Aging curve for offensive players

Postby **CrimsonKodiak** » Sat Jul 11, 2015 7:00 pm

http://socalledfantasyexperts.com/aging ... -position/

I saw this article, I had a couple of eye opening take away. Thought I would share.

Postby **Dynasty DeLorean** » Sun Jul 12, 2015 5:56 pm

interesting. it looked like the cliff for rbs was around 30 and for wrs around 33.

Postby **ajmyk** » Mon Jul 13, 2015 3:27 pm

Hard to beleive that a 33 y.o seasoned veteran QB is as bad as a his rookie self

Postby **jeffster** » Mon Jul 13, 2015 3:55 pm

I'm not sure I'm buying this. They started down an interesting path, but didn't go far enough. Does anyone think for a second that aging means the same thing to AP, Emmet Smith, Barry Sanders, Peyton Manning, Brett Farve, and so on as it does to non-elite talent? Barry put up over 2k yards at 29 and 1491 yards at 30.

At a minimum I would want to see this broken down by draft position. But that would be hugely imperfect, since we're dealing with entire careers over the course of decades here. Maybe split by guys who make two or more pro bowls versus those who don't. The point is to split out the true physical and talent freaks from the JAGs. I strongly suspect the aging story is different in these cases.

Postby **Dynasty DeLorean** » Mon Jul 13, 2015 3:56 pm

I was listening to a podcast a while ago and they had someone on who studied when players fall off the cliff, and it's not so much that the production dwindles it's more that each year you have a percentage chance to completely fall off. So when you're 30 let's say, you have a 50/50 chance you continue to produce pretty close to what you've been doing, or you suddenly suck. If you're still good that year, then the next year it's another 50/50 chance of falling off a cliff. In most cases it isn't a gradual decline.

Postby **jeffster** » Mon Jul 13, 2015 4:04 pm

They also report only means, and not variances or confidence intervals. This would be another way at getting at the differences by talent. If there's a wide variance, then some guys are being affected by age much sooner and others much later. If there's a small variance, then most guys are dropping off around the same point.

It's like the difference between a series with the values 10,10,10, which has a mean of 10, and the series 15,10,5 which also has a mean of 10.

Postby **CrimsonKodiak** » Tue Jul 14, 2015 6:57 am

jeffster wrote:I'm not sure I'm buying this. They started down an interesting path, but didn't go far enough. Does anyone think for a second that aging means the same thing to AP, Emmet Smith, Barry Sanders, Peyton Manning, Brett Farve, and so on as it does to non-elite talent? Barry put up over 2k yards at 29 and 1491 yards at 30.

At a minimum I would want to see this broken down by draft position. But that would be hugely imperfect, since we're dealing with entire careers over the course of decades here. Maybe split by guys who make two or more pro bowls versus those who don't. The point is to split out the true physical and talent freaks from the JAGs. I strongly suspect the aging story is different in these cases.

I looked into the methodology here and it's quite interesting. They accept that in order to be an NFL player when you are 30 you have to be a phenomenal talent. to survive 8 plus years of getting the snot kicked outta you and still be competitive. So a 30 year old AP doesn't measure up to a 30 year JAG. But a 30 year old AP measured up against a 29 year old AP does measure up. And similarly a 29 year old JAG also compares equally to a 30 year old JAG. So it's the change in an individual compared to himself the previous year.

I was surprised to see how early the 'peak" for players was. Everyone looks at the 30 mark as the drop off in production, but this suggests that for most (not all) the peak is substantially earlier than that.

My eye opener was rookie studs seem to out perform veteran studs in a 3-4 year window, and I'm a big fan of trading picks for proven studs. but I guess the boom/bust phenomenon of rookies plays a factor as well. a 35 year old veteran playing in the NFL still gets more points than a rookie bust that falls outta the league after 3 years (see: TRICH)

Postby **jeffster** » Tue Jul 14, 2015 7:45 am

CrimsonKodiak wrote:
jeffster wrote:I'm not sure I'm buying this. They started down an interesting path, but didn't go far enough. Does anyone think for a second that aging means the same thing to AP, Emmet Smith, Barry Sanders, Peyton Manning, Brett Farve, and so on as it does to non-elite talent? Barry put up over 2k yards at 29 and 1491 yards at 30.

At a minimum I would want to see this broken down by draft position. But that would be hugely imperfect, since we're dealing with entire careers over the course of decades here. Maybe split by guys who make two or more pro bowls versus those who don't. The point is to split out the true physical and talent freaks from the JAGs. I strongly suspect the aging story is different in these cases.
I looked into the methodology here and it's quite interesting. They accept that in order to be an NFL player when you are 30 you have to be a phenomenal talent. to survive 8 plus years of getting the snot kicked outta you and still be competitive. So a 30 year old AP doesn't measure up to a 30 year JAG. But a 30 year old AP measured up against a 29 year old AP does measure up. And similarly a 29 year old JAG also compares equally to a 30 year old JAG. So it's the change in an individual compared to himself the previous year.

I was surprised to see how early the 'peak" for players was. Everyone looks at the 30 mark as the drop off in production, but this suggests that for most (not all) the peak is substantially earlier than that.

My eye opener was rookie studs seem to out perform veteran studs in a 3-4 year window, and I'm a big fan of trading picks for proven studs. but I guess the boom/bust phenomenon of rookies plays a factor as well. a 35 year old veteran playing in the NFL still gets more points than a rookie bust that falls outta the league after 3 years (see: TRICH)

Yes, but then they take the dropoff of 29 year old AP to 30 year old AP, and average it with the dropoff of, say, 29 year old JStew to 30 year old JStew, and report that average as the dropoff for RBs from age 29 to 30.

This would make sense if you got a random RB and only knew their age, and nothing else. You could crudely estimate if they're approaching dropoff. But that's not what happens; you know if you have AP or JStew. The way age affects their likely dropoff is probably different.

On top of that you have guys like Justin Forsett and Zach Stacy. One had a boom late, the other early, which probably had little to do with their age, but potentially skews the average they report.

I'm not saying the idea doesn't have merit. It just needs the question refined farther.

Postby **dlf_jules** » Tue Jul 14, 2015 8:00 am

Here are some charts that might shed further light on the issue. The RB chart doesn't dig into elite/very good/good players, but it's at least a good picture of only fantasy relevant guys. Sorry for any formatting weirdness. Most of these are byproducts/side results of an article I wrote on WR aging a few months ago (DLF subscription required).