Metrics, Why the fascination --Grumpy Old Man Rant

One thing we on here have not touched on is: age & stage of our clientele...

NHL is metrics crazy does it make sense to apply those measures to a group of little tykes who use the same sized puck, net &ice surface as a pro?

The metrics tell us as coaches trends and provide evaluation tools on game planning & personnel.

The pro reality: players are traded, cut, drafted each year. GM's / coaches are constantly evaluating & trading talent. The metrics work bc you are comparing apples to apples. Every NFL kicker can put it thru the end zone on kick off. The point I'm trying to illustrate is their data is more stable & scaleable.

High school reality: data is not comparable in this manner due to the wide variation. For example: OL sizes my guess is some teams here are rolling with dudes 180lbs & others have 5 dudes over 220lbs. The lack of consistency in game play is also an issue.

Think about this when I started as a Skeleton athlete at the developmental level you could be back 0.8 seconds of the next guy after a run and leapfrog 3-4 guys after run 2. At the World Cup / Olympic level if the guy in front of you is ahead 0.05 that's a dog fight, rarely if ever are you jumping 3-4 places after run 1.

High school football = developmental.
Pro football / sports = higher level of consistency less errors = meaningful data

Like what? Can you please post an example that has nothing to do with tendencies and which you action for improvement?

The short answer to your question is, no-we have not been practicing to play the best defense we possibly can.  

I am taking over as DC, and one of the things that I could not stand was to hear the other coaches bitchin about how when we missed tackles, we had to "get better at tackling".  Well, to me it was because we did not spend enough time going over our run fits, and screen fits and the total scheme of "here is where you fit".  So I wanted to see for myself what our issue was.  The last 2 DC's I have worked for always laid the blame on the kids for not tackling, and the truth is we never put them in a position to succeed.  So now I have a solid "Metric" that I can work off of to correct and learn from.  

Now, get your a$$ to bed so Santa can visit you.  Merry Christmas!

If you understand math, you can acquire a decided statistical advantage. If you don't. It you know it, you can start interesting discussions that lead you to interesting places. If you don't know what you're doing and don't realize it you're spinning your wheels and you look like an idiot to those of us who understand.

1.  Our conversation rate on 4th down by yardage distance altered our formula on going for it a bit the last two years.  (two years ago we were conversting at a higher rate last year a lower rate we adjusted accordingly)
2.  Looking for average start point from our opponent on kickoffs has had a major impact on our approach to kick-off.
3.  Looking at data of punt yardage out out bounds vs. returned punts has led to us altering our approach on this.
4.  Looking at opponent punt data of non returned punts vs. rush and not rushing has led us to altering our percentage of return and rushing.
5.  Looking at gained yards after initial contact had an influence on our tackling instruction and leverage instruction (this one might be easy to say well you could see it on hudl that you were missing too many tackles but the fact was we didn't we had a very very good defense that gave up a few big plays in the pass game in our mind but when we went back there were a ton of first downs we were giving up do yardage after contact really led to us to focus on battling and winning the game of inches).
6.  Looking at completion rates created a different approach to coverage keys and assignments

That is a few specifics I don't think we go super deep in the data but I can tell you if we see a trend of something we look deeper into it rather then just reacting too it example from this year:

Week 3 we have a terrible game in one of our offensive sets (stack I) vs. a team we were superior too...  I stated to our staff that we needed to decide if we were going to dump it or not, my DC said go back and look at the data I don't think it was that bad, I pulled scrimmage and game 1 2, and 3 data (which I thought was gonna be horrible) and it was actually not all that bad and compared closesly with our base set, that game was down a tad but my perception was more based on the time and situation that it was being used in.

I went back and did a metric on which called defense gave up the most points per play.

My Victory defense (3 pass rushers, 3 deep LB's, cover 4, plus a extra deep safety) has NEVER been scored on.

My goalline 6-5 that we use inside the 3 yard line gives up points on 35% of plays.

Going to start using our Victory defense on the goalline.

I do want it to be known that I actually do think this is cool and truly am excited for jlenwood to discover this and him also being named DC.

This might derail the thread a little.

But is what he did a metric or just film evaluation?

To me, a statistical metric like we are discussing is 100% objective. Like teams average 4.7 yards per carry on 2nd and long against our 4-3 defense with a Sam blitz. A statistical metric is something a statistician can just hand you on a sheet of paper.

What jlenwood did was a film evaluation in my estimation and that has always been valuable in my mind.

Good points on both sides of the issue. Here is my take: "All things in Moderation".

We all do data analysis of some sort even if it is just mental. Most of us do keep stats and review them, watch film and make note of the good and bad that we see, we self scout and watch film in the off season in an attempt to improve in as many aspects as possible.

Granted some coaches are way more detail oriented in terms of developing a plan to overcome the weaknesses we find. But, all coaches make corrections based on what they have seen and what the film and stats show to be true.

I personally am a pretty detail oriented person. However thru the years I have found that TOO MUCH analysis can just lead to confusion and be overwhelming. Sometimes outcome is based purely on their personnel being better than yours or vice versa. Sometimes it is scheme, sometimes it is field conditions.......sometimes it is just pure luck on their part or on yours.

At any rate I would say if detailed analysis drives you and makes you better, then do it. If you hate it, don't do it.

The argument against statistical analysis seems to be that it's unnecessary to measure the value of things we know to be true (e.g., Sound tackling techniques or explosive plays or having better Jimmys and Joes than the other guy will assuredly lead to victory.) What about pre-game stretching? Everyone from pee-wee football on up to the NFL stretches pre-game. It's an accepted, general truth that stretching before exercise prevents injury and increases performance, right? Except, it doesn't. Multiple empirical studies have found that pre-exercise stretching does not prevent injury, and in some trials, it actually caused a decrease in performance. In one trial, sprinters ran five percent slower when stretching prior to running. This is to say, just b/c we (and maybe even the rest of the world) believe w/ all our heart in something, it does not necessarily make it true.

I agree with all that 

I understand the concept of perception of events, vs numerical data of the events BUT....





in all of these situations, the people who the data represents are NOT going to be the people creating the data next year correct? Punt team will be different--so will opponents return team. Kickoff team will be different, as will opponents.

That said, I DO think that the THOUGHT process here is solid. Not that coaches need to do such analysis, but I do think it benefits ALL coaches to have an actual PROCESS where you think of situations. To THINK "Hmm..our kicker isn't putting it in the endzone, would we be better doing this another way".



You need numbers for that?

Data matters.

Really?  You don't see your center not being able to reach a 1-tech watching film? What are you doing when you watch game film?  And I don't mean that in a flippant way.  Do you really think it will take 20 years for you to see that? Did you really need to run data analysis to see you guys blocked gap better than zone? Do you know why?  What was the problem with the 2 pass concepts?

All that said, I don't have a problem with this kind of analysis as much as do with points per possession, or points per yard or things like that. 

Yes.  The problem is I don't think you guys are implementing a successful blue print if you need to "crank out a HUDL report based on run direction and DL alignment" to recognize that your center was not doing his job correctly.  There is a breakdown in this process. I certainly think there is a breakdown in the process if the person who is making those adjustments/decisions is not a part of recognizing it on film.  The things you are describing should have been taken care of after a Saturday Morning (or friday night) film review.  If the person who has the authority to make those changes doesn't watch the OL, then they better have supreme confidence that the OL coach would bring that to his attention-- "Hey, OC, we are having problems with OZ because Johnny just can't block a shade.  He is stepping with the correct foot, he is bringing his hips, he just isn't strong enough, but he is the best we have.  So lets go ahead and scrap that and just GAP block"   IF the OL coach isn't doing that, then the OC needs to watch ALL the positions. 

THAT is how the process should be handled. 
Of course, I am sure you will now say how you guys have won 3 state titles, and this works for you.. so whatever.