The Honest Quant.
Open the Lab

Does the 8/21 EMA crossover actually work?

A computed answer — not an opinion. Here's what nearly two decades of honest data say.

58OUT OF 100
WEAK — PROBABLY OVERFIT

More likely curve-fitting and luck than a durable pattern.

The honest verdict

We ran the 8/21 EMA crossover through an honest backtest — nearly two decades of survivorship-free NIFTY & BANKNIFTY data, out-of-sample, with real costs. The verdict: 58.0/100 — weak — probably overfit. More likely curve-fitting and luck than a durable pattern.

Why it scored 58

Here's where it breaks:

How we tested it

Five statistical tests, no mercy: out-of-sample decay (is the edge curve-fit?), Monte-Carlo robustness (is it just luck?), real cost & slippage survival, survivorship-bias exposure, and sample-size adequacy. Run on NIFTY 50 and BANKNIFTY — survivorship-free by construction, the most honest data available — across roughly two decades. Rules are read at each day's close and filled the next open, so the strategy can never peek at the bar it trades on.

58.0/100
Honesty score
144
Trades tested
49%
Win rate
NIFTY/BANKNIFTY · ~2 decades
Data

See the full breakdown — free

Run the 8/21 EMA crossover live, pick your market, and get its honest score plus all five tests in about 10 seconds — no signup needed to see the score.

Run the 8/21 EMA crossover now →

Frequently asked

Is the 8/21 EMA crossover profitable?

On a true, survivorship-free test (58.0/100, weak — probably overfit), the 8/21 EMA crossover is weak and probably overfit. A backtest looking profitable is not the same as a strategy being real — most aren't. Test your own version free at The Honest Quant.

Does the 8/21 EMA crossover work on the NIFTY?

This verdict was computed on NIFTY 50 and BANKNIFTY index data (survivorship-free) over nearly two decades. It scored 58.0/100 — weak — probably overfit.

Is the 8/21 EMA crossover overfit?

Our overfitting test checks how much of the edge survives out-of-sample. See the full breakdown — and run your own parameters — in the lab.

Related strategies, honestly tested

buying when RSI drops below 30 · the Golden Cross · the SuperTrend indicator · the MACD crossover · buying the lower Bollinger Band · see all →

Get the monthly honest teardown

One popular strategy put on trial, in your inbox — once a month. No signals, no hype.

Educational, not financial advice. An honest backtest score measures whether a result survives statistical scrutiny — it does not predict future profit, and is not a buy or sell signal. · About · FAQ · All strategies