THE METHODOLOGY
Most weather traders look at a single app icon and guess. We do not guess. The weather bot runs every decision through 164 independent simulations spanning four modeling centers, and only acts when at least three of those four families agree. The inflation bot applies the same discipline across five macro data streams for CPI and PCE contracts.
Four independent forecast systems → one probability
31
GFS
31
AIGEFS
51
ECMWF IFS
51
AIFS-ENS
164
total members
Trade gate: at least 3 of 4 systems must agree.
NOAA GFS
31 physics ensemble members
Operational GFS perturbations via Open-Meteo — probability distributions for daily highs and lows at each Kalshi city coordinate.
NOAA AIGEFS
31 AI ensemble members (Project EAGLE)
GraphCast-class architecture on NOAA data, pulled from AWS S3 GRIB2 and parsed to 2 m temperatures at each market city.
ECMWF IFS
51 ensemble members
Independent European centre physics ensemble — another uncorrelated vote on the same atmospheric outcome.
AIFS-ENS
51 AI ensemble members
ECMWF's AI ensemble adds a fourth modeling philosophy so the gate is about structural agreement, not one vendor's bias.
Inflation bot (CPI + PCE)
Five independent signal sources
Cleveland Fed nowcast, FRED energy leading indicators, BLS CPI subcomponents, BEA PCE tables, and a homemade weighted nowcast. When those views diverge by more than 0.15 percentage points, that disagreement becomes the actionable signal.
[ INFO ] Fetching GFS ensemble members 1–31 via Open-Meteo API...
[ INFO ] Fetching AIGEFS ensemble members 1–31 from NOAA S3 (GRIB2)...
[ INFO ] Fetching ECMWF IFS ensemble members 1–51...
[ INFO ] Fetching AIFS-ENS members for grand ensemble...
[ CALC ] Building probability distributions — 164 members, 4 independent systems...
[ GATE ] SYSTEM_AGREEMENT gfs=PASS aigefs=PASS ecmwf=PASS aifs=PASS required=3of4 → PASS
[ GATE ] EDGE_SCORE ticker=KXHIGHTDC spread=0.91 volume=0.88 imbalance=0.76 mispricing=0.85 composite=0.85
[ EXEC ] ELIGIBLE ticker=KXHIGHTDC side=YES price=0.72 edge_score=0.85 → PLACING ORDER
[ INFO ] Scanning remaining tickers... majority rejected (threshold not met).
TRADE GATING
Every potential trade is scored on four dimensions before a single dollar is risked. Only trades that pass all four filters with a composite score above threshold get executed. In practice, the bot rejects 95%+ of opportunities. Patience is the edge.
How tight is the bid-ask spread? Tight spreads mean real liquidity, not phantom orders sitting in an empty book.
Is there actual trading activity, or is this a dead market? The bot requires minimum volume thresholds before entering.
Which side has more demand? If we are buying YES and the order book is stacked with YES bids, the market agrees with us.
How far is the market price from our 164-member grand ensemble probability? This is the core edge: the gap between what the market thinks and what the forecast systems say.
Example composite score calculation
spread=0.91 × 0.30 = 0.273
volume=0.88 × 0.20 = 0.176
imbal =0.76 × 0.25 = 0.190
misprice=0.85 × 0.25 = 0.213
─────────
composite = 0.852 ✓ ABOVE THRESHOLD → EXECUTEPRODUCT STATUS
The grand ensemble upgrade, inflation bot, SQLite defaults, strike consistency, regime flags, and optional dashboard are in the current Gumroad release. Existing buyers receive updates at no extra charge.
164-member grand ensemble (GFS + AIGEFS + ECMWF + AIFS)
Four independent modeling centers. Agreement gate: at least 3 of 4 must align.
Inflation bot — five macro signals
Cleveland Fed nowcast, FRED energy, BLS CPI detail, BEA PCE, homemade weighted nowcast.
Strike consistency + regime detection
Logical position checks and contradiction logging before new risk is added.
Optional visual dashboard + SQLite logging
Zero-config local database and an optional UI for monitoring runs.