Originally Posted by Rain Man (Post 9318445)

May I propose an alternative analysis, magichef? I'm not sure if you have the data to do this or not.

My theory is that there are two reasons to see increases in performance over time (prior to Father Time's inevitable emasculation). First, you have the normal factor of guys gaining more experience and getting more savvy. This is what we think we're seeing in your graph.

However, I think there's another factor in play as well, which is mildly disturbing given your results. I think that the guys who are in their 3rd or 5th or 8th year are likely to be better players overall than the guys in their 1st or 2nd year. They have more innate talent or better work ethics or are smarter or whatever. You have lots of guys who barely make a roster, play a year or two, and then wash out. The guys still playing in Year 7 are better players in general or they wouldn't have survived.

So what this means is that, even if there was no benefit of added experience, your graph should show an increase in performance related to years of experience. This disturbs me because the increase in your graph is somewhat shallow for the first seven years, which probably covers the entire careers of a vast majority of players.

What does this mean? It could possibly mean that there is no real benefit of experience. If you're good, you're good, and the increase we see with added years of tenure is the simple fact that the weaker players are being weeded out of the analysis pool as time goes on.

An alternative analysis would look at each player, and their rankings in their 1st year in the league, 2nd year, and so on. If you want to look at the trend over the first ten years of a player's career, you only examine players with ten years of experience and look at each year of development.

You said you only have five years of data, but I think that would be sufficient. I suspect that most of the benefit of experience occurs in that time frame anyway.

What do you think?

Originally Posted by Rain Man (Post 9318466)

Also, are you charting averages or medians? I think median might be preferable. It may not make a huge difference either way.

Interesting stuff, by the way.

Originally Posted by MagicHef (Post 9318476)

I'm certainly open to it. I have 2008-2012. For guys drafted in 2008, that means their rookie year through their fifth year. For guys drafted in 2003, that would be years 6-10. Would only guys drafted in 2008 be valid for this?

Also, I realized a change I should make. Right now, a reserve player that gets a handful of snaps at the end of a blowout is counted the same as a starter. Rather than counting the number of guys, I should do the number of snaps. I'll fix that tomorrow morning.

Originally Posted by ChiefGator (Post 9318509)

Very interesting.. keep it up!

Like the suggestions from the crazy, "I've got no debt and so no social responsibility" 50 year old as well.

Originally Posted by Rain Man (Post 9318490)

I think the maximum value would be if you structured it like this:

1st year ratings =

2008 rating for rookies in 2008
2009 rating for rookies in 2009
2010 rating for rookies in 2010
etc.

2nd year ratings =
2009 rating for guys who were rookies in 2008
2010 rating for guys who were rookies in 2009
2011 rating for guys who were rookies in 2010
etc.

and so on.

Originally Posted by MagicHef (Post 9318558)

That's how it's set up right now.

Originally Posted by Rain Man (Post 9318542)

I'm as free as a bird now, and this bird you cannot chay-a-aynge, whoa whoa whoa whoa...

Originally Posted by Rain Man (Post 9318578)

Oh, okay. Nice work.

Maybe then you develop a family of curves, one that includes guys with 10 year careers, another that includes guys with 9 year careers, and so on. I could be mentally deficient, but I still think we have some pollution by the fact that the left side is going to include weaker players who aren't destined for long careers.

Originally Posted by MagicHef (Post 9318806)

I'm a little disappointed no one said anything about my Star comment.

Originally Posted by htismaqe (Post 9318127)

Defensive tackles very rarely pan out in free agency. I wonder if some of those spikes are due to contract years.