Predicting the NFL Draft with Combine Results – Linear Regression Analysis for Wide Receivers

How do general managers draft players? What facts do they look at in deciding who to draft? Is it based on numbers, metrics, and statistics? Or is it more closely related to the eye test, college success, or intangibles? In this article, I explore the answer to these questions and how do NFL general managers factor statistics in deciding what prospects will be successful at the NFL level.

I decided to start this series not to make monumental observations in the world of sports analytics but to help piece together what NFL scouts and teams take into account when drafting players. I decided to take a look at the data and find what tests at the combine matter the most to each position. And how much of the story does it tell? Who knows, maybe this could help recruits best utilize their time to improve their draft stock. Today, we are going to look and see what statistics matter the most to scouts in drafting a wide receiver. If the results come back that a certain metric is statistically significant, then we can safely assume NFL teams see these statistics as predictors for success in the NFL at their position. For determining if a metric is statistically significant, the p-value needs to be less than .05.

Before we get started, speed score is a metric used to determine how efficient a player is at the 40 yard dash given how much they weigh. For example, if a heavier player has the same 40 time as a lighter player, the heavier player is a more impressive prospect because he has more momentum and most likely is more athletic/powerful. A simple linear regression with speed score as the only independent variable gives a p-value of .0079, making speed score statistically significant in determining when wide receivers were drafted in this years draft. We can see from the graph that some players with a below average speed score were draft quite high. While statistically significant, speed score does not tell the whole story, with an R Squared value of .195 (the higher this number, the more of a story it tells).

Every year the 40 yard dash seems to be the most hyped combine event. With how important speed score is becoming, this information makes sense. However, does the 40 yard dash itself still hold importance? A simple linear regression with 40 yard dash times as the only independent variable gives a p-value of .0348, making the 40 yard dash statistically significant in determining when a wide receiver was draft in last years draft. Not as important as the speed score (as I initially hypothesized). The graph does show an interesting story from this years impressive wide receiver class; most highly valued prospects had a 40 yard dash time below 4.5 seconds.

If you were curious, height and weight were both statistically insignificant. Also, please note that this years combine was impacted by the coronavirus. Hopefully, I will be able to look and compare data from the previous 20 combines in a future post.

-Cody, Founder of Sports Confidant

Success! You're on the list.

Published by Sports Confidant

Sports Analytics and Fantasy Sports Enthusiast

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: