A bit of combine number-crunching

College football and the NFL Draft

Moderator: wab

Post Reply
User avatar
karhu
Head Coach
Posts: 2049
Joined: Sat Aug 27, 2016 10:20 pm
Has thanked: 294 times
Been thanked: 373 times

I was curious about what the combine can tell us, so I put together a little database of prospects and 10 years or so of combine stats from All-Pros. Still working out some of the finer points (for my sins, I'm doing it in MS Access), but here are some preliminary findings.

To get an idea of the relative importance of each event for each position, I took a look at the average numbers for each position and those of recent All-Pros. From there, I got an adjustment factor, which I turned into a bonus factor by only applying it to marks that exceeded those of the average recent All-Pro. Most of these scaling factors are positive, in that they reflect marks that are nominally better than those of the average combine attendee, but some surprising ones, like cone and shuttle times for DBs and weight for RBs, are negative. That, I guess, is for down the line, but for now I've got some decently cool stuff.

So far, I've got complete info (I think) on height, weight, 40 times, vertical, broad jump, cone, and shuttle. I'll add the bench later, since it's the single biggest indicator for offensive linemen. That gives us seven measurements for now.

Andrei Iosivas is the only player to exceed the average All-Pro's marks at his position for six of the seven criteria. The players who racked up five such marks (along with one of my control entries) are, in descending order of their average overall scores:

Bryce Ford-Wheaton WR West Virginia
Blake Freeland OT BYU
Jon Gaines II G UCLA
Khalil Mack - CONTROL DE Buffalo
Deonte Banks CB Maryland
Sidy Sow G Eastern Michigan
Zack Kuntz TE Old Dominion
Cody Mauch G North Dakota St
Cam Smith CB South Carolina
Lukas Van Ness DE Iowa
Jack Campbell LB Iowa
Byron Young LB Tennessee

Just four players did at least one event while exceeding the average All-Pro's marks for height, weight, and the events they bothered to do. Three had three perfect events:

Deonte Banks CB Maryland
Jack Campbell LB Iowa
Byron Young LB Tennessee

Three had two perfect events, along with superior height and weight:

Christian Gonzalez CB Oregon
Jahmyr Gibbs RB Alabama
Devon Achane RB Texas A&M

(Remember, weight correlates negatively with All-Pro status for RBs, relative to all combine attendees at the position, so those little guys sneak into this list. I still need to tweak negative correlations of height and weight, though.)
So much road and so few places, so much friendliness and so little intimacy, so much flavour and so little taste.

Friendship is better than fighting, but fighting is more useful.
User avatar
thunderspirit
Head Coach
Posts: 3829
Joined: Tue Sep 06, 2016 12:51 pm
Location: Greater Chicagoland, IL
Has thanked: 605 times
Been thanked: 603 times

Interesting stuff, even in the beginning stages.

I use similar thresholds when I look at player production — also not foolproof, but a good tool for reference.
KFFL refugee.

dplank wrote:I agree with Rich here
RichH55 wrote: Dplank is correct
:shocked:
User avatar
karhu
Head Coach
Posts: 2049
Joined: Sat Aug 27, 2016 10:20 pm
Has thanked: 294 times
Been thanked: 373 times

thunderspirit wrote: Tue Mar 07, 2023 7:45 am Interesting stuff, even in the beginning stages.

I use similar thresholds when I look at player production — also not foolproof, but a good tool for reference.
Thanks. Yeah, it's my idea of fun and all, but it still only addresses the second most important question of all.

"What did he do?" - Only watching a guy play can answer this one.
"How did he do it?" - The combine can tell us something about this.

George Kittle is still my favorite example of what combine scores can mean.

What did he do? Blocked like a maniac in the Big Ten and caught the one or two balls thrown to him each game.

How did he do it? By being the fastest, quickest, most explosive TE in this or most other drafts.

Well, there's your guy, Ryan--not whoozitz from Ayn Rand U.
So much road and so few places, so much friendliness and so little intimacy, so much flavour and so little taste.

Friendship is better than fighting, but fighting is more useful.
User avatar
IE
Hall of Famer
Posts: 12500
Joined: Mon Nov 03, 2008 8:46 am
Location: Plymouth, MI
Has thanked: 523 times
Been thanked: 700 times
Contact:

Iosivas really stood out to me. I hope Poles grabs him I really do.
2023 Chicago Bears... emerging from a long hibernation, and hungry!
Middleguard
MVP
Posts: 1667
Joined: Fri Aug 31, 2012 5:10 pm
Has thanked: 34 times
Been thanked: 120 times

You're looking at All-Pros, I'm at the bottom of the barrel with UDFAs. Something I've noticed, but haven't even counted, is that WRs that do most tests (dash 40, leap vertically, and jump broadly), but fail to perform at least one of the 3-cone or shuttle at either the combine or their Pro-Day, apparently have something to hide because they tend to go undrafted. Too early to tell this year, because for some reason, a lot of WRs prefer to jump and dash at the combine, but risk looking lost at 3-cone or shuttle in the relative privacy of their Pro-Day.
EricTighe
MVP
Posts: 1219
Joined: Fri Aug 26, 2016 1:13 pm
Has thanked: 15 times
Been thanked: 62 times

Nice work.

I have used the same basic system for 40 years now.

Stats- DO they have the stats to warrant being a player?
Athletic ability- Does their workout match what you see on game day?
Watch them on TV- Do they stand out and who stands out?

So if you have a Player that has great stats. + goes to combine and puts up ridiculous numbers. + When you watch them play they are all the announcers are talking about the whole damn game (for good reasons)= Stud 99% of the time unless, of course, they get hurt. You never can account for injuries.

I am not a big fan of Arm length being a real tool to use. But that's just me.

I wonder what the stat is for short-armed guys getting called for holding versus long-arm guys. You would think the short-arm guys would have the advantage.
Post Reply