Articles related to the performance of our models
Get expert NFL Week 12 betting picks with unit recommendations and data-driven analysis. Our GridironAI models break down every matchup — spreads, totals, and player props — to find the smartest bets and optimal unit sizes. See which teams and players offer the best value for your bankroll this week.
By David · November 20, 2025
Let’s deep dive into our backtesting results to better understand how our predictions fared. When we created the model for predicting who will cover the spread, we left out all data from week six and later from 2019 to evaluate its performance. The sample is small, but it’s important not to fly blind.
By Andy Troiano · September 10, 2020
Let’s deep dive into our backtesting results to better understand how our predictions fared. When we created the model for predicting who will win games, we left out all data from week six and later from 2019 to evaluate its performance. The sample is small, but it’s important not to fly blind.
By Andy Troiano · September 10, 2020
An essential part of machine learning and artificial intelligence is evaluating the results of your models. My goal here is to provide some standard metrics from our 2019 season that are used to assess our site.
By Andy Troiano · July 29, 2020
In this article, we review some of the rookie projections from the 2019 season and provide our feedback on what worked and didn't work.
By Andy Troiano · May 08, 2020
When we build models, we apply standard evaluation metrics against a holdout dataset (data not used in any action related to the model building) to evaluate performance. Once we have what we think is “good” performance, we compare the results of the players in our hold out dataset to their actual fantasy performance and the fantasy performance of 2 other sites, ESPN and 4for4. The overall dataset we have for ESPN and 4for4 is not complete, so we filter down to just the players that exist in all three sets of data.
By Andy Troiano · August 27, 2019
When we build models, we apply standard evaluation metrics against a set of data that the model never sees. This set of data allows us to evaluate performance objectively. In addition, we have developed non-standard methods that apply directly to models created for fantasy sports.
By Andy Troiano · July 26, 2019