NFL Draft prediction rankings

Posted by

NFL Draft day has always been among my favorite sporting events of the year. While no games are played, it can often make or break a season for teams, and the combination NFL and college football, scouting, and statistics makes it the trifecta for a sports fan.

My least favorite part of the draft is Mel Kiper Jr., ESPN’s self-proclaimed draft guru. He does know a lot about college football, but I’m turned off by his bombastic nature. I wondered, does Kiper know what he’s talking about, or is he really full of hot air?

I took Kiper’s mock draft, along with 10 other football “experts” to see who came out on top for predicting the draft. Here’s the lineup I looked at…

  • Kiper – ESPN
  • Bill Simmons – ESPN (first time doing a mock draft)
  • Peter King – Sports Illustrated
  • Pete Prisco – CBS
  • Pat Kirwin –
  • Chris Steuber –
  • Doug Farrar – Yahoo
  • Cory Bonini – KFFL
  • Clark Judge – CBS
  • Chad Reuter – CBS
  • Rob Rang – CBS

I took their mock drafts they posted on Wednesday (I had to draw the line somewhere; Kiper was on version 3 or 4 by that point), and compared their selections to the final results. I scored them based on two criteria:

  • # of correct picks (either correct numerical slot or correct team – since there is so much trading that happens).
  • Average number of picks off (# of Predicted Pick – # of Actual Pick for each player). For example, Kiper guessed #4 for Russell Okung and he went to #6 Seattle, so the number of picks off is “2”). If the player was not selected in the first round, I considered him drafted in slot 33 since we will not know his true draft slot until the draft resumes.

The winner of most # of correct picks was Doug Farrar of Yahoo. Peter King was a close 2nd, and Bill Simmons was 3rd. As you can tell, it’s tough to actually get a pick right, as Doug nailed only 10 of the 32 picks.

# of correct picks is kind of like measuring Olympics success by only counting gold medals. I wanted a more holistic view.

Next, I looked at average number of picks off. The rationale here is that while you might not predict the precise slot and team for each draftee, you should be able to select player in the right range. I think this is the superior method for measuring prediction success, as every pick matters (including horrible picks like Kiper slotting Jimmy Clausen to Oakland in pick #8).

Peter King dominated this contest. For each player, he was within 4.5 slots on average, quite a feat. Bill Simmons was second, off by 4.9 slots.

So who’s the champion? I took the average and standard deviation for each set of metrics, and compared how each entrant fared, plotted on a 2 X 2 matrix. Using the interaction between the rating systems, Peter King is the overwhelming winner and the earns the title “Prognosticator of prognosticators.” Bill Simmons is the clear runner up, not bad for his first mock draft! Noticeably absent from the winner’s circle is Mel Kiper Jr. However, I do give him props for being among the top half of “experts” and admit he’s not all hot air.

Other highlights

  • Peter King, Bill Simmons, and Chad Reuter were the only ones to identify Tim Tebow as a 1st round draftee.
  • Peter King also won the title of correctly identifying the most number of players to go in the 1st round. He correctly identified 28 of 32 as 1st rounders.
  • Right team, wrong place – Only 2 players were identified as being drafted by the correct team, only in the wrong slot, by our panel. SD traded up to pick Ryan Mathews, and Dallas traded up to nab Dez Bryant.
  • Nobody picked Jacksonville’s Tyson Alualu to go in the 1st round, let alone the 10th pick.
  • Of the CBS crew, Prisco won top honors, which is not surprising since he is the lead analyst. I highly recommend his weekly score predictions as well, because he knows what he’s talking about.

Leave a Reply

Your email address will not be published. Required fields are marked *