We backtested MACD honestly — it beat buy & hold on 12 of 47 famous assets
I ran the classic MACD strategy — go long when the MACD line (the gap between a fast 12-day and a slow 26-day moving average) crosses above its 9-day signal line, and step back to cash when it falls below — measured exactly the way the rest of the site measures: every fill is charged fees and slippage, no signal acts until its bar has closed, and the benchmark is always just buying the same asset and holding it. Nothing is cherry-picked and no window was hand-chosen. MACD is a trend rule — the mirror image of RSI, trying to ride a move rather than fade it. What follows is the whole glass: every famous asset, every winner and every loser.
So — does MACD work?
On the honest evidence: mostly, no. Across the 47 famous assets with enough trades and history to judge, the MACD rule beat buy & hold on 12 of them — 26%, so 35 of the 47 did better simply held. That's a shade above the 25% we measure across every rule — but read that carefully. These are famous names that mostly trended over the window, and a trend-following rule flatters itself on a trend, so a slightly higher rate here is a fact about this survivorship-selected panel, not evidence that MACD “works”. The typical MACD run didn't win either: its median result came in at -58 pts versus buy & hold over the window. A handful of real winners exist — that's exactly why the question is worth asking honestly — but a beat on this panel is inflated by survivorship, by a mostly-trending set of names, and by the many variants you could have tried; a loss counts against MACD in a way a win does not count for it. The burden of proof is on the rule, and “it looked great on one chart” is not that proof.
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 shone, where it died
Forget the headline average for a moment — the honest signal is in the spread. One unchanged MACD rule turns into a triumph on one asset and a wreck on the next, decided entirely by which asset you happened to point it at.
Avalanche — MACD beat buy & hold by +202 pts here. A trend rule looks brilliant on an asset that trended cleanly and then dropped hard: MACD rode the move and its cross flipped it to cash before the worst of the fall. That is as much a statement about that asset's trend as about MACD's skill — you couldn't have known in advance which asset would oblige.
Nvidia — MACD came in -842 pts behind buy & hold. In a choppy, sideways stretch the MACD line crosses its signal back and forth: the rule buys late near local highs and sells late near local lows, gets chopped up on the whipsaws, and pays costs on every one — the classic way a trend rule bleeds when there is no trend.
That inconsistency is the answer to “does MACD work”: the outcome belongs to the asset and its regime, not to the indicator. MACD can help on something that trends cleanly and hurt badly on something that chops sideways — and nothing in the past tells you which future you're about to get.
Why MACD looks like it always works
- • You get shown the one chart where the crosses lined up. Choosing the asset and the dates after you already know MACD looked good there is hindsight dressed up as proof. The whole glass above is what's left when you can't choose.
- • The costs are usually hidden. MACD flips more often than a plain moving-average crossover — that's the point of the faster signal — so fees and slippage on every entry and exit bite harder. The friction was always real; most demos just leave it out.
- • A trend rule needs a trend. MACD's core assumption — that a move keeps going — simply fails in a sideways range, where it whipsaws in and out for a string of small losses, and by design it lags every real turn because it is built on smoothed averages. The same failure shows up on every choppy worst-case asset.
Nothing here is filtered — every MACD run is public, the wins sitting right next to the wrecks. Read the full MACD scoreboard, then point the rule at the asset you actually care about and watch which side of the line it lands on.
Figures are computed live from simulated backtests over roughly the last five years and update as new bars close. “MACD” here is the standard 12/26/9 rule, going long when the MACD line is above its signal line and flat when below; other settings behave differently, which is the point — you can test any of them yourself. 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.