free autonomous prediction of live data

Time-Series Resources and Live Ongoing Forecasting Competitions

Docs, Leaderboard, Blog, Ratings, Competitions,
Knowledge Center, Slack, README

Microprediction TLDR

When you publish numbers periodically to the community API (using the Python client or directly) you automatically lure reward-seeking forecasting algorithms that monitor streams and make short-term distributional predictions of future data values you are yet to publish.

  • Predictions range from a minute to an hour ahead.
  • Algorithms can be authored by anyone
  • Their success is reported on leaderboards
  • Some use the timemachines package, informed by our Elo rating system.
  • Some are kept private.
  • Some use exogenous data

The reward mechanism encourages convergence to a good steam of probabilistic predictions, and you don't have to do anything. 


If you were given a morning to engineer a system that would automatically include next year's algorithmic advances, could you do it? What about exogenous data you don't yet know about? 

You can now (example) and the use of the microprediction API is eventually hard to beat. Anyone, anywhere in the world, can launch an algorithm to improve the overall quality. Usually, they do so by modifying and running a crawler, which is merely a Python script. They can use forever functions that combine the best-performing algorithms drawn from many different open-source packages, if they wish. Then they can improve or replace them, or find relevant exogenous data. 



You've found which is an educational site for humans. The action takes place via as the machines fight it out. A browser is provided at for the benefit of stream creators and forecasters, where you can see all the currently predicted streams and leaderboards.


These efforts are sponsored by Intech Investments out of the belief that open, collective, bespoke business optimization is vastly more powerful when models and feature spaces are shared. No single company or group can solve repeated prediction, but collectively we can make a dent.

If you are researching time-series and want to release your new method into the wild, or if you wish to collaborate on open-source timeseries algorithms (such as the timemachines package, or maybe benchmarking your own) then consider joining our slack. There's an invite on our repo.


See the README or the resources on this site to get started. If you'd just like to use free on-tap prediction, then what really matters is your imaginative usage pattern. That's up to you but here are some articles from our blog to provoke some ideas. 

A Call for Contributions to a Copula Contest
Algorithms try to predict bivariate and trivariate relationships between five minutely returns of Bitcoin, Ethereum, Ripple, Cardano and Iota. Can you beat them?
Learn More
Anomaly Definition
This article explains how the microprediction platform can tackle the tricky problem of defining anomaly, and also discusses drawbacks with the state of the art.
Read More
Tears of Joy: The Easy Way to Standardize Streaming Data
Life moves pretty fast, as Ferris Bueller once said, and if you don't stop to look around once in awhile you might miss it. He was on to something, no doubt, yet when it comes to streaming data...
Learn More

Quick Start

Algorithms are waiting and watching
Get Predictions
Publish your live data and a swarm of prediction crawlers will immediately generate predictions.
Learn More
Make Predictions
Start crawling our live data and showcase your skills. You’ll be recognized for your expertise and intuition.
Learn More

Follow the Movement