What actually works — and what's hype
No regurgitated "what is RSI" filler. Every article here is backed by our own 660,005 out-of-sample backtests, so you get the honest answer — including when the answer is "it doesn't work."
Do Trading Indicators Actually Work? We Ran 660,005 Backtests
We backtested 382 indicators across 903 assets. The honest answer is more interesting than 'yes' or 'no.'
Read itThe most seductive number in trading is win rate. Our backtests show why it's also the most misleading.
Three concepts separate honest indicator research from a curve-fit sales pitch.
If markets were pure noise, nothing would beat holding. If they were easy, everyone would win.
A strategy that looks bulletproof on history and dies on live money isn't unlucky — it's usually a selection…
We backtested 382 indicators across our crypto universe. Here's what actually won most often.
We backtested 382 indicators across our forex universe. Here's what actually won most often.
We backtested 382 indicators across our stock universe. Here's what actually won most often.
RSI is everywhere on trading YouTube. So we tested it honestly — here's what 1,827 backtests say.
MACD is everywhere on trading YouTube. So we tested it honestly — here's what 1,815 backtests say.
Supertrend is everywhere on trading YouTube. So we tested it honestly — here's what 1,670 backtests say.
Bollinger Bands is everywhere on trading YouTube. So we tested it honestly — here's what 1,712 backtests say.
the Stochastic Oscillator is everywhere on trading YouTube. So we tested it honestly — here's what…
Ichimoku Cloud is everywhere on trading YouTube. So we tested it honestly — here's what 1,706 backtests say.
Moving Average Crossovers is everywhere on trading YouTube. So we tested it honestly — here's what…
WaveTrend is everywhere on trading YouTube. So we tested it honestly — here's what 1,648 backtests say.
Order blocks, liquidity grabs, fair value gaps — SMC/ICT is the loudest thing on trading YouTube.
Stan Weinstein's relative-strength chart is back in circulation — here's what the evidence actually says…
The stat is everywhere and a source is nowhere. Here's what real data does and doesn't say about how many…
Screenshots of life-changing gains drive trading YouTube. Here's how to read them honestly — and what our…
A chart went round with six green-to-red winning legs drawn on it and no numbers.
A bot's 15-day results table added up perfectly: every figure in the profit column was real and the total was…
A backtest panel showing +61.20% in two months is the most checkable thing in trading marketing, because it…
A rising cumulative-profit curve is the most persuasive and least informative artifact in trading.
A compound annual return is an average, and averages hide concentration.
Every indicator creator says their strategy works on all timeframes — 660,005 backtests show why that answer…
Adding a second indicator sounds like risk management — but across 660,005 backtests, most combinations…
660,005 backtests later, here is what happens when you put YouTube's favorite indicators through an honest…
Your strategy survived five years of history — here is why that still tells you almost nothing about what…
660,005 backtests across 903 assets reveal an uncomfortable gap between high win rates and real edge on…
We ran 660,005 out-of-sample backtests across 903 assets to find out whether a trend filter converts a losing…
Most signals are designed for momentum. Here's what to reach for when the market isn't going anywhere.
The 'avoid high VIX' rule is trading gospel—here's what 660,005 out-of-sample backtests across 903 assets…
Where to put your stop, when to close, and how to stop giving back open profits — grounded in…
Entry gets you in—exit discipline determines whether you keep what you earn.
Across 660,005 backtests on 903 assets, lagging oscillators post high win rates but rarely beat buy-and-hold…
VWAP is on every trader's screen, but the kiss-back retest narrative and mean-reversion edge don't always…
Across 660,005 out-of-sample backtests, ROC keeps surfacing as a top performer while far more elaborate tools…
A win rate above 69% sounds like an edge — until you check whether the strategy actually outperforms holding.
660,005 out-of-sample backtests across 903 assets reveal where the world's most-quoted moving average earns…
We ran 660,005 backtests across 903 assets — here is what actually happens when you replace the 200 EMA with…
Smoothness rankings are everywhere — but after 660,005 out-of-sample backtests across 903 assets, the moving…
Laguerre RSI ranks among the top commodity indicators in our 660,005-backtest dataset, but its edge is…
A model that sounds like quant gatekeeping reduces to one idea — markets behave differently depending on…
Across 660,005 out-of-sample backtests, one obscure oscillator won the ETF category more times than any other…
Across 660,005 backtests on 903 assets, Fisher Transform came out best on 17 forex pairs — more than five…
KDJ adds a third line to the classic Stochastic, but popularity on crypto dashboards is not the same as edge…
Stop clustering near key levels is real and documented — but the peer-reviewed literature stops well short of…
Intraday volatility clustering around session opens is real and documented — the leap to a profitable timing…
Gap-fill research exists — but it does not say what most FVG traders think it says.
HMA is marketed as the moving average that eliminates lag — our 660,005-backtest database puts that claim in…
Six EMAs look more authoritative than one — the backtest data says otherwise.
HMA promises to eliminate lag. Here's what 660,005 backtests across 903 assets actually found.
Get the weekly edge report
The best-performing indicator per asset, what changed this week, and the honest caveats — straight to your inbox.