Daily RSI Below 30 on US Stocks|Effect of an SMA200 Filter
ENTRIQ backtest of 100 US stocks on daily charts (5–56 years each): how RSI-below-30 buys change when limited to days closing above SMA200, versus no filter.
About this backtest
These are the results of backtesting 100 stocks in ENTRIQ (5 to 56 years of data depending on the stock) (Dataset v1.0). The universe is US stocks selected from the current S&P 500 constituents, taking the largest by market cap with up to 15 stocks per sector.
Note that this backtest covers the current index constituents and does not include stocks that were delisted or removed from the index in the past. The results are therefore subject to survivorship bias, which should be kept in mind.
The idea: why combine RSI with SMA200
When RSI(14) drops below 30, it is often treated as a sign that a stock is "oversold." However, the same condition also appears in stocks that are in a long-term decline, so whether it can be used the same way in any market environment is a separate question.
The moving average (SMA200) is a widely used guide to the long-term trend on daily charts. One approach is to add the condition "the close is above SMA200" to narrow the focus to phases in which price has been supported over the long term.
This article compares results with this filter and without it (RSI below 30 only) across 3 exit patterns, including how the sample size changes. Detailed figures without the filter are covered in Daily RSI Below 30 on US Stocks|Backtest of 3 Exits.
For the basics of RSI itself, see What Is RSI|Meaning, Reading and Formula for Beginners; for how to use moving averages, see Moving Average Strategy|Slope, Crossovers, and Backtesting.
Test conditions
- Universe: 100 US stocks selected from the current S&P 500 constituents, taking the largest by market cap with up to 15 stocks per sector
- Period: The full history available for each stock (5 to 56 years depending on the stock). Only stocks with at least 5 years of data
- Timeframe: Daily (split-adjusted)
- RSI setting: Period 14 (Wilder's smoothing)
- Entry: If RSI(14) crosses below 30 on a day when the close is above SMA200, buy at the open on the next trading day
- New signals in the same stock while a position is open are skipped (only one position at a time)
- Comparison (no filter): Entry on RSI below 30 alone, without comparing the close with SMA200
Exit patterns (3)
| Label | Exit definition |
|---|---|
| A | Exit at the open on the next trading day after RSI rises above 50 (if it doesn't, exit at the close on the 60th trading day). No stop loss |
| B | Exit at the close on the 10th trading day after entry. No stop loss |
| C | Stop loss at -5%. If the stop isn't hit, exit at the close on the 10th trading day. If the stock gaps down at the open, the fill is at that open price, so a single trade's loss can exceed -5% |
Overall results
With the SMA200 filter (this backtest)
| Metric | Pattern A | Pattern B | Pattern C |
|---|---|---|---|
| Sample size | 1,124 | 1,157 | 1,197 |
| Share of trades closed in profit | 75.4% | 59.5% | 52.5% |
| Average P/L | +2.36% | +1.19% | +0.90% |
| Profit factor | 2.25 | 1.58 | 1.46 |
| Average loss (trades closed at a loss) | -7.76% | -5.10% | -4.12% |
| Largest loss | -64.7% | -36.1% | -37.9% |
| Median MAE (maximum adverse excursion) | -3.66% | -2.88% | -2.95% |
| Longest losing streak (per stock) | 5 | 7 | 8 |
| Average holding period (trading days) | 17.6 | 10.0 | 8.3 |
Note: Past sample (N=1,124, N=1,157, N=1,197); not indicative of future results.
No filter (comparison)
| Metric | Pattern A | Pattern B | Pattern C |
|---|---|---|---|
| Sample size | 5,318 | 7,029 | 7,865 |
| Share of trades closed in profit | 72.2% | 56.9% | 45.9% |
| Average P/L | +2.21% | +1.20% | +0.54% |
| Profit factor | 1.92 | 1.51 | 1.23 |
| Average loss (trades closed at a loss) | -8.78% | -5.58% | -4.40% |
| Largest loss | -93.3% | -72.5% | -37.9% |
| Median MAE (maximum adverse excursion) | -4.73% | -3.57% | -3.76% |
| Longest losing streak (per stock) | 7 | 12 | 16 |
| Average holding period (trading days) | 19.2 | 10.0 | 7.6 |
Note: Past sample (N=5,318, N=7,029, N=7,865); not indicative of future results.
Comparing the effect of the filter
Change in sample size
Adding the SMA200 filter substantially reduces the sample size in every pattern compared with no filter. This is because RSI-below-30 signals that occur while the close is below SMA200 are excluded. The smaller number of opportunities is a cost of the filter that is worth keeping in mind.
Share of trades closed in profit and profit factor
In both patterns B and C, the share of trades closed in profit and profit factor were higher with the SMA200 filter. In pattern A as well, the share of trades closed in profit and profit factor with the SMA200 filter were higher than without the filter.
Average P/L
Pattern C had a higher average P/L with the filter. Pattern A was also slightly higher with the filter (+2.36% vs +2.21%), while in pattern B the two were almost the same, with no filter marginally higher (+1.20% vs +1.19%). Even with the same "buy below RSI 30" condition, it is worth noting that the effect of the filter shows up differently depending on the exit condition.
Risk (losses and MAE)
In both patterns B and C, the average loss (trades closed at a loss) and median MAE were larger (more negative) without the filter. The largest loss was larger without the filter in pattern B and the same in both cases in pattern C. The longest losing streak (the longest run of consecutive losses within a single stock) also involved more consecutive losses without the filter in every pattern.
Taken together, adding the SMA200 filter narrowed the opportunities, but results were milder on several measures: the share of trades closed in profit, profit factor and the depth of losses. However, not every metric favors one side; for example, pattern B's average P/L was marginally higher without the filter.
With the SMA200 filter: breakdown by trend, volatility and earnings
By trend (uptrend vs range)
Because this backtest requires "the close is above SMA200" as an entry condition, there are only a few samples outside uptrends. Refer to the range results while checking them against the minimum sample size (30 trades).
In pattern A, the share of trades closed in profit, average P/L and profit factor were all higher in uptrends than in ranges. The average loss (trades closed at a loss) was larger (more negative) in ranges, and the median MAE was also larger (more negative) in ranges.
In pattern B as well, the share of trades closed in profit, average P/L and profit factor were higher in uptrends than in ranges, and the median MAE was larger (more negative) in ranges.
In pattern C, the share of trades closed in profit was higher in ranges than in uptrends, while average P/L and profit factor were higher in uptrends. The median MAE was larger (more negative) in ranges.
Note: Past sample (uptrend / range: N=1,088 / 31, N=1,121 / 31, N=1,161 / 31); not indicative of future results.
By volatility (low, normal, high)
Pattern A
| Volatility bucket | n | Share of trades closed in profit | Average P/L | PF | Median MAE |
|---|---|---|---|---|---|
| Low volatility | 146 | 81.5% | +1.61% | 2.51 | -2.13% |
| Normal | 751 | 74.2% | +1.82% | 1.97 | -3.80% |
| High volatility | 227 | 75.3% | +4.63% | 2.86 | -4.78% |
The share of trades closed in profit was highest in low volatility, while average P/L and profit factor were highest in high volatility. The median MAE was largest (most negative) in high volatility, showing that adverse moves while holding were also larger in high-volatility periods.
Note: Past sample (N=146, N=751, N=227); not indicative of future results.
Patterns B and C (overview)
In pattern B as well, the share of trades closed in profit was highest in low volatility and average P/L was highest in high volatility. Profit factor in pattern B was highest in low volatility. As in pattern A, the median MAE was largest (most negative) in high volatility.
In pattern C, the share of trades closed in profit was highest in low volatility, while average P/L and profit factor were highest in high volatility. The largest loss and median MAE were both largest (most negative) in high volatility, and risk in high-volatility periods was observed to be notably greater.
By distance from earnings
For pattern A, periods away from earnings (outside the 5 trading days before and after an earnings date) are compared with periods near earnings (within the 5 trading days before and after).
| Distance from earnings | n | Share of trades closed in profit | Average P/L | PF | Median MAE |
|---|---|---|---|---|---|
| Away from earnings | 387 | 76.7% | +2.39% | 2.43 | -3.63% |
| Near earnings | 128 | 72.7% | +1.38% | 1.74 | -4.04% |
The share of trades closed in profit, average P/L and profit factor were all higher away from earnings, and the median MAE was larger (more negative) near earnings. Earnings date data is only available from 2008 onward, so 609 trades fall into neither the away-from-earnings nor the near-earnings category.
In patterns B and C as well, the share of trades closed in profit, average P/L and profit factor were higher away from earnings, and the median MAE was larger (more negative) near earnings.
Note: Past sample (N=387, N=128); not indicative of future results.
Limitations and caveats
Survivorship bias: This backtest covers the current S&P 500 constituents. Stocks that were delisted or removed from the index in the past are not included, so the results may look more favorable than they would if those stocks were included.
Structural bias from the SMA200 condition: The condition "the close is above SMA200" itself selects phases in which price has recovered or held up to some extent. As a result, RSI drops below 30 during downtrends are excluded, and many of the remaining situations resemble temporary pullbacks within uptrends. This is also the intent of the filter, but refer to the results with an understanding of which phases are being selected.
Stop-loss execution: Pattern C uses a -5% stop loss, but if the stock gaps down, the loss can be larger than the stop level. Patterns A and B keep holding without a stop loss, so the largest loss can become large.
Differences in period and stocks: If you trade different stocks, timeframes or market environments, the figures here do not apply as they are. You need to apply the same conditions to your own stocks to check.
For caveats about the reproducibility of backtests, see Why Backtests Fail in Live Trading: 4 Structural Reasons.
Summary
Backtesting RSI-below-30 buys with the SMA200 filter, the sample size was substantially smaller than without the filter, but a tendency for the share of trades closed in profit and profit factor to be higher was observed regardless of the exit pattern. Average P/L was higher with the filter in pattern C (and slightly higher in pattern A), while in pattern B it was marginally higher without the filter, so the difference depends on the exit condition.
On the risk side, the average loss (trades closed at a loss), median MAE and longest losing streak (per stock) were all observed to be larger without the filter (more negative, or more consecutive losses).
To go beyond the figures and see how things look under your own trading conditions, one step is to check the same conditions stock by stock with ENTRIQ's Chart Replay. Recording the results as Strategy Tags makes it easier to review the tendencies of each condition later.
Frequently asked questions (FAQ)
Q. What happens when RSI drops below 30 while the close is below SMA200?
A. In this backtest, "the close is above SMA200" is an entry condition, so no trade is entered when RSI drops below 30 while the closing price is below SMA200. For the results without the filter, see Daily RSI Below 30 on US Stocks|Backtest of 3 Exits.
Q. Would the results change with a different SMA period?
A. This backtest uses SMA200 (period 200). Changing the period changes the proportion of phases the filter lets through and excludes, so the results also change. To check your own settings, try the same conditions in ENTRIQ's Chart Replay.
Q. Why is the range sample so small?
A. Because the condition "the close is above SMA200" is added, downtrends are excluded by definition: a downtrend requires the close to be below SMA200. The filter therefore leaves only uptrends and some ranges, and as a result the range sample is small.
Q. Can I use these results directly as trading rules?
A. These figures are sample observations under specific past conditions and do not indicate future results. Results also change with the stocks, timeframe, trading costs and other factors. Run a separate backtest under your own conditions.
Q. Does not using a stop loss in pattern A make the largest loss larger?
A. Patterns A and B are designed to keep holding without a stop loss until RSI recovers to 50 or the holding period ends. Since no stop loss is triggered, positions can carry large losses while held, depending on the market environment. Check the largest loss and median MAE figures to see how large adverse moves have been in the past.
Related articles
- Complete Guide to RSI|Reading, Settings, Entries and Testing
- Daily RSI Backtest on US Stocks by Trend|How Results Vary
- Daily RSI Below 30 on US Stocks|Results by Volatility Bucket
- Daily RSI 9, 14, 21 on US Stocks|What Changes in Backtests
This article is provided for informational purposes to examine and organize historical data and does not recommend any specific trade. The figures shown are past samples and do not indicate future results. All investment decisions are your own responsibility.
Test the same condition used in this article on symbols of your choice.
Backtest this conditionReproduce this validation yourself
With ENTRIQ's chart replay, you can trade through past charts using the same rules.
Try the demo — no sign-upNo sign-up or card needed. Runs right in your browser. Start 14-day free trial →
Validate your method on past charts
The same validation in this article can be reproduced with your own method. On past charts, do the classics really work?
- ✓ Replay past charts
- ✓ AI feedback on your trades
- ✓ Validate methods with statistics
No sign-up or card needed. Runs right in your browser.