The Test Bench › Mean reversion › ConnorsRSI › Study 014
Does ConnorsRSI predict a bounce better than RSI(2)? We tested 159,794 signals across 483 S&P 500 stocks, 2006 to 2026.
STUDY 014When ConnorsRSI fell below 10 in a stock above its 200-day average, the stock rose a median 0.57% over the next 5 days, 0.29 percentage points more than on an ordinary day.
When ConnorsRSI rose above 90 (a test level set by GU Analyser) in a stock below its 200-day average, the stock fell a median 0.19%, 0.59 percentage points less than on an ordinary day.
Four of the eight comparisons we ran showed a clear gap; the other four did not.
Against RSI(2) below 10 over the same 5 days, the two oversold gaps overlapped, so neither was clearly stronger.
How ConnorsRSI is calculated, with worked examples: ConnorsRSI explained. How it is used in mean reversion rules: mean reversion trading explained. The same test for RSI(2): RSI(2) study.
An event study: it measures what followed each signal. No entry, exit, position size or trading cost is simulated; the mean reversion backtest of these signals is a separate study.
Over the 10 trading days before the signal, stocks above their 200-day average had fallen a median 3.1% to 5.4%. Over the next 5 days they rose a median 0.57%, against +0.28% on an ordinary day: a gap of 0.29 percentage points.
Over the 10 trading days before the signal, stocks below their 200-day average had risen a median 3.8% to 6.2%. Over the next 5 days they fell a median 0.19%, against +0.40% on an ordinary day: a gap of 0.59 percentage points.
The gaps are small next to the usual spread of returns, the dashed lines in the charts above: the largest clear gap was 0.59 percentage points over 5 days, measured across many thousands of signals. A single trade can land far from it.
The largest gap came on the overbought side, which the ConnorsRSI guidebook does not trade; the 90 line is GU Analyser’s own test level.
Question: after a stock’s ConnorsRSI crosses into or out of its zones, does its median return over the next 5 trading days differ from ordinary days for the same stocks in the same trend?
| Part | Definition |
|---|---|
| Indicator | ConnorsRSI(3, 2, 100): the average of RSI(3) of the close, RSI(2) of the up/down streak and the percent rank of the one-day return over 100 days (Connors Research, 2012) |
| Entering a zone | The first close below 10 (or above 90) after a close outside it |
| Leaving a zone | The first close back above 10 (or back below 90) |
| Measured | From the signal day’s close, over the next 5 trading days |
| Compared with | Every eligible day of the same stocks on the same side of the 200-day average |
| Stocks | 483 current S&P 500 members, 2006 to 2026, price returns |
The 90 line is our research test. The results above 90 come from a symmetrical level set by GU Analyser. They are not a test of a published ConnorsRSI short-selling strategy: the guidebook trades the long side only and publishes no overbought rule.
The ConnorsRSI guidebook (Connors Research, 2012) tests buy levels from 5 to 15; 10 sits in the middle and matches RSI(2)’s line.
The guidebook’s own strategy adds filters this study leaves out, among them ADX(10) above 30 and an entry on a further pullback below the signal day’s close. Here every crossing counts, with the 200-day average as the trend split used across the series.
The design was set on 6 October 2026 in our internal study record, before any return was read. The record is internal, not a public preregistration.
The 159,794 signals are not 159,794 independent observations: signals repeat in the same stock, the 5-day windows overlap, and stocks move together. The 95% intervals come from a calendar-block bootstrap that resamples whole stretches of the calendar, rerun with several block lengths and designs. A gap is called clear when every interval excluded zero on the same side. The eight comparisons are not adjusted for being tested together. Signal days are part of the ordinary-day pool, as in the app’s event study, which makes each gap slightly smaller.
Survivorship bias. This is not the historical index. It is today’s S&P 500 members projected backwards, so companies that failed or left the index after a deep fall are missing. That can materially favour a signal that buys falls.
In the Events panel, add ConnorsRSI (Momentum group), build the condition “crosses below 10”, set the horizon to 5 days and the trend precondition to above (or below) the 200-day average, and run it on your own stocks on daily bars.
| Signal | Stocks | Signal days | Median 5-day return | Ordinary days | Difference | 95% interval | Result |
|---|---|---|---|---|---|---|---|
| Enters oversold (below 10) | Above 200-day | 17,844 | +0.57% | +0.28% | +0.29 points | +0.11 to +0.47 | clear |
| Enters oversold (below 10) | Below 200-day | 20,884 | +0.44% | +0.40% | +0.04 points | −0.22 to +0.34 | not clear |
| Leaves oversold (back above 10) | Above 200-day | 17,883 | +0.42% | +0.28% | +0.13 points | −0.04 to +0.31 | not clear |
| Leaves oversold (back above 10) | Below 200-day | 20,818 | +0.42% | +0.40% | +0.02 points | −0.23 to +0.25 | not clear |
| Enters overbought (above 90) | Above 200-day | 31,475 | +0.15% | +0.28% | −0.14 points | −0.22 to −0.05 | clear |
| Enters overbought (above 90) | Below 200-day | 9,708 | −0.19% | +0.40% | −0.59 points | −0.85 to −0.35 | clear |
| Leaves overbought (back below 90) | Above 200-day | 31,307 | +0.20% | +0.28% | −0.09 points | −0.17 to 0.00 | some settings |
| Leaves overbought (back below 90) | Below 200-day | 9,875 | 0.00% | +0.40% | −0.40 points | −0.65 to −0.15 | clear |
| Checks set in advance, after 1 day | Above the 200-day average | Below it |
|---|---|---|
| Enters oversold (below 10) | +0.12 points (+0.04 to +0.20) | +0.02 points (−0.09 to +0.13) |
| Leaves oversold (back above 10) | +0.02 points (−0.06 to +0.10) | 0.00 points (−0.14 to +0.14) |
| Enters overbought (above 90) | −0.06 points (−0.09 to −0.04) | −0.14 points (−0.24 to −0.07) |
| Leaves overbought (back below 90) | −0.03 points (−0.06 to 0.00) | −0.12 points (−0.23 to −0.02) |
| Checks set in advance, after 10 days | Above the 200-day average | Below it |
|---|---|---|
| Enters oversold (below 10) | +0.24 points (−0.08 to +0.54) | −0.02 points (−0.39 to +0.38) |
| Leaves oversold (back above 10) | +0.09 points (−0.20 to +0.37) | −0.14 points (−0.59 to +0.26) |
| Enters overbought (above 90) | −0.16 points (−0.29 to −0.04) | −0.63 points (−1.00 to −0.24) |
| Leaves overbought (back below 90) | −0.09 points (−0.22 to +0.04) | −0.46 points (−0.86 to −0.06) |
At 10 days, a check set in advance, the oversold gap in stocks above their 200-day average was +0.24 points (−0.08 to +0.54), no longer clear; the overbought gap below the average held at −0.63 points.
At the 5-day checkpoint both studies share, the oversold gap in stocks above their 200-day average was +0.29 points for ConnorsRSI and +0.19 points for RSI(2); the intervals overlap. The RSI(2) study.
In the mean reversion backtest, which buys the dip above the 200-day average and sells after the first close above the 5-day average, ConnorsRSI came third of four entries, with a median total return per stock of +13% over 2006 to 2026.
Using this research. The analysis, tables and figures on this page are GU Analyser Ltd’s own work, published under a Creative Commons BY-NC 4.0 licence. You are welcome to quote it, cite it and build on it, with credit and a link. Suggested credit: GU Analyser, The Test Bench, Study 014, linking to guanalyser.com/test-bench/connors-rsi-oversold-overbought/. For commercial use, ask us at hello@guanalyser.com.
Rebuild the test
Build the ConnorsRSI condition in the Events panel, set the trend precondition and the horizon, and measure what followed it on your own stocks.
Create a free account and run your own tests → or read how ConnorsRSI is calculated →The method: how an event study works