How we score our daily stock ideas, and why we publish the misses
6 min read · updated 11 November 2026
A hit rate is only as honest as its definition. Here is exactly how a daily idea on this site is built and scored, what the back-tests found, and what the percentage you see on each card does and does not tell you.
What an idea is
Each trading day the site publishes up to three ideas per market for the next session. An idea has a reference price (the last close), an upside level two average daily ranges (ATR) above it and a risk level one ATR below it. The levels come from recent volatility, not from a price prediction.
What counts as a success
After the session closes, an idea is a success if the upside level was reached before the risk level, or, if neither was touched, if the day closed above the reference price. If the day's range touched both levels, daily bars cannot show which came first, so we count it as a miss. Every outcome stays on the track record page, wins and misses alike, and rows are never edited after publishing.
The baseline that matters
Because the success test is generous when neither level is touched, a randomly chosen stock succeeds close to half the time. So a hit rate means nothing without that baseline. In our walk-forward back-test (518 Indian sessions, rules tuned on the first 60% and judged on the rest), randomly chosen stocks scored 50.2%.
What the back-tests found
As of 11 November 2026: the India rule, a sharp dip inside a strong, established uptrend limited to the most-traded half of the universe, succeeded on 57.0% of its own picks over 606 calls, against 54.6% without the liquidity filter and 50.2% random. We tried six other filters (news gaps, market breadth, a stock's own past record, weak closes and combinations) and kept only the one that helped on both the tuning and the held-out data.
The US rule is weaker. Its held-out result was 52.9% against 51.8% for random picks, which is not a measurable edge on about a year of history. We publish it anyway, label it, and are extending the history before changing it.
Live versus simulated
Live ideas are published before the session opens. Simulated rows replay the current rule on past sessions using only the data available each morning, scored against real prices. They are labelled and kept out of the live rate, because a replay is not the same as being right in advance.
What the hit rate does not mean
It is not the chance that an idea reaches its upside level, and it is not a promise about any one stock tomorrow. A 57% rule still loses on 43 calls in 100, and a streak of losses is normal. Ideas are published for study, with the risks listed beside them, and the decision always stays with the reader.
Frequently asked questions
What is the hit rate on each idea card?
The share of past next-session calls using the same rule that counted as a success under the scoring rule above, measured on data the rule was not tuned on.
Are the ideas buy or sell recommendations?
No. They are published for information and education only. Indian stocks are shown in an educational framing. Nothing on the site is investment advice.
Why is the US rule shown if its edge is not clear?
Because hiding a weak result would make the rest of the record less believable. The back-test comparison is stated openly, and we will change the rule only when the data supports a better one.
How often are results updated?
India's ideas publish in the evening before each session, after the exchange publishes its official end-of-day data. US ideas publish before the US open. Outcomes are scored after each session.
Keep reading
m-indicator does not give any assurance of returns. Everything here is based purely on research, price trends and various market factors, and is published for information and education only. It is not investment advice. Markets carry risk, so please review independently, and consult a registered adviser if needed, before making any trade. Tax rules change; check the current rules before you act.