Independent strategy research
Is momentum trading profitable? We tested 140+ strategies.
Momentum is one of the least-bad categories we've put through the audit — and it still failed about 73% of the time. We ran 140-plus momentum strategies through the same real-cost model we use on everything, on 13 years of data, and roughly three in four were rejected once spreads and slippage were in. That's the optimistic reading. Grid and DCA bots die at a clean 100%; momentum at least leaves a few standing. Around one in five kept an edge. The rest were carried by a handful of big days, or never had an edge at all.
What counts as momentum — and why traders love it
Momentum bets that what's already moving keeps moving: buy strength, sell weakness, ride the run. On TradingView it shows up as trend-following crossovers, breakout-continuation systems, relative-strength rotations, and every variant of "the trend is your friend" written in Pine. It's popular for a simple reason. When it's right it's spectacularly right, and the winning trades feel obvious in hindsight.
That's also the trap. A momentum curve looks convincing precisely when it's most fragile, because a small number of monster days do most of the work. Take those days away and a lot of these systems are flat. We measured how many.
What actually happened when we tested 140+ of them
Roughly 73% were rejected once we modeled real spreads and slippage. Compared with the rest of the audit, that's a decent showing. Breakout systems failed at a higher rate. So did reversal, mean-reversion, and session-time strategies. Momentum sits among the categories with the best survival odds we've measured — which tells you how brutal the base rate is, because "best odds" still means about three in four don't make it.
So momentum is not a scam the way a martingale is. It's a real market phenomenon that most published implementations fail to capture after costs. Different problem, same ending for the reader who deployed one blind.
Why most momentum strategies die
Two failure modes do most of the damage. The first is the boring one: no real edge before a cent of cost. Across the whole audit, that's the single most common reason a strategy gets rejected — the system doesn't predict anything, it just rode the market's own drift and called it skill. Strip the beta out and there's nothing left to pay the spread with.
The second is the one that fools careful people: tail-concentration. The equity curve is real, the total return is real, but a few outsized days carry the entire thing. That's the chart above. Miss those sessions — a slightly worse fill, a different start date, one gap you didn't get filled on — and the edge collapses. Tail-concentration is one of the top few rejection reasons in our data, and momentum is where it lives.
Costs finish the job. A momentum system trades often, and every entry pays the spread. When the raw edge is thin, that alone is cost-fatal. And the win rate won't warn you — plenty of these show a respectable hit rate while the after-cost expectancy is negative, because the wins are small and the misses on the trend days are large.
One more thing specific to this family. Momentum systems lean on trailing stops, and a bar-by-bar backtest — the kind TradingView runs by default — overstates trailing-stop strategies by 42 to 84 percent in our tests. Switch to realistic intrabar fills and a chunk of the "edge" evaporates. That gap is the product a lot of sellers are quietly shipping.
The honest exception: the minority that survived
Not all of them failed, and we won't pretend otherwise. Around one in five momentum strategies held a conditional edge — meaning it survived costs, but only on a specific instrument, a specific timeframe, or a specific regime, not everywhere at once. That word conditional is doing real work. It's the difference between "this can be one sized-down piece of a book" and "this is a system you switch on and walk away from."
Genuinely deploy-anywhere momentum is rarer still. Across the entire audit — every category, thousands of strategies and indicators — under 1% cleared the bar to be called deployable outright. Momentum contributes some of those. It is not a large number, and it never should have been.
Grid and DCA can't say even that. If you want to see the category that fails at a full 100%, we wrote that one up too: do grid and DCA bots actually work. More of our by-type findings live in the research hub.
How we test
Every strategy is ported to Python and run against real costs — spreads and commissions modeled from tick data, not a guessed flat number. Futures come from Databento, 13 years of CME. FX comes from Dukascopy with real bid/ask. Stocks get liquidity-aware fills; crypto runs as spot and perps. A fast model does the bulk porting, then the strongest model tries to break every apparent winner, hunting look-ahead bias and impossible fills, with an independent supervisor watching the pipeline around the clock. We hash the code, so a momentum script re-published under three names gets tested once, not three times. It's the same process that rejects roughly 78% of everything we test — and about 73% of the momentum strategies specifically.
Research and education, not financial advice. No signals, no return promises. Independent, and not affiliated with TradingView.
Which momentum strategies survived — and which didn't?
You now have the aggregate truth: roughly 3 in 4 momentum strategies fail after costs, and the ones that live are usually conditional. What this page doesn't give you is the names. Which specific published momentum scripts we tested, who wrote them, and the exact after-cost numbers behind every verdict — that's The No List. Every strategy we audited, named, with the reason it lived or died.
Get The No List →FAQ
Is momentum trading profitable?
Sometimes, and less often than the charts suggest. About 73% of the 140-plus momentum strategies we audited failed a real cost model. Roughly one in five kept a conditional edge — profitable on a specific instrument or timeframe, not everywhere.
Why do most momentum strategies fail after costs?
Two reasons dominate. Many have no real edge before costs and simply ride the market's drift. Others are tail-concentrated — a few big days carry the whole curve, and a slightly worse fill erases the edge. Frequent trading and the spread finish it.
Is momentum better than grid, mean-reversion or breakout strategies?
By survival rate, yes. Momentum's ~73% rejection rate is lower than breakout, reversal, mean-reversion and session-time in our audit, and far below grid/DCA, which failed at 100%. It's one of the least-bad categories — which still means most fail.
Do any momentum strategies actually work?
A minority do, conditionally — around one in five held an edge after costs, usually on a specific market or timeframe. Genuinely deploy-anywhere momentum is rare; under 1% of everything we tested cleared that bar outright.