Gasoline Prices Rise in Tandem with Election Poll Trends! Goldman Sachs: With 26 Days Left Until the Midterm Election, Probability of a Democratic Sweep Rises to 65%

According to Zhitong Finance APP, there are only 26 days left until the U.S. midterm elections on November 3. In the latest issue of its "U.S. Election Watch," Goldman Sachs’ Washington political team sketched out a landscape deeply intertwined with energy prices: prediction markets currently give the Democratic Party a 65% probability of winning both the House and Senate, while the probability of Republicans retaining the House majority has dropped below 10%.

The most notable feature of this October 8 report is the direct overlay of Democratic election momentum with U.S. retail gasoline prices. A chart drawn by Goldman analysts Alec Phillips and Abhi Dujjilara shows gasoline prices soaring from about $3 per gallon at the start of 2026 to nearly $5 in the spring, retreating to around $4 in summer, and then climbing again in the most recent period.

During this same period, the Democrats’ lead in the popular vote widened from about 5 percentage points to nearly 9 percentage points—the two lines moving almost in sync. As the report states: "After the Iran war outbreak pushed gasoline prices higher, the Democrats’ lead gradually widened, with the two data lines converging recently."

Another chart reveals Trump’s polling struggles on key issues. Inflation and the economy are the two top voter concerns, and Trump’s net approval on both sits deep in negative territory—about -48 for inflation and -33 for the economy.

Sweep Probability Rises from 52% to 65% in a Month

On September 7, the probability of a Democratic sweep (winning both houses) had just hit a record high of 52%. A month later, Polymarket data cited by Goldman shows this probability has climbed to about 65%. Meanwhile, the scenario of "Republicans holding the Senate, Democrats taking the House"—still the market consensus into the fall—has fallen below 30%.

This has significant implications for asset allocation. Bank of America Merrill Lynch Chief Strategist Michael Hartnett characterizes a Democratic sweep as a "significant risk-off event," expecting stocks to drop more than 10%, the dollar to weaken, and bond yields to fall. His preferred risk-hedging play is shorting financial stocks and the dollar, on the assumption that investors have so far largely ignored election risk—which has now shifted, with a sweep moving from the "coin toss" 52% odds last month to a clear baseline scenario.

House All but Decided; Senate Is the Real Battleground

According to prediction markets, the House result is almost locked in. The probability of Democrats netting at least three seats and regaining the House majority exceeds 90%. Currently, Democrats lead the popular House vote by 8.9 percentage points over Republicans—compared to the Republican 2.6 percentage point victory in 2024, this is a swing of about 12 points. The 29 Republican seats Trump won in 2024 by less than a 12-point margin are precisely the ones expected to turn blue.

As for the decennial redistricting that Republicans had high hopes for, Goldman’s conclusion is blunt: "Overall, redistricting is expected to net Republicans only a handful of seats." Prediction markets currently give Democratic candidates at least a 50% win rate in 25 Republican seats, and Democrats only need to net three seats to regain the majority.

It’s notable that although Republicans lead in counter spending in most key House districts, this money is being poured into seats already deemed unwinnable by the market.

The Senate is where real uncertainty lies. Democrats currently hold 47 seats and need a net gain of 4 to win the majority. Prediction markets give them a 65% chance, but Goldman’s charts show that while Democrats lead in every key seat except Nebraska and South Carolina, most of their leads are within the margin of error.

Goldman believes the overall Senate probability is converging on the "marginal seat" probability, which "suggests the market is more confident in Democrats having multiple paths to a majority, or there is a higher correlation among state races, or both."

The "marginal seat" currently deciding if Democrats can reach 51 is Ohio. Key states to watch include Alaska, Texas, Ohio, Maine, Iowa, and Kansas—all Republican seats, many in reliably red states.

Republicans’ Last Stand: $240 Million In Final Two Weeks

Goldman data show Republicans now hold over a $200 million spending advantage in the Senate races, entirely built up in recent weeks. Republicans trailed Democrats in Senate spending all summer, only overtaking in September, with the gap widening ever since. In the past two weeks, Republicans have spent $240 million compared to Democrats’ $130 million—a ratio of about 1.85 to 1.

The direction of these funds also tells a story: Republicans added $8 million in Kansas and $5 million in South Carolina for defensive spending, totaling $13 million—Goldman notes that no one spends this kind of money on seats they believe are secure wins.

However, the effectiveness of this money may be discounted. Goldman gives two reasons: first, super PACs pay higher advertising rates than candidates directly, and Republicans’ spending is largely from PACs; second, ad slots closer to election day command a premium, and Republicans’ spending is heavily concentrated in the final phase, whereas Democrats secured their ad buys early. Additionally, mail-in ballot applications in Ohio, Maine, and North Carolina are more Democratic-leaning than in 2024, but Goldman cautions this may make up only about 25% of total mail-in ballots.

Historical Patterns Suggest Democratic Advantage May Be Overstated

Goldman does not fully embrace the prediction market’s optimism, highlighting two important historical warnings.

First, the challenge party (opposition) sweep probability historically narrows closer to election day. Reviewing the 2002, 2006, 2010, 2018, and 2022 midterms, Goldman finds "the challenge party’s sweep probability typically drops by about 5 points on average before election day." Using the current 63% Democratic sweep probability as a base, history implies a final probability around 58%—still a high likelihood but far from a sure thing.

Second, red state polling systematically overestimates Democratic performance. Goldman’s chart of historical Senate polling errors versus state partisanship shows that "midterm polls often underestimate Democratic performance in blue-leaning states and underestimate Republican performance in red-leaning states," consistent with voters "coming home" to their party late in the cycle. This pattern is more evident in recent data: R² was 0.65 for midterms from 2018–2022, but just 0.26 over the full 2002–2022 period.

This directly impacts Democratic leads in Iowa, Kansas, Maine, Ohio, Texas, and Alaska. Goldman’s partial rebuttal: 26 days out, national Senate polls only overestimated Democrats by 0.4 percentage points on average—so nationwide bias isn’t severe. But the Senate majority will be decided by those red states.

Oil Prices: The Variable That Neither Side Can Buy With Ads

Goldman’s final conclusion points to a clear investment logic: in the next 26 days, energy market trends matter more than any poll data.

The energy market currently faces multiple overlapping shocks: the Iran war continues to disrupt the Strait of Hormuz, Iranian state media report another tanker mine incident south of the strait; Trump declared on Truth Social, "There will be no action against Iran before the midterms," causing oil prices to dip briefly before rebounding on the mine reports; Hurricane Isaiah has halted 63% of U.S. Gulf offshore oil production, with Jefferies warning this is "a fuel supply shock layered on an already tight system because of the Iran war"—diesel prices are now at $6.30 per gallon versus $3.68 a year ago.

Goldman’s logic chain boils down to this: if the charted correlation between gasoline prices and the popular vote holds, then what the White House really needs is a tank of cheap gasoline, not another round of ads.

The Democratic bullish logic includes: a high and widening popular vote lead, mail-in ballots skewing blue, multiple market paths to 51 seats, and another spike in gasoline prices. The Republican bullish logic includes: a near 2-to-1 spending advantage (albeit late), the historical pattern of opposition sweep probability narrowing by 5 points before election day, and the tendency of red state voters to "come home" late in the cycle.

The ultimate swing variables are things neither side can buy with advertising: whether oil and gasoline prices rise or fall on November 3, Election Day.