TickerTruth started from a narrow, specific frustration: backtests on Indian equities that quietly fall apart because a ticker was reused, a symbol was renamed without anyone updating the mapping, or a bonus issue was applied on the wrong date. None of these show up as errors — they show up as wrong numbers that look plausible, which is the worst kind of bug in a trading system.
This blog is where we’ll write about the parts of that problem that don’t fit in release notes:
- Symbol lineage edge cases — mergers, delistings, and the handful of NSE symbols that have been reused across genuinely unrelated companies over the last two decades.
- Corporate action adjustment mechanics — how split/bonus/rights ratios chain together, and where naive cumulative-factor math breaks.
- Data quality tradeoffs — why we’d rather flag a record
LOWconfidence than silently drop it, and what that means for anyone consuming the reference tables. - Pipeline and release engineering — the boring-but-load-bearing details of running a monthly versioned data release without breaking downstream consumers.
If you’re backtesting strategies on NSE or BSE history and you’ve ever had a Sharpe ratio that seemed too good to be true, there’s a decent chance a corporate action adjustment — or a silent symbol swap — is why. That’s the layer we’re trying to make boring and correct.
Posts here will be short and specific: one problem, one worked example, one takeaway. Subscribe via RSS if you want new posts as they land, or check the methodology page for how the underlying pipeline works end to end.