Notes, mostly on things that did not work.

A research programme that only publishes its successes has no way to distinguish skill from a long enough search. These are written up the way we record them internally: the mechanism, the evaluation, and what the holdout actually said.

Notes: 4
Null results: 2 of 4
Coverage: NSE equities; mid/small focus
Figures: Not audited

01 Aug 20266 min read

A sector rotation that matched its benchmark, and why we shelved it

The holdout said the strategy was as good as holding the sectors equally weighted. That is not a win, and the holdout is now spent.

momentum
null result

25 Jul 20265 min read

A 5:1 split that read as a 79.5% drawdown

Exchange close prices are not corporate-action adjusted. If you do not fix that before the signal sees them, you are trading an artefact.

point-in-time
risk

20 Jul 20267 min read

An overlay that reversed sign out of sample

A rotation overlay on NSE strategy indices looked additive in development and inverted on the holdout. The third time this exact failure has shown up.

momentum
risk
null result

05 Jul 20266 min read

Eight factor indices, three real axes

Published NSE strategy indices carry far more overlap than their names suggest. Size is the hidden axis nobody labels.

index construction

Figures in these notes are research output on historical data, published for illustration. They are not audited, not a track record, and not indicative of future results. Nothing here is investment advice or a recommendation to transact.