usually the same house that owns the stadium: they compute the number and list the future. Every new wrapper then lives in that building.
Here the scoreboard is shared. On expiry the options that make the strip open in an auction — a live market anyone can trade. Named firms apply the same published recipe to those opening prints and sign. The registry holds the result. A venue lists. An issuer points a note, a fund, a token, a prediction at that print. Matching stays on the venue’s book. The next asset uses the same signers and the same recipe.
5. How is the float surface better than VIX?
VIX is a thermometer: one number for “average jumpiness over the next 30 days,” built as a variance-swap approximation and then square-rooted. It did a historic job — it turned a statistic into something futures could pay against. It also throws away most of the weather. Tails (crashes), term (9 days vs 90 days), implied-versus-what-happened, expected loss in the wing: the market prices all of that. VIX does not settle any of it.
The float surface is the weather map of the same option tape. The published vector Φ is a set of slices of that map. Settlement lines — each a piecewise-linear g, each with a variance-strip analogue — are wing mass (P3), curve (P4), term (P6), shortfall (P7). Premium vs realised (P1a) mixes the surface with a physical path; it advises, it does not settle. P2 (armed state) and P5 (cross-venue) are desk flags. A product that cares about crashes settles to the tail, not to a 30-day average. High-order slices on a truncated grid are not equally listable; P3 is first. “Float” means the map is not nailed to one stadium.
6. How does this help tokenized digital and real-world assets?
A token that represents a note, a fund, a prediction, a perp, or a claim on a real-world asset is a wrapper. Wrappers expire against a number. When they share one signed risk state, buyers pool, venues list without writing a new formula, and settlement travels with the product. A perp, a prediction, a tokenized note, a fund share — the same tail mass, the same term, the same signers. That is how a wave of new assets lists without each one standing up its own house index.
One signed print, names and dollars behind it, is the main. Many wrappers connect. A hedge on one venue is a hedge on another. Spreads tighten because the underlier is shared. Capital can move.
7. What the book gains when it can settle to F2
The question is not whether F2 beats DVOL. DVOL is still the best second-moment hedge a desk can buy today. The question is what that book still eats when it cannot settle to F2 — first-order density leftover a variance index does not cancel — and what it eats once an F2 note sits in the same sleeve. Same for the tail book: VaR and ES stay; P3 is the listed-wing mass they do not print. Inclusion, not replacement.
Network console · settler navigation — leftover of the book with the new print versus without it, both genera, on a daily clock.
Price is the apple. Settlement is the scoreboard. The float surface is the map. New assets are buildings on a shared main. That is the company.