Stock-Patterns: Rethinking the Next Iteration

[BennyThadikaran]

I’m incredibly grateful for the support and interest you’ve shown in Stock-Patterns — it’s by far been my most successful project, thanks to all of you who’ve starred and shared it.

While the current detection algorithms work well, there are subtle bugs when iterating over pivots — causing some chart patterns to go undetected.

To address this — and a few other limitations — I’ve been planning an incremental pattern detection system designed to work seamlessly with real-time data. The idea is simple:

• Process one candle at a time
• Build pivots incrementally
• Detect patterns as new pivots arrives

This approach brings several benefits:

• Works with data from CSVs, databases, or websockets (real-time)
• Once historical data is seeded, only the newest candle is needed for updates
• Lets you visualize all patterns — even partial ones — directly on the chart
• Built with pip, fully component-based, extensible with custom user implementations, and UI-agnostic
• Ability to read directly from bhavcopy reports to scan stocks from both NSE and BSE.

Over the next few weeks, I’ll be launching a new repository with concrete implementations and a detailed roadmap. Stay tuned!

[BennyThadikaran]

My new project Precise Patterns is now live. 🚀

I have added some partial working implementation of the code. Let me know your thoughts.

[regi-a11]

Hey Benny,

I've been following your projects for a while now, and I'm really impressed with what you've been building. I’ve been developing Telegram bots focused on gathering and processing information for options trading, and I have several ideas centered around options trading that I think could turn into something really interesting.

I’d love to connect, exchange ideas, and see if we can build something cool together. If you're interested, feel free to reach out to me at reginoldkbinoy@gmail.com.