We built one simple stock score (equal weights on trend, stability, fundamentals and the market backdrop, plus news tone for the US) and tested it the same way in two markets: the S&P 100 and India's Nifty 50.
The setup
- Development data 2020 to September 2025; the final 12 months sealed and opened exactly once after the score was frozen.
- Label: each stock against its index over the next 20 trading days.
- Measure: daily rank IC, the correlation between the score's ranking and what happened next.
The results
| Market | Development IC | Sealed-year IC |
|---|---|---|
| US (S&P 100) | +0.0085 (t 0.46) | −0.0585 (t −2.47) |
| India (Nifty 50) | +0.0187 (t 1.12) | +0.0087 (t 0.21) |
In the US the sealed year came out significantly negative: higher-scored stocks did worse. In India it was flat, indistinguishable from random rankings.
What to take from it
- Neither market shows a usable ranking. India's numbers are not "better"; t 0.21 means nothing either way.
- One market is not a replication. If we had tested only India, a flat result would have looked like a tame failure. The US result shows how far a frozen score can drift.
- India has less data. Four parts instead of five (no per-company news feed yet) and about four years of NSE filings. Small samples make wide error bars.
Both reports, with every candidate weighting we tried, are published on the research page. The score is shown on the site as a summary of each stock's data, not as a forecast.
From Quantooo Lab, the makers of quantooo.com, a free, open and educational quant lab: every formula, every test and every failure is public. See it here: https://quantooo.com/in
Educational, not investment advice.
Top comments (0)