Football - I got tired of having 6 stat tabs open while trading live matches, so I built a sub-5s momentum radar
Like many of you, I trade in-play markets (mostly on Betfair).
But I was constantly missing the exact window when a match shifted because I couldn't physically monitor the xG, dangerous attacks, and possession graphs for N concurrent games.
To solve my own headache, I spent the last few months building an automated momentum tracker. It hooks into live match data and evaluates statistical patterns in real-time.
Instead of guessing, the engine fires an alert straight to my phone when a specific scenario triggers. The rules I currently have calibrated:
- Favourite is Pushing: The favourite team is sustaining heavy pressure in the opposing third.
- Favourite Under Pressure: The underdog is generating dangerous play.
- Goal Opportunity (open game): High bilateral activity, historically preceding an imminent scoring action.
- Taking Control: One side is actively increasing its share of dangerous play after a phase of parity.
- Quiet Game: Pace has dropped significantly, dead game.
The Tech Setup
If anyone is curious about the architecture: it runs on a Java Spring Boot backend, uses an official API for data acquisition, and Redis Streams to keep the latency between a stat change on the pitch and the notification under 5 seconds. Everything is thoroughly tested and containerized.
How to try it
It’s a Telegram bot called MomEx. I’m not linking it directly here because Reddit's automod tends to instantly shadowban TG links, assuming it's a scammy tipping service (disclaimer: this is pure data analytics, no betting picks).
If you want to play around with it, you can find the link on the X profile (@mom_ex_)
The base version is completely free and lets you track a couple of rules.
If you want to test the full engine (unlimited alerts, deep stats like xG and shots on target), just type /trial in the bot for 7 days of Premium access. No credit card, no catch.
Would love to hear from other punters / traders on what metrics you find most predictive during build-up phases!