About Us

Est. 2022 | Columbus, OH

Chris Ellis

Founder

  After graduating with a 4.0 from The Ohio State University with a major in Finance, specializing in investments, Chris began his career in Columbus, OH working first as the Head of Finance at a beverage startup (Life Support) and then as the Director of Finance at an auto-insurance startup (Root Insurance). During this time Chris applied for law school while simultaneously teaching himself to program and work with big data. After earning acceptances to Harvard, Yale and UChicago law schools (among others), Chris decided to forgo law school to pursue a career in Data Science and Predictive Analytics.

  Chris joined the Data Science team at Root Insurance and spent 3 years building predictive models that influenced almost every major decision within the company. He then joined Meta in 2022 as a Lead Data Scientist where he worked in the social VR space, utilizing analytic techniques to guide product improvements and business decisions. Chris started teaching Data Science in 2022, serving as a Bootcamp instructor for Rutgers University where he taught a 24-week Data Science bootcamp.

  Since graduating from college, Chris has worked on an automated investing program that combines his years of investing experience, his formal education in investments and his expertise in Data Science and predictive modeling. When others asked for access to the program he built, this site was born.

Our Data

Use our vast collection of data to inform your investment decisions. Hundreds of servers scrape new data as it becomes available. Informative visualizations to go with each data type.

Company Data

  • Stock Price History

    View price and volume history for each stock.

  • Financial Performance

    As reported, and on a per-share basis, access financial results from each company's income statement, balance sheet and cash flow statement.

  • News

    Sourced from multiple sites and outlets

  • Discussions

    Comments and threads mentioning each company sourced from Reddit, Yahoo Finance and others so that you always have a pulse on discussions about each company.

  • Derivatives

    Call and Put option trading activity, including open, high, low and close for each option for all available expirations.

  • Insider Transactions

    See which insiders are buying or selling.

  • Fund Transactions

    See which mutual funds, index funds and/or hedge funds own the company's shares.

  • Analyst Recommendations

    Price targets and recommendations.

Market / Economic Data

  • Index Price History

    View price and volume history for each index.

  • Index Derivatives

    Puts and Calls on each index with tradable options.

  • Interest Rates

    Yields on U.S. Treasuries.

  • Inflation

    See how inflation is trending in aggregate, and at more detailed levels.

  • Employment

    Track unemployment figures and jobs reports.

Our AI Models

Our servers continuously create, train, backtest and evaluate predictive models to create stock picks, and the best of the best are made available for all users. Of the thousands of models that have been trained, hundreds make predictions on a regular cadence. As predictions are made, performance is evaluated and published, while improvements are continually made.
Stock-Specific

Models trained to predict a single stock's movement.

Multi-Directional

Our models predict probabilities of prices increase as well as decreasing.

Most Predictive Data

While our data store contains thousands of data points on every stock (in addition to economic and market data), each model utilizes a unique subset of the available data based on which data points prove most predictive of each model's target.

Multiple Time Horizons

We train models to predict price movements at multiple different time horizons, with each model utilizing a different amount of historical data.

Ensemble Models

We train models based on predictions made by multiple component models. This includes sector models, that use predictions from each stock in a sector to predict performance for the entire sector.

Continually Improving

Our models are retrained on a regular cadence, allowing the models to adjust to changing market conditions.

Backtested Results

Each model is backtested - trained only using data available through a particular date and then evaluated on predictions made after that date - to demonstrate how the model would have performed historically.