Draft Analysis #3
some final summary stats and conclusion to the analysis of the initial 8 teams
Average Rating of Drafted Players by Team
Overall
Bubble Graph - size of bubble reflects count
Conclusions
based on the initial set of draft ratings and summary data, the eagles do not jump out as the best of the initial set of 8 teams... but they're not the worst either. they're better at some things than others, but overall, my conclusion is that they're somewhere in the middle of the top 8 teams and that tentatively puts them somewhere among the top 5 drafting teams (assuming that our instincts were correct in the initial identification of best drafting teams). while they are certainly not perfect nor consistent, as a whole, the andy reid draft era has probably been the best i've seen for any eagles era and is competitive with the best of his peers.
some observations:
- one of the areas the eagles didn't do as well as one would hope -- and probably why they are perceived to be worse at drafting than the overall picture indicates -- is drafting in the first and second rounds. in the first and second rounds, the birds average player rated at 3.0 and 2.2, tying for 5th and sitting alone in 6th among the 8 teams. in hit rate, the birds only hit on 75% of their first rounders and 40% of their second rounders, again ranking 5th and 6th among the 8 teams. they do make up for it somewhat by ranking 3, 3, 2, 5, 1 for rounds three through seven, but you'd like to see them do better in the early rounds because (as the bubble graph shows) that's where you get the bulk of your top players. i'd like to see where 75% and 40% places them among the broader pool of nfl teams as we add more to the analysis.
- the best early drafting team is clearly the colts, who hit on 100% of their first rounders and 50% of their second rounders, ranking first and second in those categories. their average player rating in the first round is 3.43, easily the best among the 8 teams.
- in terms of average player rating by year, the birds ranked 2, 6, 2, 1, 7, 5, 5, 3 for the years 1999 through 2006, which averages out to just under 4, placing them at the top of the middle. the best team here was the ravens (whose drafts ranked 1, 1, 1, 5, 4, 8, 3, 4), and the worst team was the cowboys (whose drafts ranked 6, 8, 8, 8, 1, 7, 1, 7).
- the eagles ranked 4th in overall hit percentage behind the ravens, colts, and steelers. the ravens hit on an amazing 30% of their drafted players.
- the eagles ranked 2nd in ability to find pro-bowl caliber players (1 in 11), behind only the ravens (1 in 10)
- most of the teams were between 15-18 hits overall, indicating that teams should expect to get about 2 good starters per draft. well, except for the cowboys, who only hit on 10 players overall and were consistently toward the bottom in every measure except for success in the first round. of course, the cowboys also have generally drafted much higher than the other teams on this list during that time, so some of that first round success is expected.
- bumble holds the fact that the birds had just one "great" draft against them, but i don't see why that should be held against them. i think it makes sense to hold that against them if you believe that one draft was luck, but the data shows that's simply not the case. even if you eliminated two of the all-pro caliber players from that draft (making it an average draft in terms of players produced), the eagles would still be competitive with the other teams reviewed. their ranking would go down, certainly, but they'd still be better than the cowboys and the broncos by most measures.
the devil's in the details, so obviously all of these are based on the notion that the player ratings were done reasonably accurately and without bias. i'm fairly confident i've done that, and since there has been a reasonable amount of traffic visiting the site and just a few corrections/ongoing discussions, i'll assume and we can use this as a basis moving forward.
next steps will be to rate the drafts for the second group of teams -- cardinals, lions, bills, rams, saints, bengals, 49ers, texans. (on phil's suggestion, i'm including the rams, but will leave the bengals in and take out the raiders)
Average Rating of Drafted Players by Team
| Broncos | Eagles | Steelers | Ravens | Cowboys | Chargers | Patriots | Colts |
|---|---|---|---|---|---|---|---|
| 1.69 | 1.82 | 1.75 | 1.85 | 1.56 | 1.70 | 1.69 | 1.75 |
Overall
| Average of All Drafted Players | Correlation (Round-Rating) |
|---|---|
| 1.73 | -0.56 |
Bubble Graph - size of bubble reflects count
Conclusions
based on the initial set of draft ratings and summary data, the eagles do not jump out as the best of the initial set of 8 teams... but they're not the worst either. they're better at some things than others, but overall, my conclusion is that they're somewhere in the middle of the top 8 teams and that tentatively puts them somewhere among the top 5 drafting teams (assuming that our instincts were correct in the initial identification of best drafting teams). while they are certainly not perfect nor consistent, as a whole, the andy reid draft era has probably been the best i've seen for any eagles era and is competitive with the best of his peers.
some observations:
- one of the areas the eagles didn't do as well as one would hope -- and probably why they are perceived to be worse at drafting than the overall picture indicates -- is drafting in the first and second rounds. in the first and second rounds, the birds average player rated at 3.0 and 2.2, tying for 5th and sitting alone in 6th among the 8 teams. in hit rate, the birds only hit on 75% of their first rounders and 40% of their second rounders, again ranking 5th and 6th among the 8 teams. they do make up for it somewhat by ranking 3, 3, 2, 5, 1 for rounds three through seven, but you'd like to see them do better in the early rounds because (as the bubble graph shows) that's where you get the bulk of your top players. i'd like to see where 75% and 40% places them among the broader pool of nfl teams as we add more to the analysis.
- the best early drafting team is clearly the colts, who hit on 100% of their first rounders and 50% of their second rounders, ranking first and second in those categories. their average player rating in the first round is 3.43, easily the best among the 8 teams.
- in terms of average player rating by year, the birds ranked 2, 6, 2, 1, 7, 5, 5, 3 for the years 1999 through 2006, which averages out to just under 4, placing them at the top of the middle. the best team here was the ravens (whose drafts ranked 1, 1, 1, 5, 4, 8, 3, 4), and the worst team was the cowboys (whose drafts ranked 6, 8, 8, 8, 1, 7, 1, 7).
- the eagles ranked 4th in overall hit percentage behind the ravens, colts, and steelers. the ravens hit on an amazing 30% of their drafted players.
- the eagles ranked 2nd in ability to find pro-bowl caliber players (1 in 11), behind only the ravens (1 in 10)
- most of the teams were between 15-18 hits overall, indicating that teams should expect to get about 2 good starters per draft. well, except for the cowboys, who only hit on 10 players overall and were consistently toward the bottom in every measure except for success in the first round. of course, the cowboys also have generally drafted much higher than the other teams on this list during that time, so some of that first round success is expected.
- bumble holds the fact that the birds had just one "great" draft against them, but i don't see why that should be held against them. i think it makes sense to hold that against them if you believe that one draft was luck, but the data shows that's simply not the case. even if you eliminated two of the all-pro caliber players from that draft (making it an average draft in terms of players produced), the eagles would still be competitive with the other teams reviewed. their ranking would go down, certainly, but they'd still be better than the cowboys and the broncos by most measures.
the devil's in the details, so obviously all of these are based on the notion that the player ratings were done reasonably accurately and without bias. i'm fairly confident i've done that, and since there has been a reasonable amount of traffic visiting the site and just a few corrections/ongoing discussions, i'll assume and we can use this as a basis moving forward.
next steps will be to rate the drafts for the second group of teams -- cardinals, lions, bills, rams, saints, bengals, 49ers, texans. (on phil's suggestion, i'm including the rams, but will leave the bengals in and take out the raiders)
posted by The Mean Guy at 1:57 PM
6 Comments:
Very well done.
I think we'll get a much better picture once the bottom teams are looked at.
Thank goodness for that 2002 draft though. I don't care whether it was luck or skill, but we'd be hosed without those players.
TMG this is really interesting stuff. I am sure NFL teams know this, but I have never seen such comparison done and I wonder if you aren't missing your calling as a pro football statistician.
Bumble
Off topic - 2 excellent articles on Ray Didingers website. 1 on Jaws and 1 on McNabb.
Excellent analysis Pete. Maybe time for a career change ?
Captain
Nothing about the Phillies? Or are y'all to afraid that any comments will jinx them?
i'm enjoying this team and leaving the phillies talk to the many phillies blogs out there doing a great job discussing the team's exciting run to the end.
let's go phils!!!I was waiting until they actually made it to comment on it. Great comeback, gotta love that team for its fighting spirit.
Simon
Post a Comment
<< Home