Models

Daily ensemble: input models

6 signals · fed into the stacker

An LLM-designed baseline trained on gasoline futures volatility and short-term returns.

Gulf SST Storm Risk ModelCameron Anthony

Predicts whether to buy gas based on Gulf of Mexico sea surface temperature patterns that signal hurricane and refinery disruption risk.

WTI Gas Timing ModelChrista Irakoze

48% acc

Predicts whether consumers should buy gas now based on recent WTI crude oil price levels and short-term trends.

U.S. Gasoline Inventory SignalHrijoo Bosu Mullick

75% acc

Predicts whether gas prices will be higher next week by tracking U.S. weekly gasoline inventory levels from the EIA.

WTI Futures Gas SignalMarcus Holder

52% acc

Predicts whether gas prices are more likely to rise by using short-term WTI crude oil price trends.

WTI Crude Oil SignalRyan Wolff

53% acc

Predicts whether retail gas prices will be higher in one week by tracking WTI crude oil spot price trends and momentum via Alpha Vantage.

Final layer: ensemble meta-learner

stacks the daily signals

Ensemble Meta-LearnerJohn Wheeler

51% acc

The second-layer classifier. Reads every teammate's buy-probability as a feature and produces the final daily yes-or-no that shows up on the homepage.

Reads every teammate model's buy-probability as a feature and produces the final YES/NO shown on the homepage.

Tomorrow signal

hourly · local LLM

Tomorrow Signal (Local Gemma)John Wheeler

100% acc

Hourly 'should I buy gas tomorrow?' signal. A locally-hosted Gemma 4 E4B (~8B-parameter MoE) reads the news, scores sentiment, and that score is blended with intraday RBOB gasoline futures into a 24-48 hour call. Nothing in the pipeline hits a cloud LLM — it all runs on our laptop.

Runs on its own cadence and does not contribute to the daily YES/NO. Produces a short-horizon (24-48h) verdict from news sentiment + intraday gasoline futures.

Back to today's recommendation