Trump Gas Prices: A Brilliant 4-D Chess Move?

How the Iran conflict will drive gas prices to midterm highs – and why that might accelerate the energy transition

Under Biden, high gas prices were a supply-and-demand story largely beyond presidential control: a post-pandemic demand surge colliding with OPEC production discipline and a European energy crisis triggered by Russia’s invasion of Ukraine. The price spikes were global, structural, and had more to do with refinery bottlenecks and shipping disruptions than with any policy choice made in the Oval Office.

The current situation is different. The Iran conflict is a war of choice. The military escalation in the Persian Gulf has disrupted one of the world’s most critical oil transit corridors, and crude prices have responded accordingly. Brent crude is trading above $109 per barrel in steep backwardation, with the futures curve pricing in sustained supply risk. The market structure tells a clear story: traders expect tight supply now and don’t expect relief soon.

This is not a supply-demand mismatch. This is a policy-driven price shock, and the price trajectory over the next 9 months suggests it will get worse before it gets better. By the time midterm elections arrive, American voters will be staring at gas prices that make Biden-era highs look modest.

What the Model Shows

I built a forecasting model that combines three approaches – an ARIMA time series model, Facebook’s Prophet for seasonal patterns, and a regularized LASSO regression across technical and macroeconomic features. The three are blended into a weighted ensemble, with a spot-anchoring correction that prevents the model from drifting too far from current reality.

The model is benchmarked against naive alternatives (just use today’s price, use the historical average, extend the recent trend). It beats all three, with a +8.6% skill improvement over the random walk baseline.

The ensemble forecast outperforms all three naive baselines on the 30-day backtest. The +8.6% skill vs the random walk is modest but meaningful. Oil markets are efficient; beating 'just use today's price' by any margin is a real result.

The ensemble forecast outperforms all three naive baselines on the 30-day backtest. The +8.6% skill vs the random walk is modest but meaningful. Oil markets are efficient; beating ‘just use today’s price’ by any margin is a real result.

The 9-Month Forecast

The model projects Brent crude dipping slightly to around $104/bbl at the 3-month mark as summer driving demand fades, then climbing back through $113 at 6 months and reaching $123 by the 9-month horizon as heating season demand compounds the supply constraints from the Iran conflict.

Brent crude 9-month ensemble forecast. The near-term dip reflects seasonal demand patterns; the subsequent rise reflects sustained supply risk from the Iran conflict propagating through the futures curve.

Brent crude 9-month ensemble forecast. The near-term dip reflects seasonal demand patterns; the subsequent rise reflects sustained supply risk from the Iran conflict propagating through the futures curve.

The spot-anchoring diagnostic shows how the correction works. The raw ensemble (trained on 18 months of price history) wants to forecast around $86/bbl – anchored to a period when prices were lower. The spot-anchoring correction pulls the near-term forecast up to current reality ($107), then gradually releases to the model’s learned dynamics over the forecast horizon.

The raw ensemble underestimates current prices because it was trained on a period that included lower prices. The spot-anchoring correction blends the forecast with current spot, decaying at a 90-day halflife.

The raw ensemble underestimates current prices because it was trained on a period that included lower prices. The spot-anchoring correction blends the forecast with current spot, decaying at a 90-day halflife.

California vs Texas: The Price Gap That Keeps Growing

California gas prices are already the highest in the country. The model forecasts CA retail gasoline rising from $5.82/gal in the near term to $6.21/gal by the 9-month mark. Meanwhile, Texas drivers are paying $3.89/gal – a CA premium of $2.19 per gallon.

9-month forecast summary. California gas (red labels) is projected to approach $6.21/gal by the 9-month horizon. Texas gas (green) sits at $3.89 -- a $2.19/gal premium that reflects California's regulatory environment on top of the global crude price increase.

9-month forecast summary. California gas (red labels) is projected to approach $6.21/gal by the 9-month horizon. Texas gas (green) sits at $3.89 – a $2.19/gal premium that reflects California’s regulatory environment and supply-demand dynamics on top of the global crude price increase.

That $2.19 premium is structural – it reflects CARB summer blend requirements, higher state taxes, refinery concentration, and transportation costs. It doesn’t change much whether crude is at $70 or $110. What changes is the base that the premium sits on top of. When Brent rises from $80 to $120, the retail price in California doesn’t just add $40 worth of crude cost – it amplifies through refining margins, taxes calculated as percentages, and the psychological effect on station pricing.

The Political Calculus

Here’s where it gets interesting. Biden took enormous political damage from gas prices that were, by any honest economic analysis, driven by global supply-demand dynamics he didn’t control. Voters don’t care about Brent-WTI spreads or refinery utilization rates. They see the number at the pump and they blame whoever is in the White House.

The Iran conflict reverses the causation. These aren’t prices driven by pandemic recovery or European energy crises. These are prices driven by a military escalation that the administration chose to pursue. The supply disruption in the Strait of Hormuz is a direct, traceable consequence of policy decisions, not an exogenous shock.

If the model’s trajectory holds – and the backwardation curve suggests markets agree with the direction if not the exact numbers – gas prices will be near peak levels by midterm season. The question is whether voters connect the dots between the Iran escalation and the price at the pump.

The Renewable Energy Angle

There’s an irony buried in the forecast. Sustained high oil prices are the single most effective accelerant for renewable energy adoption. Every dollar added to the per-gallon cost of gasoline shifts the breakeven calculation for electric vehicles, strengthens the economic case for rooftop solar, and makes utility-scale wind and battery storage more competitive against natural gas peaker plants.

The Biden administration tried to accelerate the energy transition through subsidies and regulation (the IRA, EPA tailpipe rules, offshore wind leases). The Trump administration may accomplish more of the same thing by accident, through sustained high fossil fuel prices driven by geopolitical risk.

If you wanted to design a policy to simultaneously punish fossil fuel dependence, motivate EV purchases, and create political consequences for the party in power by midterm elections, it would look almost exactly like what’s happening now. Hence the “4-D chess” framing – though calling it “chess” implies a level of strategic intention that is quite generous.

Model Details and Assumptions

The model assumes no new macro shocks (FRED indicators held at current rates of change), decaying price momentum, unchanged news sentiment, and no additional OPEC production changes, refinery outages, or regulatory shifts. These are conservative assumptions. If the Iran conflict escalates further, or if OPEC tightens in response, the upside risk to prices is significant.

Full methodology, code, and weekly diagnostics will be published on the project’s GitHub Pages site. The model runs weekly with a growing news sentiment cache and run-to-run accuracy tracking.


Built with R and an ensemble of ARIMA, Prophet, and LASSO models. Version: v13. Data sources: Yahoo Finance (crude futures), EIA (retail gasoline), FRED (macro indicators), RSS feeds (news sentiment).

Leave a comment

Your email address will not be published. Required fields are marked *