Scalping strategy: a 5-minute pullback plan, costs included

What scalping is, and who it suits

Scalping is the shortest-term style of discretionary trading. A scalper opens a position, aims for somewhere between 3 and 15 pips, and is usually out within the hour. Ten or twenty trades in a session is normal. The idea is that small moves happen far more often than big ones, so you can collect many of them.

The trade-off is that every trade pays the same entry cost, however small the move. On a swing trade with a 60-pip stop, a one-pip spread barely registers. On a scalp with a 6-pip stop it is a large part of the risk, and it is paid on every single trade. That is why most of this page is about costs and timing rather than about the entry signal, which is the easy part.

Scalping suits someone who can give the screen full attention for one or two hours at a fixed time each day, makes decisions quickly, and can accept a loss without needing to win it back on the next trade. It does not suit anyone who trades between meetings. A scalp you check every twenty minutes is not a scalp, it is a badly managed day trade. If that is your situation, the swing trading strategy fits better.

Do the cost arithmetic before anything else

Write down your all-in cost per round trip: the typical spread on the pair at the hour you will trade, plus commission converted to pips. On a raw-spread account with 0.2 pips of spread and $7 commission per lot, that is about 0.9 pips on EUR/USD. On a standard account with a 1.2-pip spread and no commission, it is 1.2 pips. Our guide to choosing a broker covers the account types.

Now see what it does to the break-even point. Take a scalp with a 6-pip stop and a 9-pip target, a 1.5 to 1 ratio:

• Before costs, a 1.5 to 1 trade breaks even if 40% of trades win (6 divided by 15).
• With a 1-pip cost, a winner nets 8 pips and a loser costs 7. Break-even moves to 47% (7 divided by 15).
• With a 2-pip cost, a winner nets 7 and a loser costs 8. Break-even is 53%, and the trade now has to be right more often than not just to stand still.

Compare a swing trade with a 60-pip stop and a 120-pip target: a 1-pip cost moves its break-even from 33.3% to 33.9%. Same spread, almost no effect. The practical rule is simple: if your all-in cost is more than about a fifth of your stop, scalping that pair at that hour is not worth testing. Choose a tighter pair, a cheaper account, or a longer timeframe. A cashback rebate also matters more here than in any other style, and the rebate calculator shows what it is worth at your volume.

The setup: 5-minute pullback to the EMA

This is a trend-pullback scalp, the same logic as a swing trade compressed into minutes. It needs four things.

1. The right hours

Trade only when the market is liquid: the London open and the overlap with New York, roughly 07:00 to 11:00 and 12:30 to 16:00 UK time. Spreads are at their tightest and moves are large enough to reach a target. Our page on forex trading sessions has the full timetable.

2. Direction from the 15-minute chart

Only buy when the 15-minute chart is making higher highs and higher lows and price is above its 50-period EMA. Only sell in the reverse. If the 15-minute chart is flat, there is no trade.

3. The pullback, on the 5-minute chart

Plot the 20-period and 50-period EMAs. In a healthy intraday trend, price pushes away from the 20 EMA, then drifts back to it. You want price to touch or come within a pip of the 20 EMA while staying above the 50 EMA.

4. The trigger

Enter when a 5-minute candle closes back in the trend direction after touching the EMA. A close above the previous candle's high is the clearest version. Do not anticipate it; the close is the signal.

In the chart above, EUR/USD pulls back to the 20 EMA and the pullback bottoms at 1.0842. A bullish candle closes and the entry is taken at 1.0846. The stop goes at 1.0840, two pips under the low, for 6 pips of risk. The target is 1.0855, 9 pips away, which is 1.5 times the risk.

Stop, size and exit

The stop goes one to two pips beyond the pullback low for a long, or beyond the pullback high for a short. On EUR/USD during London that usually gives a stop of 4 to 8 pips. If the pullback is so deep that the stop would need 12 pips or more, skip the trade: the move is no longer a pullback to the EMA, and the target would be too far away for a scalp.

Risk a smaller share of the account per trade than you would on a swing trade, because you take many more trades and losing streaks arrive in an afternoon rather than a month. Half a percent is a sensible ceiling. With a $10,000 account, 0.5% is $50; with a 6-pip stop on EUR/USD, where a pip is worth about $10 per standard lot, that is 0.83 lots. The position size calculator and pip calculator do this in a few seconds, and you should have the number before the trigger candle closes.

Exit at a fixed target of 1 to 1.5 times the risk, set as a take-profit order the moment you enter. Scalping is not the place for trailing stops and letting winners run: the edge, if there is one, comes from taking the move the setup usually offers. Two other exits are worth having. First, a time stop: if the trade has gone nowhere after six 5-minute candles, close it, because the momentum you entered for has gone. Second, a daily stop: after three losing trades, or a loss of 1.5% in a day, stop for the session.

When scalping fails

The commonest way scalping goes wrong is trading at the wrong time. In the quiet hours, especially the late New York afternoon and the Asian session on European pairs, EUR/USD can sit in a 6-pip range for hours while the spread widens. Every entry looks like a pullback, none of them has room to reach a target, and costs take a bite out of each one.

The chart shows the pattern: a tight Asian range in which two pullback entries are stopped out, then the London open, when the same pair moves 25 pips in an hour. The scalper who waited for London needed one trade. The one who traded through Asia paid for three.

The other failures are execution and behaviour. Slippage of half a pip on entry and exit changes the arithmetic above as much as a wider spread does, so record the price you actually got on every trade. News releases are worse: in the minute after a jobs report or a central bank decision, spreads can widen to several pips and stops can fill well past their level. Close scalps before high-impact releases and wait at least fifteen minutes after. Finally, overtrading. A losing streak tempts you to take the next setup that is nearly right, and with ten or more trades a day, nearly right adds up fast. The daily stop exists for exactly that.

How to backtest a scalping strategy

Scalping is harder to test honestly than slower styles, because small errors in cost and fill assumptions are large relative to the trade. Do it in this order:

1. Choose one pair and your trading window, for example EUR/USD from 07:00 to 11:00 UK time.
2. Scroll the 5-minute chart back three months and replay it bar by bar, hiding the bars to the right. Mark every trade the rules produce, with entry, stop, target and result.
3. Charge every trade your real all-in cost, plus half a pip of slippage on each side. If you only have one price feed, assume the worse fill.
4. Log at least 100 trades. Scalping produces them quickly, so there is no reason to stop at fewer.
5. Then trade it on a demo account for two weeks at the real hours and compare your demo fills with the backtest.

Judge the result on the average net win against the average net loss, the longest losing streak, and the worst single day. If the strategy is only positive before costs, it does not work, whatever the entry looks like. Our guide to backtesting a trading strategy covers the method in more depth.