HYPERLIQUID · HIP-3 PREDICTION MARKET · OUTCOME #303
Draw
33.1¢
66.9¢
▸ Advanced metrics · M2M bundle
hyperliquid · pred-draw-303 · fresh · feed 1s old24h sparkline · 60 pts▼ —
realized vol (ann.)
85.68%
max drawdown
3.42%
sharpe
—
ulcer index
2.12%
RMS drawdown
pain index
1.38%
mean drawdown
mod. VaR 95%
0.00%
Cornish-Fisher
martin ratio
—
ret / ulcer
CDaR 95%
3.42%
cond. drawdown
gain/pain
1.22
Σgain / Σ|loss|
sterling
—
ret / CDaR
omega (θ=0)
1.22
upside/downside
roll spread
0.0 bps
implied (price-only)
bars used
2000
store
spread
—
24h Δ
—
flow lean
—
carry
flat
signalNEUTRALconfidence 20%
Same bundle via M2M API:
/api/m2m/hl-pred-draw-303/bundle · venue execution: hyperliquid →LIVEPOLL0SRCFRESH⏱--:--:-- UTCNEXT8.0sUP0s--:--HIST
▶ STREAMING·HYPERLIQUID·POLYMARKET·0 POLLS·SRC FRESH·UPTIME 0s·NEXT POLL 8.0s·CC0 OPEN DATA·HYPO.MARKETS·▶ STREAMING·HYPERLIQUID·POLYMARKET·0 POLLS·SRC FRESH·UPTIME 0s·NEXT POLL 8.0s·CC0 OPEN DATA·HYPO.MARKETS·
Yes mid · live
33.1¢
No mid · live
66.9¢
24 bars · close 33.60¢ · 24h -1.09%
Σ 100.00% · fair
Σ-sides total = 100.00% (tight rounding)
H(p) entropy = 0.916 / 1.00 bits (92%) · high uncertainty
Yes33.1%33.1¢3.02×▬ +0.00pp
No66.9%66.9¢1.49×▬ +0.00pp
primary vs counter implied %
Σ 3353 · peak 1347
1.4s
33.093¢
66.907¢
100.00%
0.00pp
24 bars
33.60¢
-1.09%
§1 · 24h time-series
range [32.54¢, 34.30¢] · span 1.76pp · MA(5) latest 33.03¢
24 bars · last close 33.60¢
Σ vol = 3353 · peak 1347 · mean 139.7
§2 · Distribution of one-bar increments Δp = pₜ − pₜ₋₁
n=23 · positive 5 · negative 2
reference line = identity (perfect normality). Heavy upper-right tail = fat positive tail.
§3 · Sample moments (prices)
SAMPLE MOMENTS · N=24LEFT-SKEWED (G₁=-0.72)
FIVE-NUMBER SUMMARY · BOX PLOT
SKEWNESS · G₁-0.723left-skewed
−3−10+1+3
EXCESS KURTOSIS · G₂-0.264mesokurtic · normal-like
−30+2+4+6
μ = mean · σ = standard deviation · CV = coefficient of variation · skew (G₁): >0 right-tail · kurt (G₂, excess): >0 leptokurtic. 95% CI uses 1.96·SE around μ. σ × 1.349 ≈ IQR under normality.
§6 · Time-series structure
TIME-SERIES STRUCTUREREGIME: MEAN-REVERTING · ρ(1) -0.30 + ADF rejected
HURST EXPONENT [0, 1]
H = 0.871STRONGLY PERSISTENT
0
anti-persistent0.45
mean-reverting0.5
random walk0.55
persistent1
strongly trending
anti-persistent0.45
mean-reverting0.5
random walk0.55
persistent1
strongly trending
AUTOCORRELATION FUNCTION · ρ(k) for k=1..5
OLS TREND · t-STAT · [-5, +5]
−5 reject−1.960 retain H₀+1.96+5 reject
ρ(k) = lag-k sample autocorrelation · H = R/S Hurst exponent · t = OLS-trend t-statistic. Significance bands at ±2/√n approximate the 95% white-noise envelope. α=0.05 critical |t|=1.96; α=0.01 |t|=2.58.
§7 · Microstructure
MICROSTRUCTURE · MARKET QUALITYPERFECT · ARB-FREE Σ=100.00%
TWO-SIDED PRICING
33.09¢
66.91¢
Σ-SIDES ARBITRAGE TEST
Σ = 100.00% · |1 − Σ| = 0.000pp
24H ACTIVITY · LIQUIDITY
Σ-sides = primary + counter implied probabilities. Perfect arb-free Σ = 100%. |1−Σ| > 2pp suggests synthetic outright arbitrage.
§8 · Position sizing & edge analysis
FAIR MARKET · no edge
Probability scale (Yes)
Implied decimal odds
Yes3.02×(33¢)No1.49×(67¢)
Kelly bet-size (% of bankroll) K* = 0.00%
Entropy H(p̂) = 0.916 bits (92% of max) · high uncertainty
Σ-sides = 100.00% · |1 − Σ| = 0.00pp · tight cross-venue rounding
K* full = (b·p − q)/b where b = (1−p̂)/p̂ are the net odds implied by p̂. ½K and ¼K are industry-standard conservative fractions.
§9 · Resolution criteria
This outcome resolves to Yes if the Game ends in a draw, or if the Game is canceled or not completed by July 19, 2026 at 23:59 UTC without FIFA officially declaring a winner.
§10 · Hourly return heatmap
5 up bars · 2 down · best 1.00% · worst -1.69% · typical |Δ| 0.224%
§11 · Equity curve & underwater drawdown
final equity 0.9960 (-0.40%) · max DD -1.77% · time-under-water 15/24 bars
§12 · Rolling-window statistics (w = 5 bars)
latest -0.422 · range [-41.86, 58.90] · μ 3.240 · positive Sharpe = excess-return-per-risk earned by buying-and-holding through this window
latest 62.26% · range [0.00%, 90.67%] · μ 41.07% · σ̂ scaled to annualised (×√8760)
latest 0.001 · |ρ| > 0.3 ⇒ regime with persistence (ρ > 0) or reversal (ρ < 0) · |ρ| ≤ 0.1 = consistent with random walk
§13 · Hypothesis tests (α = 0.05)
2 of 6 REJECT · mixed evidence2 reject·4 pass·α = 0.05
Jarque-Bera
REJECT H₀***H₀: Δp ~ Normal(μ, σ²)
STATISTIC
43.5739
p-VALUE (log scale)
< 0.0001
Ljung-Box(h=5)
FAIL TO REJECTnsH₀: No serial autocorrelation up to lag 5
STATISTIC
2.8581
p-VALUE (log scale)
0.7244
Dickey-Fuller (τ_μ)
FAIL TO REJECTnsH₀: p has a unit root (non-stationary)
STATISTIC
-2.4691
p-VALUE (log scale)
0.1297
Wald-Wolfowitz runs
FAIL TO REJECTnsH₀: Sign sequence of Δ is random
STATISTIC
1.2153
p-VALUE (log scale)
0.2243
KPSS (μ stationarity)
REJECT H₀*H₀: p IS level-stationary
STATISTIC
0.5792
p-VALUE (log scale)
0.0245
Variance ratio q=2
FAIL TO REJECTnsH₀: Δp is a random walk · VR = 1
STATISTIC
-1.3452
p-VALUE (log scale)
0.1786
Each row states an explicit null H₀, the test statistic, an approximated p-value, and the decision. REJECT means evidence against H₀. KPSS complements ADF (rejecting both ⇒ ambiguous; rejecting one ⇒ clean verdict).
§14 · Spectral analysis (DFT periodogram)
dominant period ≈ 2.56h (freq 0.391) · concentrates 27.3% of total energy · Σ|X̂|²/n = 2.839e-4
▸ Depth section using sovereign-store price series (5000 bars · effective 5257847 bars/year) — annualisation reflects native polling cadence, not upstream timeframes.
§15 · Honest position analytics
A binary-market analytics module framed in horizon time (days to resolution, not annualised). Estimators that need a model probability q as a first-class input (Kelly, KL divergence, Bayesian posterior, Mark-to-Market MC) only render when q is provided externally. Sweep an exploratory q at the interactive simulator →
§16 · Horizon returns
Horizon 7.0 d · σ/bar 0.028pp · expected |Δp| over horizon 0.36ppterminal variance p(1−p) = 0.2214 · n = 5000✓ n = 5000
μ per bar
+0.000pp
average Δp · drift
σ per bar
0.028pp
one-bar volatility · logit-free
Per-day movedaily
0.14pp
σ × √24
Per-horizon move7d
0.36pp
σ × √168
Terminal variancebinary
0.2214
p(1−p) at resolution
Current pricep
33.1¢
latest snapshot
Note: annualised Sharpe/Sortino are omitted — they are not meaningful for a bounded fixed-horizon binary contract that snaps to {0, 1} at resolution.
Annualised metrics are intentionally omitted — they don't apply to bounded probability series that resolve at a fixed date.
§17 · Tail risk
VaR₉₅ 0.05pp · ES₉₅ 0.06pp · method parametric · drift-correcteddrift +0.000pp/bar · quantised: yes · median step 0.01pp · unique ratio 0.01✓ n = 5000
VaR 95%
0.05pp
1.645·σ (parametric) of Δp
ES 95%
0.06pp
mean of the tail
Max drawdown
3.8pp
peak 33.5¢ → trough 32.2¢
Median step
0.01pp
price bucket granularity
Price series is bucketed (cent grid). Empirical quantiles collapse to grid points — parametric N(0, σ²) used instead.
Empirical quantiles unless the price series is bucketed (PM cent grid), in which case parametric N(0, σ²) is used to avoid grid collapse.
§18 · Odds conversion
Implied probabilityP
33.1%
= price
Decimal oddsEU
3.022
total return per $1
AmericanUS
+202
$100 wins $202
FractionalUK
2.02 / 1
profit per $1 risked
Profit per $100stake
+$202.18
clean dollar framing
underdog (+)favorite (-)your price
Price → implied probability → decimal odds → American moneyline → fractional. Five views of the same number, plus the moneyline curve.
§19 · Binary entropy
Market entropyH(p)
0.916 bit
max 1.0 at p = 0.5
Your entropyH(q)
0.916 bit
Δ +0.000 bit vs market
Surprise · YES−log₂ p
1.60 bit
self-information
Surprise · NO−log₂(1−p)
0.58 bit
self-information
Market entropy only — model entropy requires an external q.
§20 · Model-dependent surfaces
External model required
The position-economics, Kelly, KL-divergence, Bayesian and Monte-Carlo surfaces require a model probability q as input — a number independent of the market price p.
The previous build defaulted q to a tape-momentum heuristic derived from p; that produces apparent edge that is structurally guaranteed to be small and is not a useful skill signal. The auto-derived path has been removed.
To explore these surfaces with a hypothetical q, open the interactive simulator and drag the MODEL P(YES) slider. To wire a real model, POST to the NOSTRADAMUS hook (TBD) or pass ?q=… on the simulator URL.
§∞ · Provenance & attestation
- Upstream
- https://hyperliquid.xyz
- Snapshot fetched
- 2026-06-14 10:11:14 UTC
- Snapshot age
- 1.4s
- Page rendered
- 2026-06-14 10:11:15 UTC
- History points
- 24 closes · 24 counter-side closes
- Storage policy
- no persistence — fetched on every request
- SHA-256 attestation
e07e39e26098371a685d3e521acd1bd5c0e64b0e393a017732a6a273fdafc547· deterministic hash of source snapshot- Open data licence
- CC0 / public domain
§∞-2 · Related markets · explore more
HL PREDGermany94.4¢HL PREDSpain90.1¢HL PREDUruguay67.7¢HL PERPBTC-USD perpetual$64,540HL PERPETH-USD perpetual$1,676.1HL PERPHYPE-USD perpetual$60.63
Also see: /arb opportunities · RSS feed
Risk metrics
▸ sovereign store · 5,000 barsperiods/year ≈ 5.26MRealized vol (annualised)
189.99%
σ per bar = 0.000829
Mean return (annualised)
31.79%
μ per bar = 0.000000
Sharpe (rf=0)
0.17
annualised; risk-free assumed zero
Max drawdown
3.75%
peak 0.33 → trough 0.32 over 1571 bars
/api/asset/hl-pred-draw-303/risk · same metrics, JSON