🏠 Home 🔍 Today's Scan 📰 Daily Report 📈 Hit Rate 📊 Base Rate ❓ Methodology 📚 Learn 🪙 Crypto Scanner 📋 All Tools ⏪ Investment Simulator 🧾 Tax Calculator🧮 Pension vs. Direct ⚡ Leverage ⚖️ Rebalancing 💹 DCA 📉 Averaging-Down B/E 💰 Dividend Calendar
← Back to the Learn hub
📊 Real Measured Data · Statistical Traps

The Volatility Illusion
— Not a Good Signal, Just a Stock That Whipsaws

If a signal has a high hit rate, is the signal good, or did it just pick a stock that naturally moves a lot? Real measured data shows the latter effect dominates overwhelmingly.

Written by Dawn · IT Engineer · Published
💡 Key takeaway — The target-hit rate is driven far more by a stock's volatility than by any signal. Comparing hit rates without controlling for volatility means nothing.

Real Measurement — a Hit Rate That Splits on Volatility Alone

We split the entire universe into 4 quartiles by ATR (average true range) and re-measured each bucket's base rate (the rate of reaching +15% within 20 trading days). We didn't look at any signal at all.

Sample: 63,912 historical backfilled episodes (15,978 per bucket) · as of 2026-08-14.

Volatility BucketBase Rate (+15% Reached)Meaning
Q1 (lowest)4.0%About 1 in 25 stocks
Q217.8%
Q335.3%
Q4 (highest)51.7%About 1 in 2 stocks

Same target (+15%), but the hit rate spreads from 4.0% to 51.7% — more than a 12x gap. And we didn't look at a single signal.

The Illusion This Creates

Now say some signal has the property of showing up often on high-volatility stocks. A volume spike, a gap up, a large breakout — that describes most surge-related signals.

Its hit rate then naturally comes out high. But that's simply the signal picking out stocks that already move a lot on their own — not the signal predicting the future. It's like reading "more people carry umbrellas on rainy days" as "umbrellas cause rain."

⚠️ The Practical Implication — A high-volatility stock hits +15% often, but it also hits −15% often. Entering based on the hit rate alone means buying the downside risk right along with it. You need to look at a volatility metric alongside it.

How to Filter This Out — Stratification

The fix is to compare within the same volatility bucket.

  1. Split the universe by volatility (e.g., into 4 quartiles).
  2. Within each bucket, compare the hit rate for stocks with the signal versus without it.
  3. Only accept a signal as real if it shows a consistent edge across every bucket.

If the lift survives even in the low-volatility bucket, that signal is more likely carrying real information, not just acting as a volatility proxy. Conversely, if it only shows an edge in the high-volatility bucket, that's just another way of saying "it picked high-volatility stocks."

Why This Trap Is So Common

  • It produces a number that looks good — without stratifying, the hit rate comes out high, so there's no incentive to bother splitting it.
  • The sample shrinks — splitting into 4 quartiles cuts each bucket's sample to 1/4, making it harder to reach statistical significance.
  • Volatility isn't visible — just looking at a chart, it's hard to feel the volatility difference between two stocks.
📌 Summary — When you look at a hit rate, also ask "what kind of stock does this signal mostly show up on?" If it's mostly small-cap, high-volatility names, a large part of that high hit rate may be coming from the stock's character, not the signal. This is why DawnScan requires lift to hold up across every volatility bucket as a condition for adopting a signal.
📮 Daily US Market Morning Brief — We send an analysis of the previous day's top 10 US gainers (TOP10) and what they had in common, every day at 8am (KST). Telegram @dawnbrief · Free · No ads · Not stock recommendations.

Frequently Asked Questions

Why does high volatility mean a higher target-hit rate?

Volatility is literally how wide a price's swings are. A stock that moves 5% a day builds up a very large cumulative range over 20 trading days, so naturally it's more likely to brush past +15% at some point. A stock that only moves 0.5% a day, by contrast, structurally has a hard time reaching +15% in the same window. In our real measurement, the lowest volatility bucket came in at 4.0% and the highest at 51.7% (as of 2026-08-14, 63,912 backfilled episodes).

So should I just buy high-volatility stocks?

No. A high-volatility stock hits +15% often, but it also hits -15% often. Entering based on the upside hit rate alone means taking on an equal-sized downside risk too. Volatility is both opportunity and risk, and the hit rate alone can't tell you which way it'll actually resolve.

What is stratified verification?

It's a method of splitting the comparison group by confounding factor and comparing only within the same condition. Split into 4 volatility quartiles and compare signal-present versus signal-absent within each bucket, and you can separate the signal's effect from volatility's effect. Only a signal with a consistent edge across every bucket is trusted.

Can an individual investor check this too?

A simple way is to check the ATR or beta of the stocks a signal you're watching shows up on. If the signal's stocks are mostly small-cap and high-volatility, you should suspect that a large part of that signal's hit rate is coming from the stock's character.
📏
ATR Volatility
How to read the average-true-range indicator
📊
Base-Rate Proof
Real measured lift per signal, published
⚡
Beta and Volatility
A comparison standard for stock volatility

Related Reading