jump to navigation

Thin-Slicing Recruiting: Mini-Pitch to Coach John Beilein December 15, 2009

Posted by mgodoublems in Experiments.
Tags: ,
add a comment

I went to the set of Coach John Beilein’s weekly radio show today to see if I could get a chance to talk to him.  For those who don’t know, he is Michigan’s Men’s Basketball Coach.  Coach Beilein and I have spoken a few times at volleyball games throughout the year, and when I first came up with this idea he was one of the first people that I wanted to pitch it to.

I made a short pitch to him before the start of the show, essentially a short version of what I wrote about in the project introduction.  After the show, I showed him this:

After about 20 seconds of explaining, he seemed fairly impressed, and said that I was on to something.  It’s clear that he’s looking forward to what I have to say, and hopefully I will be able to incorporate him into the planning for the project after the season.  It will have to wait until then, of course.  He has more important things to do right now.

The idea is that Coaches and Recruits want exactly the opposite out of each other – Coaches want to see full games and grades, because they will show the flaws in the potential recruit.  Recruits would prefer if the Coach stuck to the highlight reels and stat sheets until their foot is in the door and they have a scholarship – otherwise, their flaws might be too apparent.

What I would like to do is see if it’s possible for a coach to section a small part of a highlight reel that they receive from a player that will teach them far more about the player than the entire highlight reel ever would.  Then, the coach wouldn’t need to see the entire reel, he could section off a small portion of it, make a decision without any cajoling for information the player doesn’t feel as comfortable giving, and then he could send an assistant to see the game.

If a coach can make a reliable decision based on a highlight reel clip, he could analyze more players in a much shorter time period, leaving him time to focus his attentions on the ones most worth targeting.  It’s also more likely that the coach could find ‘sleeper’ players, ones that nobody else is recruiting, but that are significantly beyond the skill level attributed to them.

Thin-Slicing Recruiting: An Introduction December 3, 2009

Posted by mgodoublems in Experiments.
Tags: , ,
1 comment so far

Earlier, I mentioned that I was planning on beginning an experiment based on the principles of Blink.  This is my quick introduction to ‘Thin-Slicing Recruiting’. The premise of this project is to look initially at two sports – Basketball and Football – and the way that collegiate coaches analyze the abilities of potential players.

Hypothesis: Experienced coaches are able to determine the capabilities of a player in their respective sport with a small quantity of information.  Visual information will have a significantly higher success rate than statistical information, and the amount of time required to make a judgement on a player will be short – 30 seconds to a minute.

Experimental Procedure Expose coaches to a variety of information sets on recent high school players.  These information sets will include, but are not limited to, various combinations of: short videos of playing time, short videos of camp time, playing statistic tables, and non-playing statistics (height, weight, strength, speed, etc.).  Include with the videos a questionnaire concerning each target player, tabulate responses, and compare to actual information.


I will be dividing football players into twelve categories.  I will produce video clips of varying lengths of each of the players, until I have a group of 20 players from each category with clips ranging from 5 seconds to 2 minutes.  I will also collect statistical data on the players’ in high school, as well as their data in college for comparison.

The twelve categories are: QB, HB, WR, TE, OT, OG/C, DT, DE, CB, LB, S, K/P


I will be dividing basketball players into the five numerical position categories (1-5).  As with football, I will produce video clips ranging from 5 seconds to 2 minutes, and will collect all necessary statistical data.

I will be posting some of the videos as I produce them.  If any members of the community are willing to devote some time to helping me edit video clips or locate target players, it would be much appreciated – please e-mail me at mathieum906@gmail.com if you would be willing to contribute.