Break of structure's honest cousin beat buy & hold on 11 of 47 famous assets
A break of structure that truly beats buy & hold would be worth a great deal — the trouble is that the drawn version is almost impossible to pin down, because the swing high or low it 'breaks' can still shift as later bars arrive. (The idea is simple: a close beyond a prior swing extreme, read as the trend shifting or carrying on.) So the honest thing to test is the half that stays put — the Donchian breakout: a close above the highest high of a set number of already-closed bars, a line fixed the moment those bars closed. It is the first Smart Money Concept we've backtested this way, with more on the way and the fair value gap already covered. The method is the same as everywhere here: the level comes only from closed bars, every fill pays fees and slippage, and every result is set against simply buying and holding — no kind window, no quietly dropped losers. Below is the whole glass: every famous asset, every winner and every loser.
So — does break of structure work?
On the honest evidence: the drawn break can't be pinned down, and the version that can mostly trails holding. Of the 47 famous assets with enough history and trades to judge, the Donchian breakout — the non-repainting stand-in for a break of structure — beat buy & hold on 11, which is 23%. A trend-following rule is expected to catch some of the famous names that spent the window in one long climb — but catching a trend once it is under way is the asset's doing and the era's, never a property of a level drawn by hand, so the burden of proof stays on the break, and one tidy chart is not it. That sits below the 26% we measure across every strategy on every asset — so even the honest, non-repainting version clears buy & hold less often than a rule picked at random. The middle of the pack sits here too: the median Donchian breakout run came in at -44 pts versus buy & hold over the window.
These are famous names that still trade today. Assets that delisted, went bankrupt or (for crypto) died or were rug-pulled aren't in the set (survivorship bias) — and that history isn't available from the free public data this tool runs on — so this rate isn't a representative base rate for every asset that has ever traded.
Where it ran, where it whipsawed
Averages hide the real story here; the spread tells it. One breakout rule, held fixed, goes from clearing buy & hold by a wide margin on a sustained trend to bleeding out in false break after false break on a choppy name — the only thing that changed is the asset.
Avalanche — the breakout cleared buy & hold by +176 pts here, riding a single long trend from near its start and sitting through the pullbacks that would have shaken out a quicker rule. As much a statement about that asset's run as about the breakout.
Nvidia — the breakout came in -871 pts behind buy & hold. In a choppy, sideways market, price pokes above the lookback high and turns straight back down; the rule buys each false break near the top and is stopped out near the bottom, so a range feeds it one whipsaw after another.
That swing from a clean run to a wipeout is the real answer to whether a break of structure means anything: the outcome rides the trend that happens to follow, not the break itself. Clearing a prior high pays off when a long move comes next and bleeds when the move was a feint — and the past can't say which is coming. So where the cousin beats, it is borrowing the asset's trend; it is never evidence that a hand-drawn structure line predicts anything.
Why a break of structure looks like it always marks the turn
- • The example always shows the one swing whose break led straight into a clean trend. But which swing counted was obvious only after that trend arrived — picking the pivot with the move already on the screen is hindsight, not evidence. The whole glass above is what is left when the pivot can't be chosen after the fact.
- • The breaks that failed get quietly re-labelled. When price clears a high and immediately reverses, chart-readers tend to decide the 'real' structure was a deeper swing all along, so the eye only ever meets breaks that led somewhere — the same survivorship trick, moved from deleting a losing trade to redrawing which level mattered. The backtest counts every break the fixed rule ever took, the failed ones included.
- • On a replayed chart the swing looks like it was always there, so the entry seems obvious and early. In real time it isn't: a high only becomes 'the' swing once enough later bars fail to beat it, so the break a tutorial marks wasn't knowable at the candle where it appears. That gap between the replay and the real clock is exactly what a closed-bar rule removes — and what it costs in lag.
Why the drawn break can't be backtested the same way twice
The break of structure is one idea inside Smart Money Concepts, and it has an honest half and a fragile one. The fragile half is the drawing; the honest half is whatever can be fixed by a rule and read back identically from bars that have already closed. Here is where the line falls, and the testable rule that stands in for it.
A break of structure is drawn from a swing pivot — a high or low flanked by a few bars that don't exceed it. The catch is that a pivot isn't settled when it forms: a later, higher bar can promote a fresh high and demote the one that looked decisive, and changing how many bars a pivot must clear on each side selects different pivots altogether. So the 'structure' a break clears keeps getting redrawn as the chart fills in — the level slides, or it only becomes legible well after the bar where a tutorial later marks it. Graded by whether it can ever be shown wrong, the drawn break fails on repeatability long before performance enters the picture: the same history, read twice, can hand back two different break levels. Measuring it would mean measuring one analyst's choice of pivot on one pass through a chart, not a rule the data can reproduce.
The break keeps its honest half the moment the 'prior extreme' is set by a rule instead of by eye. The Donchian breakout does exactly that: it takes the highest high of a set number of already-closed bars as the level, and treats a close above it as the break. It is still the same plain idea — price pushing past where it last turned — but the level was fixed when those bars closed, so it cannot slide to flatter what happened next. That is the version on the scoreboard above: fully testable, and the live figures show exactly how often it clears buy & hold.
Deterministic here carries exactly one meaning: the rule reads the same level from the same bars on every pass, so it can be measured without fooling anyone. It does not mean the break foresees the next move — the scoreboard above shows how seldom it clears buy & hold. A fixed, testable rule and a rule worth risking money on are different claims, and only the first is on offer here.
This boundary is not fixed. Any concept that can be written as a rule over closed bars — one that returns the same level on every pass — can cross from the drawn side to the tested side, exactly as the break of structure did here. Naming it is never a pledge to stop adding concepts, and never a hint that a concept already tested beats simply holding.
Every Donchian breakout backtest is public — every asset that ran and every one that whipsawed, nothing hidden. Read the whole scoreboard, then run the breakout on a chosen asset and see honestly which side of buy & hold it lands on.
Figures are computed live from simulated backtests over roughly the last five years and update as new bars close. 'Break of structure' is tested here through the Donchian breakout — a close above the highest high of the last 20 closed bars for entry, a close below the lowest low of the last 10 for exit, both windows excluding the current bar so nothing peeks ahead; other lookbacks behave differently, which is the point, and any of them can be tested in the sandbox. A breakout rule trades the whipsaws, so fees and slippage on every false break are counted in. Hypothetical / simulated results — Past performance is not a reliable indicator of future results. This is an educational, analytical tool — not investment advice, and not a recommendation to buy or sell anything. You make all decisions and execute on your own broker.