Predicting the NFL Draft with Combine Results – Linear Regression Analysis for Tight Ends

Welcome to the third and final edition of this series! I may bring it back to provide data since 2000 if the series performs well. To avoid any confusion, I recommend going back to the first two articles if you have not read them already.

Today, we are going to look and see what statistics matter the most to scouts in drafting a tight end. 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.

SUMMARY OUTPUT
Regression Statistics
Multiple R0.10422155
R Square0.010862131
Adjusted R Square-0.099042076
Standard Error57.66092372
Observations11
ANOVA
 dfSSMSF
Regression1328.597249328.5972490.09883272
Residual929923.039113324.782124
Total1030251.63636  
 CoefficientsStandard Errort StatP-value
Intercept189.6947013205.94398440.921098530.38102638
Speed Score-0.6277487451.996804226-0.3143767110.76040042
Speed Score Simple Linear Regression

Unfortunately, we cannot conclude that speed score was statistically significant in determining when a tight end got drafted. In fact, none of the variables I tested end up statistically significant. If we had a bigger sample size I think we would have better odds. Let’s take a look at the two players that might have thrown this off that led us to the conclusion that speed score is not statistically significant or relevant for tight ends.

Stephen Sullivan – With a speed score over 105, it seems like Stephen would have been drafted sooner. I believe the combination of being overshadowed by Thaddeus Moss (who went undrafted) and posting the worst 3 cone and shuttle combine times, of the drafted tight ends in his class, led him to slipping despite having an above average speed score.

Albert Okwuegbunam – This Missouri tight end, with a fantastic speed score of nearly 127, slipped in the draft. He may have slipped because he did not participate in any of other NFL Combine drills and may have lingering injuries (knee). The Denver Broncos take a chance on potentially yet another injured tight end (Jake Butt) and former teammate of quarterback Drew Lock. I will be watching Albert closely to see if he can reach his high potential.

The small sample size and uniqueness of certain prospects may have thrown our data off a bit. Throwing out the irregularities may have found some of these variables statistically significant, but i don’t find it fair to pick and choose the data. If I was confident of a unique reason these players weren’t drafted higher up then maybe I could do it with an asterix next to it.

What I learned from this experience is that it’s not all about the data. Here, we can clearly see there are intangibles and other variables that we cannot see with statistics. It’s likely with more data we could see a different story, but the potential story shown here is that there may have been more of an emphasis on the whole picture and not certain metrics with certain prospects in this years draft class.

-Cody, Founder of Sports Confidant

Processing…
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:

WordPress.com Logo

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

Google photo

You are commenting using your Google 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: