The best RSI settings for day trading are not the same as the best settings for swing trading, and on 15-minute Nasdaq futures charts the honest answer from our backtest is that no RSI setting produced a large edge. Some were mildly profitable, one popular choice lost money, and the two most impressive-looking results rest on small samples.
We took a simple intraday version of the RSI pullback idea — buy an oversold reading while the short-term trend is up — and ran it on 15-minute NQ bars for every regular session from June 2010 to April 2026. We tested RSI(2), RSI(3), RSI(6) and RSI(14), the same periods from our daily RSI settings comparison, with thresholds scaled to each period.
The 15 minute RSI rules
- Market: NQ continuous front-month futures, regular session only (09:30–16:00 ET), 15-minute bars, 3,884 sessions from June 7, 2010 to April 16, 2026.
- Indicator: RSI(n) on 15-minute closes using Wilder-style smoothing, calculated across regular-session bars only (overnight bars are excluded).
- Trend filter: the bar must close above its 50-bar exponential moving average.
- Entry: long at the close of any bar before 15:30 ET where RSI(n) is below the threshold and the trend filter is true. Long only.
- Exit: at the close of the first bar where RSI(n) is above 70, or at the session close (15:45 bar close). More than one trade per day is possible, but only one position at a time.
- Stop: none. Every position is flat by the close.
- Costs: 1.0 NQ point per round trip.
All results are in NQ index points for one contract. MNQ is $2 per point and NQ is $20 per point, so MNQ dollars below are simply points × 2. Every trade has 1.0 point deducted for costs (slippage plus commission).
Best RSI settings for day trading: the results
| 15-min setting | Trades | Win rate | Avg pts | Total pts | Profit factor | Closed-trade DD (pts) | 2010–2019 pts | 2020–2026 pts |
|---|---|---|---|---|---|---|---|---|
| RSI(2) < 5 | 697 | 62.4% | +0.58 | +405.5 | 1.05 | 1,258.5 | -103 | +508.5 |
| RSI(2) < 10 | 1,599 | 63.5% | -1.84 | -2,937 | 0.85 | 3,765.5 | -744.5 | -2,192.5 |
| RSI(3) < 10 | 362 | 60.2% | +1.21 | +439.5 | 1.07 | 744.8 | -93.2 | +532.8 |
| RSI(3) < 15 | 802 | 64.1% | +1.03 | +824.5 | 1.07 | 1,141 | +180 | +644.5 |
| RSI(6) < 20 | 86 | 59.3% | +8.87 | +762.8 | 1.6 | 196.5 | +37.5 | +725.2 |
| RSI(6) < 25 | 229 | 57.2% | +4.52 | +1,035 | 1.28 | 486 | -111 | +1,146 |
| RSI(14) < 30 | 0 | – | – | – | – | – | – | – |
| RSI(14) < 35 | 0 | – | – | – | – | – | – | – |
| RSI(14) < 40 | 44 | 72.7% | +21.57 | +949.2 | 5.56 | 91.5 | +71 | +878.2 |
| RSI(14) < 45 | 404 | 54.2% | +3.52 | +1,423 | 1.21 | 842.5 | -209 | +1,632 |
Read this table with the trade counts in mind:
- RSI(2) day trading at < 10 lost money. It was the busiest setting, with 1,599 trades and a 63.5% win rate, yet it lost 2,937 points (profit factor 0.85). The average win was 16.39 points and the average loss -35.17 points — high win rate, small wins, bigger losses. It lost in both 2010–2019 and 2020–2026.
- RSI(2) < 5 and RSI(3) were roughly break-even. Profit factors of 1.05–1.07 leave almost no room for worse fills than we assumed.
- RSI(6) < 20 looks good but is a small sample. 86 trades in almost 16 years, profit factor 1.6, 762.8 points. Of that, 484 points came in 2025 and 290.5 in 2020. The 2010–2019 total was only 37.5 points.
- RSI(14) < 30 and < 35 produced zero trades. A 14-period RSI on 15-minute bars basically never gets that low while price is still above its 50-bar EMA.
- RSI(14) < 40 has the best-looking numbers and the smallest sample. 44 trades, a 72.7% win rate and a profit factor of 5.56. That is a tiny sample for a 16-year test — roughly three trades a year — and we would not treat the profit factor as reliable.
- RSI(14) < 45 made the most points (1,423 on 404 trades), but 1,588.8 of those points came in 2025 alone. Excluding that one year, it would have been a small net loser. It also lost 209 points in 2010–2019.
Is there an MNQ RSI strategy worth trading here?
Put the numbers in MNQ dollars (points × 2). The best total, RSI(14) < 45, made about $2,846 on one MNQ contract across almost 16 years. RSI(6) < 25 made about $2,070. For comparison, a plain 15-minute opening range breakout held to the close made about $30,410 per MNQ contract over the same sessions in our tests. The intraday RSI pullback is a small-edge idea at best on this market.
There is also a structural reason the numbers are modest: the trend filter and the oversold condition work against each other. The more oversold RSI gets, the more likely price has already fallen below its 50-bar EMA, which cancels the signal. Longer RSI periods suffer most, which is why RSI(14) only fired once we raised the threshold to 40.
Our take: If you want an RSI for 15-minute Nasdaq charts, the short-period “RSI 2 day trading” setup that works nicely on daily bars did not carry over here — RSI(2) < 10 was the worst result in the table. The medium settings (RSI(6) at 20–25, RSI(14) at 40–45) were positive, but either the samples are small or the profit is concentrated in one or two years. We would use intraday RSI as a timing tool inside a bigger idea, not as a stand-alone strategy.
What is not modelled
- This is a backtest; past results do not guarantee future results.
- Entries and exits are at 15-minute bar closes with 1 point of costs; real fills may be worse.
- No stop loss is used, so a single trend day against the position is absorbed until the close. There is no overnight gap risk because every trade is closed by 16:00 ET.
- Only long trades were tested. No taxes, data or platform fees.
- Ten combinations were tested; the best one may look good partly by chance.
Key takeaways
- RSI(2) < 10 on 15-minute NQ: 1,599 trades, 63.5% win rate, but -2,937 points after costs.
- RSI(14) < 30 and < 35 never triggered with the 50-EMA trend filter; < 40 and < 45 did.
- The best-looking results are small samples: RSI(6) < 20 had 86 trades, RSI(14) < 40 had 44.
- RSI(14) < 45 made 1,423 points, most of it in 2025.
- 1.0 point per round trip deducted; MNQ = $2/pt, NQ = $20/pt.
Put the research to work on your Desk
Explore the tools behind the research, compare the available access, or open your Desk if you already have an account.
FAQ
What RSI setting is best on a 15-minute chart?
In our 2010-2026 NQ test, no setting showed a strong edge. RSI(14) below 45 made the most points (1,423), but most of that came in 2025. RSI(14) below 40 had the best profit factor (5.56) on just 44 trades, which is a small sample.
Does RSI(2) work for day trading?
Not in this test. On 15-minute NQ bars, RSI(2) below 10 had a 63.5% win rate but lost 2,937 points over 1,599 trades after 1 point of costs per trade.
Why did RSI(14) below 30 produce no trades?
The system also required the bar to close above its 50-bar EMA. On 15-minute Nasdaq bars, RSI(14) never closed below 35 on an eligible entry bar while that trend filter was true.
How much would this make on MNQ?
MNQ is $2 per NQ point. The best total, RSI(14) below 45, equals about $2,846 per MNQ contract over almost 16 years, before taxes. Past results do not guarantee future results.
Loomi.Ai provides educational market research, not financial advice. Trading futures and options involves substantial risk of loss, and backtested results are historical and do not guarantee future returns.
Sources
- Loomi backtest: blog_strategy_tests.py + results JSON, Loomi research
- J. Welles Wilder Jr., Wikipedia