Downside Protection ETFs: Four Mechanisms, What Each Absorbed, and Where Each Fails

A downside protection ETF uses a written rule or an options contract to absorb part of a market loss. Four mechanisms do the work. A contractual buffer. A signal rule that cuts exposure. A low volatility screen. An options overlay that sells upside for premium.

All four work. All four send a bill, and the bill is the part advisors skip. This page measures the bill.

What is a downside protection ETF?

It is a fund built so a decline in its benchmark does not pass through in full. The structure is disclosed in advance. That is the whole product.

The name oversells it. Only the contractual buffer protects a defined amount. The other three mitigate, which means they seek to reduce a loss by an amount nobody can promise beforehand. Read "protection" as "mitigation" and the category gets easier to compare.

Each is a building block, not a portfolio. See risk-managed ETF model portfolios for how they get assembled.

What does systematic downside mitigation actually mean?

It means a written rule decides, not a portfolio manager on a bad morning. The rule exists before the decline starts. It fires on the same input every time.

That's the whole claim. It is a narrower claim than most marketing makes it sound. A rule does not know a decline is coming. It only knows what already happened.

So the useful diligence question isn't whether a fund has a rule. Every fund in this category has one. The question is what the rule costs when it's right and what it costs when it's wrong. More on that in systematic versus discretionary investing.

How does a buffer ETF cut a decline?

By contract. A buffer ETF uses options to absorb a fixed slice of loss. The outcome period is usually one year. A common structure absorbs the first 9% to 15%.

The math is deterministic. With a 10% buffer, an index that falls 25% leaves the holder down roughly 15% before fees. Not 25%. Not zero.

You pay for that with a cap. The upside is sold off to fund the buffer, and the cap resets when the period resets. Buy one mid-period and you inherit whatever buffer is left, which may be almost none.

How does a signal rule cut a decline?

By leaving. A signal-based fund reduces equity exposure when a trend or volatility measure crosses a line. Some rotate to defensive sectors. Some go to cash or short-term Treasuries.

This is the only mechanism that can remove market exposure entirely. It is also the only one that has to be right twice. Once to exit and once to come back.

The lag is structural, not a flaw in a particular fund. A trend rule needs a confirmed move before it fires. The move it needs is the first part of the drawdown. A trend following ETF and a volatility-targeting sleeve both carry that lag.

How does a low volatility screen cut a decline?

By owning steadier stocks. A low volatility ETF ranks the universe on realized or forecast volatility and holds the calm end. Utilities, staples, and insurers usually dominate.

It stays fully invested. Beta drops, it does not go to zero. In a broad selloff a low volatility sleeve falls less than the index and it still falls.

The screen also looks backward. It ranks on volatility measured in a window that did not contain the shock you're worried about.

How does an options income overlay cut a decline?

By collecting premium. A covered call fund sells calls against its holdings and keeps the cash. That premium cushions a decline by the amount collected.

In a mild drawdown that helps. In a 25% drawdown a few points of premium is a rounding error against the loss. And the calls cap the recovery on the way back up.

Can a covered call ETF give downside protection?

A few points of it, in a slow decline. Premium income is the entire cushion. Collect 1% a month and you have absorbed about 12 points in a year.

In 2020 the market did not give you a year. The index fell 33.9% in 23 trading days, and the written calls then capped the rebound. Treat it as income with a small side benefit.

What happened to each mechanism in 2022?

Use real dates. The S&P 500 closed at 4,796.56 on January 3, 2022. It closed at 3,577.03 on October 12, 2022. That's a 25.4% peak-to-trough decline spread across roughly nine months.

A 10% buffer running the full period absorbed 10 points of that and left about 15. A signal rule on the 200-day average triggered in late January 2022. By then the index was already down about 8%. A low volatility screen stayed invested the whole way. An options overlay collected premium into a market that kept grinding lower.

Now run 2020 through the same rules. The index fell from 3,386.15 on February 19 to 2,237.40 on March 23. That's 33.9% in 23 trading days. The 200-day average broke on February 27, with the index already down 12%.

Same rule. Same mechanism. Two very different outcomes, because the shape of the decline changed.

How much of each decline did each mechanism absorb?

We measured it. The signal rule is a 200-day moving average on the S&P 500. Long or in cash, acting on the close after the cross. The low volatility figure is the largest low volatility screen on US large caps, on adjusted closes. The buffer row is contract arithmetic, not a fund result.

Read the low volatility row twice. In the fast 2020 crash the calm-stock screen fell more than the market it was meant to cushion. In the slow 2022 decline the same screen absorbed nearly 40% of the loss.

The signal rule did the opposite. Over calendar 2022 it finished level with buy and hold, so it bought nothing. In 2008 it finished down 2.5 percent against the market's 36.8 percent.

That is the finding. Each mechanism is matched to a shape of decline, not to risk in general. Fast crash, slow grind, choppy flat year. Pick the wrong shape and you paid for nothing.

Past performance does not indicate future results. These measure generic rules on index data, not any THOR fund.

What does a signal rule cost in the years it is wrong?

This is our own mechanism, so here is its bill. We ran the 200-day rule on the S&P 500 from November 1993 to December 2025. Just over 32 years.

The rule cut the worst drawdown by more than half. It gave up about 1.9 points of growth a year to do it. That is the trade.

The cost concentrates in flat, choppy years. We counted seven where buy and hold finished between minus 10% and plus 10%. In those the rule fired 13 times on average and lagged by 5.6 points.

• 2007. Rule 12.2 points behind, 17 signals.
• 1994. Rule 11.3 points behind, 14 signals.
• 2015. Rule 10.1 points behind, 8 signals.

Nobody markets that list. It is the honest price of the only mechanism that can go fully to cash.

Is there an ETF with 100% downside protection?

There are structures that aim at it. A fully protected ETF holds instruments that return your principal at the end of the outcome period. It spends the interest on a call option for upside.

The cap is funded by that interest. Nothing else pays for it. Which is why the category is new. The 13-week Treasury bill yielded 0.05% in January 2021, so there was no cap worth selling. By January 2023 it yielded 4.57%.

Three limits before anyone calls it free. You give up every dividend. The protection holds only if you bought at the reset and stay the full period. And the cap is low by construction.

What about an S&P 500 ETF with downside protection?

Most of these funds are built on S&P 500 exposure already. Buffers are written on it because its options are the deepest. Signal rules are tested on it because its history is the longest. Low volatility screens rank inside it.

That matters practically. Stack three of these on one benchmark and you have diversified nothing. You own the same index three times, with three overlays and three fees. Want a different bet? Managed futures and sector rotation read inputs these overlays never see.

When does each mechanism fail?

Buffers fail on timing. The buffer is measured from the start of the outcome period, not from your purchase date. Buy after the fund has run most of the way to its cap and little upside is left. You can also give back your own gains before the buffer engages at all.

Signal rules fail on whipsaw. A sharp drop that reverses inside a few weeks costs you the exit and the re-entry. Repeat that three times in a flat year and the drag is real. About 5.6 points, measured above. February and March 2020 punished slow rules on the way down. Then again on the way back up.

Low volatility screens fail when correlations converge. In a liquidity event investors sell what they can, not what they want to. Defensive names get sold with everything else. That is not a theory. In early 2020 the screen fell 35.8% while the index fell 33.7%.

Options overlays fail in a fast crash. Premium is small against a 30% move. Strikes that looked comfortable last month are far away this month.

None of that argues for holding the index and hoping. It argues for knowing which failure you signed up for. The same logic drives sequence of returns risk for a client drawing income. And capital preservation for one who cannot rebuild.

How should an advisor choose a downside protection ETF?

No single fund is best for every client. Any list that claims one is selling something. The choice follows the decline you are defending against.

Four questions get you most of the way.

• What is the trigger? A contract, a moving average, a volatility rank, or an option strike.
• How fast can it act? Daily, monthly, or only at the end of an outcome period.
• Where does the money go? Cash, Treasuries, defensive equity, or nowhere at all.
• What does being wrong cost? A capped rally, a whipsaw round trip, or a partial cushion.

Then match the answer to the shape. A buffer suits a client who needs a known floor over a known year. A signal rule suits one who fears a long bear market more than a choppy one. A low volatility screen suits one who will stay invested anyway. See also rules-based risk management platforms for RIAs, and tactical asset allocation as the industry calls it.

THOR runs the signal side of that list. THOR AdaptiveRisk Dynamic and THOR Index Rotation both cut equity exposure on a rules-based signal. Both seek to reduce drawdowns that way. Whipsaw is the bill we agree to pay for it. The 32-year table above is roughly that bill.

Definitions

• Downside mitigation. Any structure that seeks to reduce loss in a falling market. Often written as downside protection, which overstates it.
• Buffer. A contractual slice of loss absorbed by an options structure over a defined outcome period.
• Cap. The maximum return over that outcome period, sold to pay for the buffer.
• Outcome period. The fixed window, usually twelve months, over which a buffer and cap apply.
• Full protection structure. A fund that returns principal over an outcome period and funds its cap from interest on that principal.
• Whipsaw. An exit signal followed quickly by a re-entry signal, where both trades lose money.
• Downside capture. The share of a market decline a strategy actually took. Below 100% helped. Above 100% made it worse.
• Beta. Sensitivity to the broad market. A beta of 0.7 falls about 7% when the market falls 10%.
• Signal rule. A pre-written condition that changes exposure when a market measure crosses a threshold.

Talk it through

THOR builds adaptive index strategies for advisers and allocators. Fifteen minutes with our founder and CIO gets you the mechanism, the index history, and an honest read on where it fits. Book a time.