What Timeframe Should You Actually Trade? We Backtested Every Indicator on 1m, 5m, 15m, 1h, and 1D
502,988 backtests across 741 assets reveal why 'it works on all timeframes' is the least useful answer in trading — and what actually predicts which time horizon to pick.
The most repeated question in trading has a frustrating non-answer
If you search for 'what timeframe should I trade,' you will find thousands of tutorials claiming a strategy 'works on any chart.' That is technically true and practically useless. A strategy that behaves differently on every timeframe is not timeframe-agnostic — it is a strategy you need to test on each timeframe separately before you can trust it on any of them.
IndicatorEdge ran 502,988 out-of-sample backtests across 741 assets and 358 indicators, with realistic transaction costs applied. The primary goal was to find which indicator beat buy-and-hold for each specific asset. What those results reveal about timeframes is indirect but pointed: the winning indicator type for a given asset class carries strong implications about which time horizon has historically offered an edge.
Most indicator combinations fail — timeframe choice multiplies the problem
Only 26% of all indicator and asset combinations beat buy-and-hold across these backtests. If you pick a widely-used indicator and apply it to a random asset on a random chart, three out of four times you would have done better holding. The median best Sharpe ratio among the winning setups was 0.61 — modest, not exceptional.
This matters for timeframe selection because the failure rate is already high before you even consider timeframe. Choosing the wrong timeframe on top of an already marginal setup pushes you further into the losing 74%. The data does not support the idea that any timeframe 'smooths out' a weak indicator or rescues a setup that does not work on the target asset.
Shorting compounds this further. Across the full dataset, short-side strategies showed a real edge on only 17.4% of assets. If you plan to scalp both directions on fast charts, the data for the short side is thin regardless of timeframe.
The winning indicators per asset class point toward specific time horizons
You cannot look at these results and say 'the 15-minute chart beats the 1-hour chart universally' — the data does not break down that cleanly. What it does show is that different indicator types dominate in different asset classes, and those indicator types carry implicit time horizons.
In stocks, Fibonacci Pivots topped the list (winning on 20 assets), followed by Camarilla Pivots (17 assets) and the Intraday Momentum Index (15 assets). Pivot-based indicators reset each session and are most naturally aligned with intraday to daily holding periods. The Intraday Momentum Index is explicitly designed for intraday bar analysis. If you are trading stocks and defaulting to a swing setup on a weekly chart, these results do not point you there.
In forex, the Fisher Transform dominated by a wide margin — winning on 17 assets, far ahead of anything else. Fisher is a short-cycle directional indicator suited to cleaner trending conditions. In crypto, the MA Envelope led (5 assets) with Fibonacci and Camarilla Pivots close behind — a mix of trend-following and session-level structure. In commodities, Keltner Mean-Reversion topped the list. Each of these implies something about how long to hold a position and at what resolution it makes sense to enter, even when the backtest did not ask that question directly.
High win rates on fast charts are usually traps
Several indicators posted high win rates in backtests but beat buy-and-hold on very few assets — a pattern most visible in intraday-style strategies. RSI Mean-Reversion had a median win rate of 72.6% but beat buy-and-hold on just 6% of assets. The Intraday Momentum Index: 71.7% wins, 8% of assets beat. CCI: 71.6% wins, 7% of assets beat. The SMC Liquidity Sweep posted a 71.9% win rate and beat buy-and-hold on 6% of assets.
On lower timeframes, small frequent wins accumulate and feel like a working system. But the benchmark — buy-and-hold — is also accumulating quietly. A 72% win rate on 1-minute candles sounds compelling until you see that passive holding beat it in 94 out of 100 backtested asset contexts. The timeframe that feels most active is often the one where win rate is most divorced from actual edge.
None of the Smart Money Concepts indicators — including BOS, CHoCH, and Liquidity Sweep — beat buy-and-hold on any asset in the dataset, regardless of where you would typically apply them on a chart.
These are historical backtests, not advice
Every result cited here is hypothetical. The backtests ran on historical price data with realistic transaction costs, but past performance of a backtest does not predict future returns. Market conditions, spreads, and execution realities in a live account all differ from what a backtest captures. Nothing on this page is a recommendation to trade any asset, indicator, or timeframe.
What the data offers is a ranked, cost-adjusted starting point: what worked historically, for which assets, and with what consistency. Whether a given setup survives the next market regime is a question no backtest answers. Use it as a filter, not a signal.
Questions, answered
Is day trading or swing trading better according to the data?
The backtests do not return a universal answer. The winning indicators per asset class hint at time horizons — pivot and intraday indicators dominated in stocks, short-cycle directional indicators in forex — but no single holding period won across all 741 assets. The more useful question is: which indicator has historically worked for the specific asset you want to trade, and what timeframe does that indicator's logic imply?
Does RSI work better on faster timeframes?
RSI Mean-Reversion had a median win rate of 72.6% across backtests, which looks compelling. But it beat buy-and-hold on only 6% of the 741 assets tested. That gap — high win rate, almost no edge over passive holding — tends to show up most sharply on intraday setups where small wins accumulate fast while the underlying asset's drift still compounds against you. The data does not confirm RSI as a systematic edge at any timeframe.
What about SMC strategies on the 15-minute or 1-hour chart?
Smart Money Concepts indicators, including the Liquidity Sweep (71.9% win rate), did not beat buy-and-hold on any asset in these backtests. That held across every asset class tested. These approaches may be applied discretionarily in ways a mechanical backtest cannot capture, but there is no evidence of systematic edge in these results.
Are these real trading results?
No. All figures are from hypothetical out-of-sample backtests on historical data with realistic costs applied. They represent what a mechanical strategy rule would have returned in the past — not what any real account produced, and not a guarantee of any future return. Not financial advice.
Every figure here comes from our own out-of-sample backtests, costs included — not a course or a guess. Educational information only — not investment advice. Hypothetical backtested results; past performance does not guarantee future results. Trading involves risk of loss.
Keep reading
Get the weekly edge report
The best-performing indicator per asset, what changed this week, and the honest caveats — straight to your inbox.