Does Stochastic RSI Work? 20 and 80 Tested on S&P 500 Stocks

After Stochastic RSI fell below 20

Over the 20 trading days before the signal, stocks above their 200-day average had fallen a median 0.9% to 1.9%. Over the next 10 days they rose a median 0.61%, against +0.52% on an ordinary day; the gap was not clear.

After Stochastic RSI rose above 80

Over the 20 trading days before the signal, stocks below their 200-day average had risen a median 0.7% to 2.4%. Over the next 10 days they rose a median 0.74%, against +0.79% on an ordinary day; the gap was not clear.

Interpretation

The lines stay close to zero. In this test, Stochastic RSI described the recent move and was not followed by a clear difference over the next 10 days.

How we tested it

Stochastic RSI can sit at 0 or 100 for several days in a row. A crossing here is the smoothed %K line moving through 20 or 80, so it marks the day the reading turns.

GU Analyser’s default uses an RSI length of 14, a stochastic length of 14 and 3-day smoothing for %K, with zones at 20 and 80. The settings are common defaults. No standard horizon exists for Stochastic RSI: 10 days is our choice, between the 5 days used for RSI(2) and the 20 for RSI(14), with day 5 as a secondary.

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 424,227 signals are not 424,227 independent observations: signals repeat in the same stock, the 10-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.

Limitations

• Today’s S&P 500 members projected backwards (survivorship bias); one period; price returns, dividends excluded.
• An event study: no trading rule, cost or short sale is simulated, so a weaker return after an overbought reading tests no short-selling rule.
• The calm and volatile label uses each stock’s whole-period median, which is known only in hindsight.
• No adjustment for the market’s own move or for other risk factors.
• One horizon, 10 days, chosen by us: no standard exists for Stochastic RSI. The 5-day secondary is one check on that choice; other windows could read differently.
• RSI(2), RSI(14) and ConnorsRSI are tested the same way in their own studies.

How to reproduce this

In the Events panel, build the condition “Stochastic RSI %K crosses below 20”, set the horizon to 10 days and the trend precondition to above (or below) the 200-day average, and run it on your own stocks on daily bars.

Appendix

All results

Checks set in advance

A 5-day check set in advance found a small gap for entering oversold (below 20) in stocks above the 200-day average (+0.12 points); by day 10 it had faded.

Compared with RSI(14)

At the 5-day checkpoint both studies share, the oversold gap in stocks above their 200-day average was +0.12 points for Stochastic RSI and +0.50 points for RSI(14); RSI(14)’s interval sat wholly above. Day 5 is a secondary check for Stochastic RSI and RSI(14). The RSI(14) study.

In the backtest

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, Stochastic RSI came second of four entries, with a median total return per stock of +44% over 2006 to 2026. One possible explanation for the difference between the two tests: the backtest’s short exit captures rebounds that a fixed 10-day window dilutes.