Outcomes (HIP-4)
Trade HIP-4 binary outcome markets and read the implied vs reference probability shown next to every contract.
Outcomes are Hyperliquid HIP-4 binary contracts. Each contract pays 1 USDC if its condition resolves YES at expiry, or 0 if it resolves NO. Today the only deployed class is priceBinary, which settles YES if the underlying perp's mark price is greater than or equal to a strike price at the expiry time.
You see them in the Outcomes tab of the lower activity area in Trade whenever the selected market has at least one matching binary listed.
What Each Row Shows
Each row condenses a contract into one line.
| Field | Meaning |
|---|---|
UNDERLYING ≥ strike (or ≤) | The condition that pays YES. Direction follows whether the strike sits above or below current mark. |
Period badge ( 3m, 15m, 1d, …) | Recurring cadence at which the contract is re-listed. |
Countdown | Time left until expiry. |
Move % | Underlying move needed from current price to reach the strike. |
Implied | Market-implied P(YES). It is the YES leg's ask price, clamped to [0.001, 0.999]. Outcome prices are already a probability — buying YES at 0.62 means the book is pricing YES at 62%. |
ref ~ | The reference probability — see Implied vs Reference Probability. |
σ distance | How many expected standard deviations the strike is from the current mark, scaled to time-to-expiry. |
YES / NO pills | Current ask on each leg. Click to send a market buy at the configured ticket size. |
The ref ~ and σ values are hidden on narrow viewports. Open the full market detail (the expand icon) to see them.
Payoff Math
A binary contract is worth 1 USDC if it hits, 0 otherwise.
- Buy YES at price
P→ payP, receive1if YES,0if NO. Return on success:(1 − P) / P. - Buy NO at price
Q→ payQ, receive1if NO,0if YES. Return on success:(1 − Q) / Q.
The two legs are linked by no-arbitrage: P(YES) + P(NO) ≈ 1. When one side is missing a quote, the panel infers it from 1 − other side.
Implied vs Reference Probability
The two probabilities shown on each row answer different questions.
Implied
The implied probability is what the market is currently charging for YES. It comes straight from the YES book — specifically the best ask, clamped to [0.001, 0.999] so the math stays defined at the boundaries. Nothing model-based; it is just the price.
Reference
The reference probability is what a textbook zero-drift Brownian model would price YES at, given:
- the current underlying mark price
S - the strike
K - the time remaining to expiry
t - a daily volatility estimate
σ_d
Under that model the underlying's terminal price S_T is log-normal and the closed form is:
P_ref(YES) = Φ((ln(S/K) + ½ · σ_t²) / σ_t)
where:
σ_t = σ_d × √(t / 1 day)— daily σ scaled to the time horizon, so a 15-minute outcome is judged against minutes of vol, not a full day's worth.Φis the standard normal cumulative distribution function.
Where σ Comes From
σ_d is a single-sample proxy off the perp's markPx and prevDayPx:
σ_d = |ln(markPx / prevDayPx)|
Floored at 0.5%/day so a flat market still produces a non-zero σ band, and capped at 50%/day to avoid blowups. Defaults to 2%/day when either price is missing.
This is not a statistically rigorous vol estimate. It is deliberately rough, because it only feeds the reference probability and the σ-distance label — never order sizing or risk numbers.
What the Model Assumes
| Assumption | Practical Effect |
|---|---|
Zero drift ( μ = 0) | The reference is direction-agnostic. It does not bake in any view that the underlying is trending. |
Log-normal terminal | Implicit Brownian motion in log-prices. Ignores fat tails, skew, and jumps. |
One-day single-sample σ | The vol number is a yesterday-vs-today move, not realised or implied vol. |
No funding, fees, or rebates | The reference is a clean theoretical price, not an after-cost expectation. |
How to Read the Gap
Treat the reference as a mild prior for sanity-checking the book, not a forecast.
- Implied ≈ Reference → the market is pricing the contract in line with a vanilla zero-drift assumption. No edge signal.
- Implied >> Reference → YES looks expensive vs the prior. The panel may surface a
speculate NOchip if the gap exceeds 5%. - Implied << Reference → YES looks cheap vs the prior. The panel may surface a
speculate YESchip if the gap exceeds 5%.
The 5% threshold is fixed and applies to the absolute difference |ref − implied|.
Suggestion Chips
Below each row, the panel can surface one or two ranked chips. Each chip is a deterministic suggestion based on your open perp position and the implied/reference gap — not a recommendation engine.
| Category | Trigger | What It Does |
|---|---|---|
Hedge | You hold a perp on the underlying and the strike falls inside your loss zone — long with strike < entry, or short with strike > entry. | Sizes a NO (long hedge) or YES (short hedge) buy that pays out exactly when the perp is losing the most. Capped at 5% of position notional by default. |
Speculate | The implied vs reference gap exceeds 5% in either direction. | Buys the cheaper side at a small ticket — half the hedge cap when you have a position, $25 when you do not. |
Amplifier | You hold a perp and the strike is on the same side as the position's profit direction — long with strike ≥ entry, or short with strike ≤ entry. | Buys the leg that pays when the perp wins, only if the cost is ≤ 0.70 so there is still convex value left. Tagged neutral. |
The math on each chip is explicit (net if win, net if lose), so the trade is auditable before you click.
Caveats
- Outcomes are fully collateralized. No leverage, no liquidation, no funding. The most you can lose on a leg is the price you paid.
- Reference probability is heuristic. Do not size off it. Use it the same way you would use a one-line "is this offer reasonable?" check.
- σ distance is informational. A
+0.50σstrike is "half a daily move away, scaled to expiry" — useful for ranking strikes against each other, not for risk modelling. - Pre-expiry mid can disagree with both implied and reference. That is the market doing price discovery; both numbers will track the book over time.
For execution mechanics on the YES/NO sides see Order Types. For the underlying-price concepts (mark vs oracle vs last) see Trading Fundamentals.
