});
Dashboard
Weather Markets
0
Edge Found
0
Live Edges
0
Models Active
82
Activity Log
--:--:--Waiting for scan...
Edge Analysis 4-Model Active
Scan first, then click Analyze Top 15
Open Positions 0
No open positions
Trade History
No closed trades yet
Live Signals SCANNING
Signals
--
Avg EV
--
High Conf
--
Markets
--
Loading signals...

TOP WEATHER TRADERS

Live leaderboard from Polymarket Analytics

SOURCES & EDGE
STATION PARSERACTIVE
ICAO Station Confidence
Confidence tiers 0–3 from nearby airport METAR/TAF observations. No-trade gate if station confidence < 2. Ensures forecasts are anchored to real surface data.
4-MODEL CONSENSUSACTIVE
GFS-31 + ECMWF-51 + UKMO + MeteoFrance
Ensemble consensus blend via Open-Meteo. Weighted uncertainty from 4 global NWP models (113 total members). Produces calibrated probability distribution per market.
NOWCASTING ENGINEACTIVE
AR(1) + Monte Carlo (5000 samples)
Activated when resolution ≤ 24h. AR(1) residual model with observation anchoring. Half-life 2h near-peak, 4h otherwise. Blends real NWS obs with model forecasts.
REGIME CLASSIFIERACTIVE
4-Feature Weather Regime Detection
Features: ensemble spread, wind shift, cloud cover, precip. Classifies as frontal / marine / convective / clear-sky. Shapes probability distribution per regime type.
CROSS-MARKET FILTERACTIVE
Delta Z-Score Correlation Check
Only applied when inter-market correlation r > 0.90 in same regime. Prevents correlated position clustering. Uses delta z-score to flag redundant exposures.
CLOB EXECUTIONACTIVE
Passive Limit Orders
30–90s time-in-book. Up to 3 reprice attempts. Fill-probability proxy for sizing. Position cap 15–25% of visible depth. No market orders.
V2 EDGE CALCULATION FLOW
01
Station Parse
Confidence tier from nearest ICAO station METAR/TAF
02
Market Discovery
Gamma API slug lookup for active weather contracts
03
4-Model Forecast
Consensus blend with weighted uncertainty (113 members)
04
Nowcasting
If ≤ 24h to resolution: AR(1) + 5000 Monte Carlo samples
05
Regime Classification
Front / marine / convective / clear-sky distribution shaping
06
Cross-Market Filter
Delta z-score check (r > 0.90 same regime only)
07
Theo EV Calc
SUM[size × (p × (1/P − 1) − (1−p))] − cost_proxy
08
Diurnal Gate
Pre / near / post-peak staging by latitude window
09
CLOB Execution
Passive limit, fill-prob proxy, size cap 15–25% depth
Gamma API
Market discovery Events Tags
Slug-based lookup 7-day window
ACTIVE
CLOB API
Orderbook Prices Trading
Real-time bid/ask Market orders
ACTIVE
Open-Meteo GFS
30-member ensemble Global
00z/06z/12z/18z UTC 16-day
ACTIVE
NWS (US only)
Real obs + hourly forecast blend
Airport stations KLGA/KORD/KMIA
ACTIVE
KDE Probability
Bayesian Beta-Binomial + KDE
integrate_box_1d uncertainty output
ACTIVE
Kelly Criterion
Fee-adjusted Fractional
f* = (p×W − q×L) / (W×L)/(WL) ¼ Kelly default
ACTIVE
Edge Calculation Flow
01
Discover
Slug-based Polymarket lookup: highest-temperature-in-{city}-on-{month}-{day}-{year} for all 23 cities 7 days ahead
02
Parse
Extract city, date, temperature unit (C or F), bin ranges from market question titles
03
Forecast
Fetch GFS 30-member ensemble via Open-Meteo at ICAO airport coordinates (KLGA not Manhattan)
04
NWS Blend (US)
For US cities: combine real station observations with hourly forecast to get daily maximum with time-weighting
05
KDE Probability
Bayesian Beta-Binomial + Gaussian KDE over 30 ensemble members P(temperature in bin) with uncertainty
06
MOS Correction
City/season bias correction + spread factor + skill decay by lead day (day 7: 48, day 10: 16)
07
Edge & Kelly
edge = GFS_prob → market_price EV = (p→P)/P → fee fee-adjusted Kelly position size
CONFIGURATION

⚙️ SETTINGS

(Applies to bot Edge trades)
(Hard cap per position)
(ICAO tier 0–3, default 2)
(Passive limit order duration, 30–90s)

💰 WALLET

⚠️ Never enter your private key anywhere. Connect wallet directly on Polymarket.com and paste your proxy address here only.

📡 DATA SOURCES

GFS Ensemble (Open-Meteo)ON
ECMWF IFS (51-member)ON
UKMO (Met Office)ON
MétéoFrance (Arpège)ON
Polymarket Gamma APION
NWS (US cities only)ON
HOW IT WORKS
How WeatherEdge v2 Finds Edge
The complete pipeline behind every trade signal. WeatherEdge v2 automates the entire Polymarket weather trading pipeline using a 4-model ensemble consensus, nowcasting, regime classification, and passive CLOB execution.
COVERAGE
58 cities with ICAO airport coordinates across the US and key international markets. Each city is mapped to the nearest airport METAR/TAF station for real surface observations. Station confidence tiers (0–3) gate whether a market is tradeable — no trade below tier 2.
FORECAST MODELS
4 global NWP models provide the probability backbone:
GFS — 31-member ensemble (NCEP)
ECMWF IFS — 51-member ensemble (European Centre)
UKMO — UK Met Office deterministic (Met Office)
MeteoFrance — Arpege/Arome (Météo-France)
All accessed via Open-Meteo API. Consensus blend produces calibrated probability with weighted uncertainty across 113 total ensemble members.
EDGE METRIC
The primary gate metric is theoretical_full_ev: SUM[size × (p × (1/P − 1) − (1−p))] − cost_proxy. This replaces the simple edge percentage from v1. Only positions with positive theo EV after cost estimation pass the gate. Sizing uses ¼-Kelly fraction by default.
NOWCASTING
When a market is ≤ 24 hours from resolution, the nowcasting engine activates. It runs an AR(1) residual model on the latest NWS observations with 5,000 Monte Carlo samples. Observation anchoring uses a half-life of 2h near the diurnal peak and 4h otherwise, blending real surface data with model forecasts for high-confidence last-mile predictions.
DIURNAL STAGING & REGIME
Markets are staged into pre-peak, near-peak, and post-peak windows based on latitude-derived solar timing. A 4-feature regime classifier (ensemble spread, wind shift, cloud cover, precip) tags each market as frontal, marine, convective, or clear-sky — shaping the probability distribution accordingly.
EXECUTION
All trades use passive CLOB limit orders — never market orders. Time-in-book ranges 30–90 seconds with up to 3 reprice attempts. Position sizing is capped at 15–25% of visible orderbook depth. Each trade gets a frozen model snapshot via a unique trade_group_id so post-trade analysis can reconstruct the exact signal state.
Activity Log
Full Activity Feed
--:--:-- INFO WeatherEdge initialized. Wallet not connected.