RSI Trading Strategy 2026: How RSI Works, Settings and Limits

Checked on October 1, 2026 against Wilder's method as documented in StockCharts' ChartSchool RSI reference. This rewrite removes invented win rates for RSI setups, a fabricated six-month backtest, dollar profits on example trades and claims that combining RSI with other tools lifts win rates above 70%.

What RSI Measures

RSI is a momentum oscillator: it measures how strongly price has been rising or falling, not where it will go next. J. Welles Wilder Jr. introduced it in New Concepts in Technical Trading Systems (1978). For each bar it looks at the change from the previous close, averages the gains and the losses over the look-back period, and turns the ratio into a number between 0 and 100. A reading of 50 means average gains and losses are equal; readings near 100 mean almost all recent movement was up.

How RSI Is Calculated

1. Split each change into a gain or a loss. If a close is above the previous close, the difference is a gain and the loss is zero; if below, the difference is a loss, recorded as a positive number.
2. Start with simple averages. First average gain = sum of gains over the past 14 periods ÷ 14. First average loss = sum of losses ÷ 14.
3. Smooth after that (Wilder's method). Average gain = (previous average gain × 13 + current gain) ÷ 14, and the same for losses.
4. Convert to RSI. RS = average gain ÷ average loss, and RSI = 100 − 100 ÷ (1 + RS). If the average loss is zero, RSI is 100.

A worked example, starting from a 14-period average gain of 0.60 and average loss of 0.40:

Because each value carries the earlier averages forward, RSI depends on how much history your platform loads. StockCharts, for example, uses at least 250 data points before a chart's start date, so two platforms can show slightly different readings for the same bar. You never need to calculate RSI by hand, but knowing the formula explains why a single large bar can move it sharply and why it reacts less as the look-back gets longer.

RSI Settings and What Changing Them Does

A period counts bars, not time. Fourteen periods cover about 70 minutes on a 5-minute chart and 14 sessions on a daily chart. Changing the setting after every losing trade until the past looks good is curve fitting; fix the setting before you test.

Four Ways Traders Read RSI

1. Overbought and oversold

Wilder considered RSI overbought above 70 and oversold below 30. The trap is treating those levels as automatic sell and buy signals. In a strong trend RSI can stay above 70, or below 30, while price keeps going, and StockCharts notes that overbought readings can be a sign of strength rather than a top.

2. Trend ranges

Constance Brown, in Technical Analysis for the Trading Professional, observed that RSI tends to hold between 40 and 90 in an uptrend, with the 40 to 50 zone acting as support, and between 10 and 60 in a downtrend, with 50 to 60 acting as resistance (as summarized by StockCharts). Read this way, a dip to 40 in an uptrend is a possible pullback, not an oversold signal. The exact ranges vary with the market, the setting and volatility.

3. Failure swings

A bullish failure swing: RSI drops below 30, rises back above 30, pulls back but stays above 30, then breaks above its previous high. A bearish failure swing is the mirror image: RSI rises above 70, pulls back, bounces without exceeding 70, then breaks below its previous low. Failure swings are read from RSI alone, without reference to price. Wilder considered them strong indications of an impending reversal; in practice they still fail, so they need a stop like any other signal.

4. Divergence

A bullish divergence is a lower low in price with a higher low in RSI; a bearish divergence is a higher high in price with a lower high in RSI. Both show momentum fading. The limitation matters more than the pattern: in StockCharts' words, “divergences are misleading in a strong trend,” and a strong uptrend can show several bearish divergences before any top. Andrew Cardwell described the opposite patterns, which he called positive and negative reversals: a positive reversal is a lower low in RSI while price makes a higher low, sometimes labeled hidden bullish divergence.

Turning RSI Into Rules You Can Test

An RSI reading is a condition, not a strategy. A usable rule set names the market context, the trigger, the stop, the size and the exit. The two examples below are educational templates to test, not recommendations or proven edges.

Example A: pullback in an uptrend

1. Context: price is above a long moving average you define in advance, such as the 200-day.
2. Trigger: RSI(14) dips below 40, then closes back above 40.
3. Stop: below the low of the pullback.
4. Exit: a fixed multiple of the risk, or a rule you set in advance, such as RSI closing above 70.

Where it breaks down: when the trend is ending, pullbacks turn into reversals, and the moving-average filter reacts late.

Example B: fade at the edge of a range

1. Context: price has been moving sideways between a support and resistance level you can mark.
2. Trigger: near support, RSI(14) closes below 30 and then back above 30.
3. Stop: below the range low.
4. Exit: the middle or top of the range.

Where it breaks down: ranges eventually break, and an oversold reading at support is exactly what a breakdown looks like at first.

How to test the rules

• Write the rules so someone else could follow them, and fix every setting before you look at results.
• Record every signal the rules produce, including the ones you would have skipped.
• Subtract commissions, fees and slippage from every trade. On futures, one tick of slippage is $12.50 per ES contract and $1.25 per MES.
• Measure expectancy: (win rate × average win) minus (loss rate × average loss), after costs. A high win rate with larger losses can still lose money.
• Check the rules on data you did not use to design them, then forward-test in simulation.
• Remember the limits of simulated results: CFTC rules require advisors who show them to warn that they do not represent actual trading and are designed with the benefit of hindsight (17 CFR 4.41).

Risk Management Decides More Than the Signal

No RSI rule avoids losing streaks, so size every trade from its stop. Dollar risk per contract is the stop distance in ticks times the tick value: an 8-point stop on the Micro E-mini S&P 500 is 32 ticks of $1.25, or $40 per contract, so a $200 limit allows 5 contracts before costs. Set a daily loss limit and stop when you reach it. Our ES futures trading strategy guide covers sizing from the stop in more depth, and the futures trading for beginners guide explains margin and tick values.

RSI Mistakes to Avoid

• Selling every reading above 70 in an uptrend, or buying every reading below 30 in a downtrend.
• Treating divergence as timing: it can persist through a long trend.
• Re-optimizing the setting after each loss until the backtest looks perfect.
• Ignoring the timeframe: a 14-period RSI on a 1-minute chart describes the last 14 minutes, not the trend.
• Trading without a stop because the indicator “should” turn.
• Comparing RSI values across platforms without checking how each calculates it.

RSI vs Stochastic and MACD

RSI compares the size of up closes with down closes. The stochastic oscillator compares the latest close with the high-low range over its look-back, and MACD tracks the gap between two exponential moving averages. All three respond to momentum and often move together, so stacking them adds less independent confirmation than it seems. Our MACD guide covers that indicator; for reading the order book instead of an oscillator, see the order flow guide.

RSI Trading Strategy FAQ

Related Reads

• ES futures trading strategy
• ES futures scalping strategies
• Futures trading psychology
• Full trading library