Edge Lab / Clenow / Briefing

The Shock Absorber

A weekly momentum strategy that captured most of the index’s return at roughly half the worst-case drawdown. Twenty-one years, real data, on the Nasdaq-100 and S&P 500.

OverviewDeep Dive →

Executive summary

Buy-and-hold is a wonderful strategy right up until it isn’t. Across the last two decades, simply owning the index compounded beautifully — and twice asked investors to stomach a fall of more than half their capital. This strategy answers a narrower, more practical question: how do you stay invested in strong markets without being destroyed by the weak ones?

It does that with a mechanical discipline — ride the strongest-trending large-cap names while the market is healthy, and step aside when it turns. Over 21 years on real, survivorship-bias-free history, it captured most of the index’s return — 11.4%/yr on the Nasdaq-100 and 9.3%/yr on the S&P 500, against the index’s 15.3% and 10.9% — while cutting the worst-case loss roughly in half, with a risk-adjusted return (Sharpe) equal to or better than buy-and-hold. And in the twelve-month stretches when the index fell hardest, this strategy came out ahead of buy-and-hold every single time.

In one line: it earns close to the index’s return with about half the worst-case drawdown — built to keep capital intact through the crashes that wreck long-term compounding.

The true cost — and what you keep

The drawdown-halving effect. Owning the S&P 500 outright meant living through a −55% collapse in 2008. The same dollars run through this strategy saw a worst-case −23%. On the Nasdaq-100 the worst loss shrank from −53% to −30%. That is the product: a structurally shallower bottom.

The trade-off on the top line is now modest. Over the full window buy-and-hold compounded somewhat faster — roughly 11–15% a year versus 9–11% for the strategy — because it was fully exposed during the long bull runs that dominate the record. But the strategy’s risk-adjusted return matched buy-and-hold on the Nasdaq-100 (Sharpe 0.77) and edged it on the S&P 500, and its worst 12-month stretch was roughly half as deep. You give up a slice of bull-market upside; you keep most of the compounding and shed most of the drawdown.

Most capital can’t sit through a 55% decline. Drawdowns force selling, trigger redemptions, and end careers. What this strategy buys is the ability to stay in the game at close to the index’s rate of return.

Growth of $100k: strategy vs. buy & hold
Growth of $100,000, log scale. The strategy (solid) tracks much of buy-and-hold’s climb (dashed) — through far gentler declines.
Drawdowns: strategy vs. index
The same story as pain. The index plunges past −55% in 2008; the strategy’s deepest hole is roughly half as deep.
See if it fits your mandate. We can re-run this on your own universe and risk limits — request a bespoke study →

Where it adds most: the bear-market record

It adds most exactly when you need it to. Sort every 12-month period by how the index did. In every single 12-month window where the index fell more than 20%, this strategy returned more than buy-and-hold — by about 15 points on the Nasdaq-100, and by roughly 29 points on the S&P 500, where it held its loss to single digits through years the index lost a third of its value.

This is the signature of a defensive allocation: it captures most of the index in calm and rising markets and pulls clearly ahead in the downturns. The protection isn’t an accident of one lucky year — it shows up across the whole history, every time the market broke.

12-month returns by market condition
Average 12-month return, strategy vs. buy-and-hold, grouped by how the index itself did. Far left (the index falling hard) is where the strategy adds the most.

How it works, in plain terms

No black box. Three simple ideas:

Methodology is Andreas Clenow’s, from Stocks on the Move. We implemented the weekly version faithfully and tested it without curve-fitting, on split- and dividend-adjusted prices.

Tested the hard way

Real history, including the losers. The test uses point-in-time index membership that includes the companies that later failed, were acquired, or dropped out — so the results aren’t flattered by quietly only trading today’s winners.

Skill, not luck. We re-ran the strategy 200 times with the stock-picking replaced by random selection, holding everything else fixed. The momentum ranking added return above those random versions in every test, landing in the top 1–2% of them — and the trend filter contributes most of the downside protection.

Costs, slippage and cash, stated plainly. We modelled Interactive Brokers commissions — they lower the return by under ~20 bps a year, immaterial. Slippage is modest too: the system trades a handful of liquid large caps, so even a pessimistic fill assumption costs about a point a year at most. The book is near-fully invested, and the small idle balance earns 0% in the headline — for a complete picture, add the true cash return you would earn at your own rate. The Deep Dive shows the full breakdown.

View the full 2005–2026 month-by-month performance

The complete monthly grid for both universes, with every year’s return and intra-year decline, lives on the Deep Dive — along with the random-portfolio stress test, the in-sample/out-of-sample checks, and the full risk-adjusted breakdown.

Who this is for

This suits a portfolio manager whose first job is to protect capital through the cycle — one who wants close to large-cap equity returns without sitting through a 55% drawdown to get them. Used as a defensive equity allocation or a drawdown-aware sleeve, it does a job buy-and-hold cannot. For a pure maximum-growth mandate with a genuine tolerance for 50%+ drawdowns, straight index exposure compounded somewhat faster over this window.

Let’s stress-test this for your mandate

No off-the-shelf backtest matches a real risk tolerance, capital constraint, or regulatory mandate. This study is a baseline demonstration of our research process. If you manage institutional capital, a family-office allocation, or a private portfolio, we can adapt the engine to your exact parameters.

Request a bespoke portfolio study

Or audit it yourself

Want the Python, the daily trade logs, and the data behind every number here? We’ll send the full package.

Request the artifact package & custom study parameters

Typically delivered as an interactive notebook and CSV package within a few business days.

→ Full Deep Dive with the complete methodology, monkey baseline, ablation, and risk-adjusted tables.