Technical & Price Action
RSI Oversold Stocks is a Technical & Price Action screen. It checks every NSE-listed stock against 1 condition computed from nightly data, after excluding heavily pledged promoters and very small market caps. This page explains the rule, each figure and its source. It lists no stocks and is not advice.
The Relative Strength Index (RSI) measures how fast and how much a stock has moved relative to its own recent history. An RSI below 30 signals that the stock has fallen sharply and quickly — sellers have dominated to an unusual degree. Historically, extreme oversold readings tend to precede short-term bounces, not because the business has changed, but because the selling has been excessive relative to the stock's own pace. This is a mean-reversion screen, not a quality filter. It works best when combined with a check that the underlying business is still fundamentally sound.
A stock must satisfy every condition below. Conditions are grouped under the pillar they test.
Oversold Condition
Hard excludes
Ranked by RSI (14), lowest first.
RSI (14-Day)
Relative Strength Index calculated over 14 trading days. It ranges from 0 to 100. Above 70 is considered overbought (stock has risen too fast); below 30 is considered oversold (stock has fallen too fast).
Low is better
Type hypothetical values to see which conditions they satisfy. These are not real stocks and nothing here is stored.
Enter values to begin
How many stocks pass if this threshold were set differently. Counts only; no stocks are named.
Loading the worked example…
Sign in with a completed risk profile to check any stock against this rule.
In the past data we checked, the stocks this screen picked did worse than the ones it left out, by more than luck alone would explain. We looked back over 64 months. Each month we checked whether the stocks this screen picked did better, over the following month (21 trading days), than the stocks it didn't pick. We score that from -1 (the picks always did worse) to +1 (always did better); 0 means no pattern. This screen scored -0.0291. As a measurement, the score is very small, but clearly below zero. If the screen had no real edge, a score at least this far from zero would show up less than 1 in 1,000 by luck alone. That does not mean the screen is wrong; it still works for narrowing a list of stocks by your own rules.
This is a record of the past, not a prediction.
Not yet testable — 5 of 6 points needed to chart a rolling information coefficient.
This is a historical, statistical measurement produced by FINOME TECHNOLOGIES PVT LTD's research harness on past data. It is not a prediction, and past results do not indicate future results. The universe is today's Nifty 500 constituents, so the sample is affected by survivorship bias. It describes how a screen's selections behaved in the past and does not constitute investment advice or a recommendation.
Rule version 1.0.0