We are tilting the models into November — and the more important half of that sentence is that we are tilting them back.

A six-point spread is the honest headline

The midterm pattern is better documented than most seasonal market folklore, and we are not going to talk you out of it. But when the average answer moves six percentage points depending on which decade you start counting from, you are looking at a genuine tendency measured on a thin sample — not a rule you should reposition a retirement account around.

What holds up better than the number is the mechanism. Uncertainty peaks before the vote and drains away once the outcome is known, whatever the outcome is. Policy stops being a coin flip and attention goes back to earnings. That is why the effect appears no matter which party takes the House — and it is the honest answer when a client asks what happens if the other side wins.

Which leads somewhere most people find counterintuitive: the right response to an election year is not to get defensive. A retirement account with a decade or more left to run is a buyer, not a seller. Volatility between now and November is not a threat to it — it is the mechanism by which every paycheck contribution buys more shares than the last one. Moving to cash converts a temporary price into a permanent loss.

So the tilt is not about lowering risk. It is about making the ride holdable: quality and dividends over speculation, a few genuinely defensive sectors, and short duration on the bond side so that the safe sleeve is not itself the risk. The whole design goal is that the participant does not sell in October.

Uncertainty peaks before the vote and drains away once the outcome is known — whichever way it goes. That is the whole mechanism, and it is why the effect does not care who wins.

The mechanism

Every forecast has an expiry date. The one durable finding is that resolution matters more than result.

And because it is a posture rather than a portfolio, it comes with a date and an exit. The tilt below is written to be reversed once the vote resolves — which is the half of this that takes actual discipline, and the half we spend the most time on.

In a taxable account, the constraint is tax, not availability. Everything is buyable, but a tilt executed by selling appreciated positions can cost more in realised gains than the tilt is worth. So it gets built with contributions and dividends first, and only then with sales — which means it phases in over weeks rather than landing in a day.

In an IRA or a rollover, there is no friction at all. Full access to every instrument, and rebalancing carries no tax consequence, so the posture goes on exactly as designed and comes off the same way. This is where the model is expressed in its purest form, and it is the version the other two are approximations of.

In a workplace plan, the menu decides what is possible. On the plan list above, only 2,561 of 3,572 funds were open, every ETF was restricted, there was no money market fund, no total-bond index and no small-cap index. The tilt still gets built — but from mutual funds only, with cash living in short Treasuries and small-cap exposure reached through mid-caps. It is the same posture wearing whatever the plan happens to stock.

Constraints do not change the thesis. They change the instruments — and knowing which instrument a constrained account can actually use is most of the work.

Core US, 38%. You do not time an election with the core. Nearly forty points sit in plain index exposure and stay there through November regardless of what the polls say. The one active decision inside it is a value tilt — value has carried a real earnings-yield advantage while rates stayed high, and it holds up better than growth when a multiple compresses, which is what a volatility event actually is.

Dividend and quality, 16%. A dividend-appreciation screen is a quality filter wearing an income costume: it selects companies that have raised payouts for a decade, which is why it falls less in drawdowns. Inside any tax-deferred account those dividends compound untaxed, so the usual tax objection to an income tilt does not apply — and in a taxable account we site the sleeve deliberately for the same reason.

Defensive sectors, 16%. Staples and utilities are the classic pair, and utilities have a second job now as the least speculative way to own AI power demand. Health care is the genuinely debatable one — drug pricing is live political material in an election year, so the headline risk is real. But that risk is precisely why the sector is cheap, and it has already begun to turn. We hold it, and we would not defend it past a small weight.

International, 11%, and real assets, 7%. Developed markets have outrun the S&P over the past year on a cheaper starting multiple and a softer dollar. Energy and real estate are in the book for what November will not settle: deficits, and what they eventually do to the price level.

Short bonds, 12%. With no money market fund on the menu, short-duration bonds and short TIPS are the closest thing to dry powder. Their job is to be worth what they say they are worth in November. Reaching for yield here would make the safe sleeve the risky one — which is the mistake that hurt people in 2022.

How the reversal actually runs

It is staged, not a single trade. The defensive sector overweights come down first, because they are the most explicitly tactical part of the book and the cheapest to be wrong about. The short-duration bond sleeve is next, and it moves back toward the standard duration as the reason for holding cash-like assets disappears. The dividend and quality tilt comes down last and least — some of it is a permanent preference rather than an election trade, and in a tax-deferred account the untaxed compounding argument for it holds in every year.

The core never moved, which is the point. Roughly forty percent of the book sat in plain index exposure through the whole episode precisely so that being wrong about the election could not do real damage. The tilt is expressed at the edges, on purpose, because that is where a mistake is survivable.

And if the reversal turns out to be early or late, it is early or late on sixteen points of sector weight rather than on the whole portfolio. The size of the bet is itself a risk decision, and it is one made before the trade rather than after it goes against you.