PDA

View Full Version : PMD's Draft Research


Paranoidmoonduck
06-15-2010, 12:34 AM
Success and failure rates get thrown around a lot in regards to the NFL Draft, and it's always something that's bugged me. The definition for success versus failure in the NFL is a somewhat nebulous thing and it definitely isn't as black and white as people tend to think. A few years ago, I started a project to actually calculate success and failure rates, then abandoned the project for much more ambitious version, then abandoned the whole thing. I was poking around old documents today and found the cursory work I had done for the original idea. With a couple hours of work, I finished the project and figured some people here would find it interesting.

The first step was setting up a somewhat objective manner with which to decide whether a player was successful or not. I came up with a three tiered system that looked like this...

1st Level: Produced like a starter for any NFL team for six or more seasons
2nd Level: Fulfilled the first level and was elected to one or two Pro Bowls
3rd level: Fulfilled the first two levels and was elected to three or more Pro Bowls
Bust: Any player that failed to reach the 1st level criteria

Now, I know that Pro Bowls are generally a pretty crappy way to measure success, but I honestly couldn't find one that wouldn't rely on my own football knowledge (since I went back to years of football I had never experienced firsthand). I had a much more complex system that involved a combination of Pro Bowls and All-Pro awards (which are more reliable), but that was more to find numbers on those who were exceptional. In this simple system, Pro Bowls was about as good a barometer as I was going to find.

I left the definition of the 1st Level a bit vague on purpose. There are players who started over 10 games for at least 6 years in the NFL (Joey Harrington was one) who didn't produce like successful starters and there are players who didn't start bunches of games but put up great numbers. I wanted to have some wiggle room to include or preclude players who were on the bubble. It still isn't perfect as some players missed the cut, but as I said, I wanted this to be as objective as possible while still being useful.

The biggest struggle was players who fulfilled the 2nd or 3rd level criteria's, but didn't match the 1st level. Between 1990 and 2002 (the range I choose to analyze), there were 11 such players (and if I had completed this when I started it, Ricky Williams would have joined that list; his 2009 season pulled him into matching the 1st Level criteria). My solution was to not include them in the numbers of qualified players, but to include the Pro Bowls they made in the Pro Bowl numbers.

There's more I could say, but I figured I'd leave some questions to be asked. Without further delay, here's the extent of my findings...

Between 1990 and 2002, there were 385 players drafted in the 1st round of the NFL Draft. Of those players drafted, 209 qualified for at least the 1st Level (54%) and 136 made it to the Pro Bowl at least once (35%).

Over the same stretch of time, there were 130 players taken in the top 10. Of those players picked that high, 85 qualified for at least the 1st Level (65%) and 63 made it to at least one Pro Bowl (49%).

I'm including the document I wrote all this up at, which can be accessed here (http://docs.google.com/Doc?docid=0AT5oTy064ue7ZGZ2cjV4OXpfNTRkdm5qa3FkMw&hl=en), if anyone wants to take a closer look.

ncst8fan83
06-15-2010, 02:34 PM
Success and failure rates get thrown around a lot in regards to the NFL Draft, and it's always something that's bugged me. The definition for success versus failure in the NFL is a somewhat nebulous thing and it definitely isn't as black and white as people tend to think. A few years ago, I started a project to actually calculate success and failure rates, then abandoned the project for much more ambitious version, then abandoned the whole thing. I was poking around old documents today and found the cursory work I had done for the original idea. With a couple hours of work, I finished the project and figured some people here would find it interesting.

The first step was setting up a somewhat objective manner with which to decide whether a player was successful or not. I came up with a three tiered system that looked like this...

1st Level: Produced like a starter for any NFL team for six or more seasons
2nd Level: Fulfilled the first level and was elected to one or two Pro Bowls
3rd level: Fulfilled the first two levels and was elected to three or more Pro Bowls
Bust: Any player that failed to reach the 1st level criteria

Now, I know that Pro Bowls are generally a pretty crappy way to measure success, but I honestly couldn't find one that wouldn't rely on my own football knowledge (since I went back to years of football I had never experienced firsthand). I had a much more complex system that involved a combination of Pro Bowls and All-Pro awards (which are more reliable), but that was more to find numbers on those who were exceptional. In this simple system, Pro Bowls was about as good a barometer as I was going to find.

I left the definition of the 1st Level a bit vague on purpose. There are players who started over 10 games for at least 6 years in the NFL (Joey Harrington was one) who didn't produce like successful starters and there are players who didn't start bunches of games but put up great numbers. I wanted to have some wiggle room to include or preclude players who were on the bubble. It still isn't perfect as some players missed the cut, but as I said, I wanted this to be as objective as possible while still being useful.

The biggest struggle was players who fulfilled the 2nd or 3rd level criteria's, but didn't match the 1st level. Between 1990 and 2002 (the range I choose to analyze), there were 11 such players (and if I had completed this when I started it, Ricky Williams would have joined that list; his 2009 season pulled him into matching the 1st Level criteria). My solution was to not include them in the numbers of qualified players, but to include the Pro Bowls they made in the Pro Bowl numbers.

There's more I could say, but I figured I'd leave some questions to be asked. Without further delay, here's the extent of my findings...

Between 1992 and 2002, there were 385 players drafted in the 1st round of the NFL Draft. Of those players drafted, 209 qualified for at least the 1st Level (54%) and 136 made it to the Pro Bowl at least once (35%).

Over the same stretch of time, there were 130 players taken in the top 10. Of those players picked that high, 85 qualified for at least the 1st Level (65%) and 63 made it to at least one Pro Bowl.

I'm including the document I wrote all this up at, which can be accessed here (http://docs.google.com/Doc?docid=0AT5oTy064ue7ZGZ2cjV4OXpfNTRkdm5qa3FkMw&hl=en), if anyone wants to take a closer look.

How can 130 players be picked in the top 10 in a span of less than 13 years?

Paranoidmoonduck
06-15-2010, 03:34 PM
How can 130 players be picked in the top 10 in a span of less than 13 years?

Sorry, that's a typo. The span was 1990-2002, as it said earlier in the post and indicated in the document.

Way to quote the entire OP in the first response, by the way.

BeerBaron
06-15-2010, 06:22 PM
Interesting...basically speaking, just a little over half of first rounders succeed in any capacity, and just a little over a third of first rounders reach at least 1 pro bowl.

I think it confirms just how big of crapshoot the draft really is...

Paranoidmoonduck
06-15-2010, 06:32 PM
Interesting...basically speaking, just a little over half of first rounders succeed in any capacity, and just a little over a third of first rounders reach at least 1 pro bowl.

I think it confirms just how big of crapshoot the draft really is...

What really struck me as I was doing this was how many former 1st round picks barely managed to stay in the NFL more than two or three years. It wasn't like most of the guys not qualifying for the first tier didn't because they were mostly bit players, some guys just failed spectacularly.

I don't have numbers on it, but I saw less and less of that as I looked over more contemporary drafts. It might be a sign that the importance placed on analyzing a prospect's attitude and mental toughness is gaining steam with good results.

BeerBaron
06-15-2010, 06:36 PM
Could also be that teams are a lot less willing to give up on a guy given the money they have to pay him, particularly the high first rounders. And besides the money, it puts the GM/Coach/Scouts all on the line for making a bad call, and admitting failure is more likely to get them canned.

Even a decade ago, it seemed like coaches and front office people were given longer to try and make things work....the era of "5 year plans" and such. Now, you maybe get one honeymoon year and if you don't show improvement after that, your on the hot seat.

bored of education
06-15-2010, 07:06 PM
holy hell pmd. i'd like to break this down but i really cant since my add wont allow. but wow. excellent work. some interesting **** for sure. ill highlight some stuff when i get a chance, but wow.

its hard to charecterize the levels using certain perameters but i see what you did. very nice

btw, you will be bussing Jerry Hughes tables at his country wide steak house after he gets in to Canton!

bored of education
06-15-2010, 07:12 PM
i am awesome *edit*

Paranoidmoonduck
06-15-2010, 07:12 PM
BCE, this is not a thread for you to post your predictions. You're welcome to start your own thread if you want to.

And thanks boe, I'd love to hear your thoughts when you get a chance.

Paranoidmoonduck
06-15-2010, 09:56 PM
sorry for the lost posts, PMD. thought i'd clean up your thread, given that the premise is fairly intriguing. i fully understand the lack of interest in the more in depth research, but i'm still a little disappointed it isn't here, just for curiosity's sake.

I sort of anticipated that there wouldn't be a huge number of posts, but the important thing was to get the information out there. I know a lot more people have read through it than have commented on it.

And thanks for the clean-up, it's most appreciated.

Shane P. Hallam
06-15-2010, 10:08 PM
Love the work and something I will look over more in a few weeks for sure. Thank you so much PMD.

LonghornsLegend
06-27-2010, 03:19 PM
This is good stuff right here PMD, was a very interesting read. I had actually read it over a few times before commenting. Your one of the few around here who I enjoy reading their stuff, especially mock drafts and stuff like this.