keggerz Posted July 10, 2013 Share Posted July 10, 2013 First of all, props for doing this data research. Second, I was also confused by the metric name and description in the first post, but I understand it now. I'd suggest you come up with a new name for it, at least. I think the easy way to determine the value of this metric is to make a list of all the RBs from 2011 and 2012 whose end-of-year performance was drastically different from their preseason ADP (both breakouts and busts). Then, find their FD/TD efficiency metric from the year before their "surprise" season. That will either define this metric as a solid predictor or a meaningless indicator of O-line/scheme/etc. Carry Efficiency (CE) would be a good metric...I also think to really give this teeth you need to find a factor to apply to each carry based on down and distance. A 7 yard carry on 1st and 10 needs to be positively weighed, not just the carry that goes for 3on 2nd and 3 to get the first. Hate to say but field position would also be another great thing to build into it. At first maybe CE is measured by any carry that results in a 1st down or goes over a set number, say the league average for YPC...maybe. Just some thoughts is all Quote Link to comment Share on other sites More sharing options...
bostonsoxandy Posted July 10, 2013 Author Share Posted July 10, 2013 Keggerz and Swiss -- I like the name Carry Efficiency (CE). That seems most accurate for this stat. However at this point I am slightly reluctant to to digg in and apply any type of factors for field position, down, or distance simply because of the time it would take. Right now it is relatively quick to calculate as both total carries and first downs are available as stats for each player every year. If I applied any type of factor like you are saying I would have to go game-by-game, probably play-by-play to determine the #'s. I agree that it would be more valuable that way but I think I might go insane in the process. And Swiss -- I will do that sometime this week (busts/booms from ADP and their CE the year before). That is definitely a good way to find any predictive power. Quote Link to comment Share on other sites More sharing options...
bostonsoxandy Posted July 10, 2013 Author Share Posted July 10, 2013 (edited) I'll add more when I have the time: 2012 RBs: (ADP: # taken in RBs) LeSean McCoy (ADP:3, Finish: 21) CE% year before: 37.0% *missed 4 games in 2012 Darren McFadden (ADP:4, Finish: 28) CE% year before: 23.9% (small sample size) *missed 4 games in 2012 Chris Johnson (ADP:5, Finish: 13) CE% year before: 19.5% DeMarco Murray (ADP: 7, Finish: 25) CE% year before: 26.2% *missed 6 games in 2012 Maurice Jones-Drew (ADP:8, Finish: 50) CE% year before: 24.2% *missed 10 games in 2012 Adrian Peterson (ADP: 11, Finish: 1) CE% year before: 29.3% Fred Jackson (ADP: 12, Finish: 39) CE% year before: 27.1% *missed 6 games in 2012 Ryan Mathews (ADP: 13 Finish: 31) CE% year before: 26.1% *missed 4 games in 2012 Doug Martin: no stats available/rookie Frank Gore (ADP: 20, Finish: 10) CE% year before: 21.6% Reggie Bush (ADP: 23, Finish: 14) CE% year before: 21.3% Stevan Ridley (ADP: 24, Finish: 9) CE% year before: 21.8% (small sample size) Cedric Benson (ADP: 25, Finish: >50) CE% year before: 21.6% *missed 11 games Peyton Hillis (ADP: 26, Finish >50) CE% year before: 22.4% *missed 3 games Shonn Greene (ADP: 28, Finish: 15) CE% year before: 20.6% CJ Spiller (ADP: 40, Finish: 7) CE% year before: 28.0% 2011 RBs: Chris Johnson (ADP: 4, Finish: 16) CE% year before: 20.9% Marshawn Lynch (ADP: 30, Finish: 4) CE% year before: 21.2% Ryan Mathews (ADP: 20, Finish: 8) 20.9% Michael Bush (ADP: 40, Finish: 9) 25.3% Darren Sproles (ADP: 44, Finish: 10) 20.0% Reggie Bush (ADP: 31, Finish: 13) 34.3% Fred Jackson (ADP: 29, Finish: 14) 21.2% Will add 2011 too. Surprisingly, there aren't a whole lot of guys that made a significant jump from ADP to actual results last year that weren't rookies. All I got are Spiller, Greene, Ridley, Bush, Gore, and AP. Edited July 10, 2013 by bostonsoxandy Quote Link to comment Share on other sites More sharing options...
keggerz Posted July 10, 2013 Share Posted July 10, 2013 (edited) Keggerz and Swiss -- I like the name Carry Efficiency (CE). That seems most accurate for this stat. However at this point I am slightly reluctant to to digg in and apply any type of factors for field position, down, or distance simply because of the time it would take. Right now it is relatively quick to calculate as both total carries and first downs are available as stats for each player every year. If I applied any type of factor like you are saying I would have to go game-by-game, probably play-by-play to determine the #'s. I agree that it would be more valuable that way but I think I might go insane in the process. And Swiss -- I will do that sometime this week (busts/booms from ADP and their CE the year before). That is definitely a good way to find any predictive power. i know all about going insane with the process...Swiss can attest to that too...for the record I wasn't saying you need to do that now...just something I "see" that I think could really make it an even more valuable metric is all. Edited July 10, 2013 by keggerz Quote Link to comment Share on other sites More sharing options...
Axe Elf Posted July 10, 2013 Share Posted July 10, 2013 The only stat that REALLY matters anyway is the AEAI (Baby Jane Approval Index). Outside of that, it's all a guessing game. Quote Link to comment Share on other sites More sharing options...
Old Money Posted July 11, 2013 Share Posted July 11, 2013 Does the number of first downs stat include receiving first downs by the running back or just the total first downs (both rushing and receiving)? Same thing with the touchdowns (total TD's or just rushing)? Quote Link to comment Share on other sites More sharing options...
bostonsoxandy Posted July 11, 2013 Author Share Posted July 11, 2013 Does the number of first downs stat include receiving first downs by the running back or just the total first downs (both rushing and receiving)? Same thing with the touchdowns (total TD's or just rushing)? All the stats provided are only for rushing. Absolutely no receiving totals were thrown into these #'s. That is probably why Ray Rice has lower numbers than one would expect. It really measures pure rushing efficiency in their team's system at this point. Quote Link to comment Share on other sites More sharing options...
Old Money Posted July 12, 2013 Share Posted July 12, 2013 All the stats provided are only for rushing. Absolutely no receiving totals were thrown into these #'s. That is probably why Ray Rice has lower numbers than one would expect. It really measures pure rushing efficiency in their team's system at this point. Just wanted to make sure your numbers weren't being skewed by receiving first downs being included. Very interesting stuff. I'm interested to see what trends you can determine with this. Good work. Quote Link to comment Share on other sites More sharing options...
WashingtonD Posted July 12, 2013 Share Posted July 12, 2013 Michael Turner - 22.5% Steven Jackson - 19.8% So the corpse of Michael turner got 22.5% while SJax for less than 20 behind a poor STL line. Sjax CE in 2014? Over under at 25% Quote Link to comment Share on other sites More sharing options...
stevegrab Posted July 12, 2013 Share Posted July 12, 2013 Ineteresting stuff. One player I see as an outlier on the list is Trent Richardson. He is near the bottom (17%) but there is no way he is losing touches. But overall there could be some merit to showing/predicting trends with some players. Quote Link to comment Share on other sites More sharing options...
bostonsoxandy Posted July 13, 2013 Author Share Posted July 13, 2013 Alright, so I had a bit of time tonight to try to test the correlation between 2011 CE and 2012 Results. The correlation between RB RANK (e.g. AP's rank last year is 1) and CE% is a moderate correlation with a correlation coefficient of 0.47. While this isn't a strong correlation, it definitely is a correlation and a potential predicting indicator for future success. I also came up with these following group numbers by 10's below for last year: Average RB Rank: 5.5 --- Average CE%: 24.8% Average RB Rank: 15.5 --- Average CE%: 23.3% Average RB Rank: 24.5 --- Average CE%: 27.5% Average RB Rank: 35 --- Average CE%: 26.2% Average RB Rank: 45 --- Average CE%: 18.0% Funky stuff. I think it gets higher at the end because of the individual players in that group. In the group 20-30, some guys I think that threw it off were: LeSean McCoy (just a horrid year for the eagles after a GREAT year with a 37.0% CE), Sproles (small amt of carries gave him a 33.3% CE for 2011), and DeMarco Murray (that Rams game really is a large part of his sample size rookie year CE of 26.2%) In the group 30-40, some guys I think that threw it off were: Ryan Mathews, Pierre Thomas (30% CE!), Ingram (specialty 3rd down/goalline guy has inflated CE of 28.7%), Jacquizz Rodgers (28.1% CE on small sample size rookie year), and Mike Tolbert (fullbacks or fullback-esque players have inflated CEs) A few interesting things to note from the data: In guys who finished in the top-10, the lowest CE% the year before was 21.8%. Everyone in the top-20 except Chris Johnson (b/c of very inefficient offense) finished with a CE of over 20.5%. The highest CE%s from 2011 were LeSean McCoy (37%), Darren Sproles (33.3%), Pierre Thomas (30%), Ahmad Bradshaw (29.8%), and Adrian Peterson (29.3%) Will do 2011 Rb Rank vs. 2010 CE% in my free time and hopefully draw a higher correlation. Also found a nifty database that had First downs listed for each player from last season and easily importable into excel. Here are the high-low guys in CE% for last year for everyone's info: HIGHEST: Tolbert (57.4%), Andre Brown (46.6%), Delone Carter (43.8%), Evan Royster (43.5%), Jacob Hester (41.2%), LaMichael James (37.0%), Jamie Harper (36.8%), Brandon Bolden (35.7%), Stevan Ridley (32.8%), Bryce Brown (32.2%), Danny Woodhead (31.6%), Ryan Grant (31.3%), Anthony Allen (31.3%), Ben Tate (30.8%), Cedric Peerman (30.6%), Shaun Draughn (30.5%), David Wilson (29.6%), Alfred Morris (29.6%) LOWEST (THAT ARE NOTABLE NAMES): Ryan Williams (13.8%), Cedric Benson (15.5%), Jonathan Stewart (16.1%), Darren McFadden (17.1%), Rashard Mendenhall (17.6%), Jacquizz Rodgers (18.1%), Trent Richardson (18.4%), Chris Johnson (18.8%), Ryan Mathews (19.6%), Matt Forte (20.2%), Steven Jackson (20.6%) Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.