That is fifteen names out of 1,707. On the same scan, 6.8% were overbought above 70, and the median reading across the whole universe was 54.9.
If you have been taught that RSI tells you when to buy and when to sell, those numbers should give you pause. The signal most people wait for fired on fewer than one stock in a hundred.
What the indicator actually measures
RSI compares the average size of recent gains with the average size of recent losses, usually over fourteen periods, and puts the result on a scale from 0 to 100. Fifty means gains and losses have been about balanced.
The more useful description is what it leaves out. RSI knows nothing about volume, nothing about valuation, nothing about why a stock moved, and nothing at all beyond its lookback window. It is a compact summary of how one-sided recent price changes have been — no more than that, and treating it as more is where the trouble starts.
Why 30 and 70 are conventions, not physics
Welles Wilder proposed those thresholds in 1978. They have survived through repetition rather than through testing, and there is nothing structural about either number. Some traders use 20 and 80 to make signals rarer; others move the bands depending on market conditions.
A threshold that fires on 0.9% of a universe is not a filter you can build a routine around. It is an event you will wait months for, and the waiting is what makes people abandon the rule at exactly the wrong moment.
The number that explains the number
On the same scan, 67.6% of tracked names were trading above their 200-day EMA and 66.8% above their 50-day. That is a broadly rising market — and in a rising market, almost nothing gets oversold.
This is the part that gets lost when the indicator is taught in isolation. The 30 line does not move, but what reaching it means changes completely with conditions. In a sharp decline the same threshold can fire on hundreds of names in a single session, and most of them are simply falling together rather than each becoming individually attractive.
Same reading. Same rule. Entirely different information, depending on a regime the indicator cannot see.
What to do with that instead
- Read it relatively. Where a stock’s RSI sits against its own recent range says more than where it sits against a number chosen in 1978.
- Pair it with context. An oversold reading in a broadly rising market is a different event from the same reading when most of the market is below its 200-day.
- Do not wait for a threshold to grant permission. If your plan requires a sub-30 print, your plan requires something that happened to 0.9% of stocks on the day we measured.
How we measured this
These figures come from our own scan of the equities on the platform — 1,707 names with a valid RSI reading on 16 August 2026, computed on daily bars. Every percentage on this page carries the sample it was drawn from, because a percentage without one is not evidence.
Numbers from a single day describe that day. We publish the date and the sample size so you can weigh them accordingly, rather than treating one scan as a general law.
Frequently asked questions
Does RSI below 30 mean a stock is cheap?
No. RSI measures the speed and persistence of recent price changes, not value. A reading below 30 says the falls have been fast and one-sided lately — which is equally consistent with a temporary overreaction and with a company whose situation has genuinely deteriorated. The indicator has no information about which. It is a description of price behaviour, not an assessment of worth.
Why is 30 the threshold rather than some other number?
Because Welles Wilder chose it when he introduced the indicator in 1978, and it stuck through repetition rather than through testing. There is nothing structural about 30 or 70. Some traders use 20 and 80 to make signals rarer, others shift the bands depending on whether the market is trending. The thresholds are a convention, and the measurements on this page are one reason to treat them as such.
If oversold readings are that rare, is RSI useless?
It is not useless, but the rarity means the threshold is doing far less work than most people assume. The more informative use is relative rather than absolute: where a stock's RSI sits against its own recent range, or whether momentum is diverging from price. Waiting for a sub-30 print as an entry trigger means waiting for something that, on the day we measured, was true of fifteen names out of 1,707.
Would these percentages look different in a falling market?
Substantially, and that is the point. We measured on a day when 67.6% of tracked names were above their 200-day EMA — a broadly rising market. In that regime almost nothing gets oversold. In a sharp decline the same 30 threshold can fire on hundreds of names at once. The threshold is fixed; what it means is not.
How is RSI calculated?
It compares the average size of recent gains with the average size of recent losses over a lookback window, conventionally 14 periods, and expresses the result on a 0 to 100 scale. A reading of 50 means gains and losses have been roughly balanced. What matters more than the formula is what it excludes: no volume, no valuation, no news, and no information about anything beyond the lookback window.
Related
Deciding how large a position should be is separate from deciding whether to take it — the position size calculator handles the arithmetic. More lessons are on the learn page, and the free calculators are on the tools page.