FFAnalytics Web App Accuracy Tab
Curious about which fantasy football projection sources actually deliver the most accurate results? In the past, we have published articles on the accuracy of projection sources, but we’ve also built tools that give users a way to examine the accuracy of projections themselves through the FFAnalytics web app.
The Accuracy tab gives the user access to historical projections and actuals from various sources to get a view of how well the sources performed in a given season. We’ve recently made updates to this tool with new features that make it easier to evaluate projection sources with enhanced visualizations, filters, and statistical summaries.
Below is an overview of the accuracy tab on the web app and how you can use it to examine and compare historical accuracy across sources. In order to use the Accuracy tab in the web app, you’ll need an FFA Insider subscription.
Calibration Plot
The updated calibration plot shows how closely actual player performance aligns with projected fantasy points.
- Each point represents a player’s actual vs projected fantasy points.
- The diagonal line reflects perfect accuracy and points on the line mean the projections were perfectly accurate.
- The green curve shows the relationship between actual and projected points. The closer it follows the diagonal line, the more accurate the projections are.
- Points above the diagonal line indicate players who outperformed projections, while points below show players who underperformed.
- The R² value measures overall accuracy, where larger numbers reflect more accurate projections.
- The calibration plot uses the user’s current settings defined on the Settings tab.

Mean Absolute Error (MAE) Bar Chart
The MAE bar chart quickly highlights the most accurate projection sources and how the projections compare to one another.
- Each bar shows the average difference between action and projected points.
- Shorter bars indicate greater accuracy.
- The plot updates based on your selected year and week.
- FFA Average: Uses the straight average across all sources available for the selected season/week.
- FFA Weighted: A weighted average using FFA’s weighting system, which gives more influence to historically more accurate sources.
- Current Setting: Uses the settings defined by the user on the Settings tab. You can exclude specific sources from the average calculation and adjust the weight given to each remaining source.

Projection Bias Violin Plot
The violin plot provides insight into the spread of projection errors.
- Points reflect the difference (bias) between actual and projected points.
- Positive values: players who exceeded their projections.
- Negative values: players who fell short of their projections.
- The width of the violin at different points shows how common certain levels of bias are.
- A wider shape near zero indicates highly accurate projections.
- The plot updates based on your selected year and week.
- FFA Average: Uses the straight average across all sources available for the selected season/week.
- FFA Weighted: A weighted average using FFA’s weighting system, which gives more influence to historically more accurate sources.
- Current Setting: Uses the settings defined by the user on the Settings tab. You can exclude specific sources from the average calculation and adjust the weight given to each remaining source.

Filtering Options
You can filter projections based on:
- Actual and Projected Points Range: Focus your analysis on a specific range of projections or actual performance.
- Position and Player Filters: Narrow your view to a specific position (QB, RB, WR, TE) or individual players.
Comprehensive Accuracy Metrics Table
For users who love diving into the numbers, the stats table offers detailed metrics to evaluate projection accuracy:
- Includes metrics such as R², RMSE, MASE, ME, MAE, MSE, MPE, MAPE, SMAPE, RMSLE and Predictive R².
- For more information on these metrics, use the Show Column Definitions icon on the table to see a detailed explanation of each metric.
Try It Out for Yourself
Using the Accuracy tab on our web app gives you access to tools to assess and examine how accurate various projection sources really are. You now have new and improved visualizations to compare sources to help you better understand historical projections.
Check out the new Accuracy tab today and we hope you enjoy digging into the historical accuracy data!