Tracked Signals

Strategy tests

Data Trader 5-minute scalping strategy (4-hour range): review and backtest

No edge YouTube strategy · tested 11 Oct 2026 · figures re-checked 11 Oct 2026

What does Data Trader's 5-minute scalping strategy claim?

The video "The BEST 5 Minute Scalping Strategy Ever (Simple and Proven)" (2025) presents a rule-based scalping method for gold, crypto and forex with no indicators. Its gold example is 10 trades, six wins and four losses for +8R, described as a 60% win rate; it also shows 7 crypto trades (about 72%) and 6 EUR/USD trades (about 83%). The days were picked by the presenter, and no trading costs are included.

How we tested it

Our version of the rules: mark the high and low of the first 4-hour block of the New York day (00:00 to 04:00 New York time). From 04:00 to midnight, on 5-minute bars, when a candle closes above the range and a later candle closes back inside it, sell; when a candle closes below the range and a later one closes back inside, buy. Entry at the next bar's open, stop at the far end of the breakout move, target 2 times the risk, one trade at a time, more than one trade a day allowed. The video's 'nearest key level' stop for very large breakouts is a judgement call and was not tested; a mechanical stand-in was run separately. All of these choices were written down before the first test ran. Main test: gold on 5-minute bars, January 2020 to August 2026. Checks: the stand-in stop, 15-minute and 1-hour gold, and the same rule on 8 markets at 30 minutes: 5 versions in all.

33.3% winsthe '60% win rate', measured on gold 5-minute744 of 2,232 trades reached the 2R target; 95% range 31.4-35.3%
33.3% / 38.3%win rate needed to break evenbefore / after costs, with a 2R target
-0.150Raverage per trade, after costsabout zero before costs (-0.0003R); 2,232 gold 5-minute trades
gone by Feb 2022$10,000 risking $100 a trade, after coststhe $10,000 was lost by trade 647 (22 Feb 2022); before costs it ended at $9,943
0 of 7years profitable after costs2020 to 2026; before costs 3 of the 7 were above zero
7 of 8markets negative after costs on 30-minute barsthe exception, EUR/USD, made +0.007R a trade on 68 trades
Before trading costs unless stated. R = the amount risked on a trade.

Costs are spread, slippage and overnight financing, about 0.15R per trade on gold 5-minute: the median stop was only 4.97 points (dollars per ounce), so ordinary costs are a large share of each trade. They turned an average of about zero per trade into -0.150R. At 1.5 times these costs it was -0.225R. The worst losing run on the 5-minute test was 340R.

Does Data Trader's 5-minute scalping strategy work?

Not in our test: it did no better than chance. The 60% comes from 10 gold trades picked in the video; even taken at face value, 6 wins in 10 has a 95% range of about 31-83%. With a target twice the risk, a strategy needs 33.3% wins to break even before costs and, here, 38.3% after them. On gold 5-minute it won 33.3%: exactly the first line, so it made about nothing before costs, and costs turned that into a loss in every year from 2020 to 2026. Random entries in the same hours, on the same side and with the same stop sizes, averaged -0.179R a trade after costs, and taking a coin-flip side at the rule's own moments averaged -0.139R; the rule's -0.150R is within what chance produces in both comparisons. All four gold versions won between 33.2% and 34.0%.

Does the 4-hour range scalping strategy win 60% of trades?

Not in our test. The 60% is 6 wins in 10 gold trades chosen in the video. Run on every setup on gold 5-minute charts from January 2020 to August 2026, 2,232 trades, it reached its 2R target 33.3% of the time (95% range 31.4-35.3%).

What win rate does the 5-minute range strategy need?

With a target twice the risk, 33.3% wins breaks even before costs. After spread, slippage and financing on gold 5-minute it needs 38.3%; it won 33.3%.

Does it do better on 15-minute or 1-hour charts?

It loses less but still loses after costs: -0.104R a trade on 15-minute gold (1,533 trades) and -0.071R on 1-hour gold (910 trades). The win rate stayed between 33.6% and 34.0%; wider stops only make costs a smaller share of each trade.

Limits of this test

  • The video does not say which 4-hour candle a gold chart should use. We used 00:00 to 04:00 New York time, the block on the video's own crypto example and the default of a separate public re-creation of these rules; on some brokers' gold charts the first 4-hour candle starts an hour or two differently.
  • Our version of the rules also assumes: a close on a range line counts as back inside; the stop is the highest high (lowest low) from the first close outside through the candle that closes back inside; the target is exactly 2 times the risk; trades are held until stop or target, with no time exit; a setup that appears while a trade is open is skipped.
  • The 'nearest key level' stop was not tested: it is a judgement call. Our mechanical stand-in (a recent swing point between the entry and the breakout extreme) won 33.2% of 3,505 trades, -0.174R a trade after costs; it is our approximation, not the video's rule.
  • Longer bars lost less only because costs take a smaller share of wider stops: 15-minute gold won 33.6% of 1,533 trades (+0.010R before costs, -0.104R after), 1-hour gold 34.0% of 910 trades (+0.020R before, -0.071R after).
  • The 8-market check uses 30-minute bars from January to May 2026, because we do not hold a long 5-minute history for those markets; it tests the same rule on coarser bars.
  • The rules come from a transcript of the video's narration. This checks a claim on price data already used in our earlier research; no held-back data was opened. Gold prices are Dukascopy's; costs use our own spread, slippage and overnight-financing model, not one broker's.
  • Not financial advice: these are measured results on past prices, for one version of the rules, and they do not predict future results.

Measured results, not advice. We have no commercial link with Data Trader. If you made this tool and think a figure is wrong, write to corrections@trackedsignals.com; corrections are published within 7 days. How we test.

More strategy tests

All strategy tests