3 year Draft Success Table

[Astrobot]

Joe Mays has updated the 3 year Draft Success table to include roster cutdowns that have taken place after training camp.

 

http://www.drafttek.com/ConsolidatedNFLDraft-stillwithteam.ASP

 

The Bills' draft success rate is 73.1%, the percentage of players drafted 2009-present that are still with the team. Atlanta is #1 with 95.2% (only the 2009 5th rounder, William Middleton CB from Furman is gone). Worst is Denver.

 

Other AFC East teams:

Miami---69.6%

NJets---92.3% (Scotty McKnight)

NEPats--66.7%

 

Astro

[The Big Cat]

Joe Mays has updated the 3 year Draft Success table to include roster cutdowns that have taken place after training camp.

 

http://www.drafttek.com/ConsolidatedNFLDraft-stillwithteam.ASP

 

The Bills' draft success rate is 73.1%, the percentage of players drafted 2009-present that are still with the team. Atlanta is #1 with 95.2% (only the 2009 5th rounder, William Middleton CB from Furman is gone). Worst is Denver.

 

Other AFC East teams:

Miami---69.6%

NJets---92.3% (Scotty McKnight)

NEPats--66.7%

 

Astro

 

Wang is listed as being on IR on bb.com...

[Albany,n.y.]

It's a lot worse. Only 16 of 26 are on the active roster. He included Jasper (practice squad) L.Brown -gone & Easley (IR-has yet to play a real down) That's 61.5% Even worse, there are only 4 starters of those 26, a measly 15.4%. That includes none from last year. Worse, if you don't include this draft, since most picks stick at least through their rookie year, the past 2 year only 8 of 17 picks are on the roster 47.1%. How many teams have already cut their 2009 top pick?

 

The bottom line is that the Bills have been among the worst drafting teams in the league. So far, Buddy Nix's 1st draft wasn't much better, yielding no starters. Until the Bills start drafting better, they'll be no better than mediocre.

 

Wang is listed as being on IR on bb.com...

Wang is waived injured. As soon as he's healthy enough to play, he's off the team.

 

http://wiki.answers.com/Q/What_does_it_mean_when_a_NFL_player_is_waived_injured

Edited September 15, 2011 by Albany,n.y.

[The Big Cat]

It's a lot worse. Only 16 of 26 are on the active roster. He included Jasper (practice squad) L.Brown -gone & Easley (IR-has yet to play a real down) That's 61.5% Even worse, there are only 4 starters of those 26, a measly 15.4%. That includes none from last year. Worse, if you don't include this draft, since most picks stick at least through their rookie year, the past 2 year only 8 of 17 picks are on the roster 47.1%. How many teams have already cut their 2009 top pick?

 

The bottom line is that the Bills have been among the worst drafting teams in the league. So far, Buddy Nix's 1st draft wasn't much better, yielding no starters. Until the Bills start drafting better, they'll be no better than mediocre.

 

 

Wang is waived injured. As soon as he's healthy enough to play, he's off the team.

 

http://wiki.answers.com/Q/What_does_it_mean_when_a_NFL_player_is_waived_injured

 

Thanks for the clarification.

[Astrobot]

Wang is listed as being on IR on bb.com...

Uh, no. Waived/Injured is considered gone.

Look at Transactions page

09/03/2011 waived/injured T Ed Wang

[Orton's Arm]

Joe Mays has updated the 3 year Draft Success table to include roster cutdowns that have taken place after training camp.

 

http://www.drafttek.com/ConsolidatedNFLDraft-stillwithteam.ASP

 

The Bills' draft success rate is 73.1%, the percentage of players drafted 2009-present that are still with the team. Atlanta is #1 with 95.2% (only the 2009 5th rounder, William Middleton CB from Furman is gone). Worst is Denver.

 

Other AFC East teams:

Miami---69.6%

NJets---92.3% (Scotty McKnight)

NEPats--66.7%

 

Astro

I realize there's no such thing as a perfect statistical measure, and that most measures are variants of the "good but imperfect" theme. Nevertheless, I feel the urge to improve this particular statistical measure.

 

Suppose Team A had one pick per round in each round of the draft. Whereas Team B has one pick per round in rounds 1 - 3, and two or even three picks per round in rounds 4 - 7.

 

As you know, first and second round picks are much more likely to meaningfully contribute than are sixth or seventh rounders. Assuming Team A and Team B are equally good at drafting, Team B's success rate will look a lot worse because of all those late round picks it had.

 

The solution is something I'll call "adjusted success rate." Here is how it should be calculated:

 

1. Compute the average success rate by round. For example, let's say you wanted to find the adjusted success rate for the 2007 draft. You would begin by determining the percentage of 2007 first round picks which are still with their original teams, the percentage of second round picks, and so forth.

 

2. Compare each team's success rate by round against the average rate for the round. For example, if the average success rate for the first round is 75%, and if 100% of team X's first round players from 2007 are successful, that team's first round success rate is +25%.

 

3. Multiply the above numbers by the number of players in each round. If Team X had one sixth round pick (bust) and four seventh round picks (all successes), its -20% success rate for the sixth round should be multiplied by 1, and its +85% success rate for the seventh round should be multiplied by 4.

 

4. Add each team's numbers from step 3, and then divide by the number of players that team drafted. This is its adjusted success rate. For example, Team X's adjusted success rate would be:

 

[ +25% (first round pick) -20% (sixth round pick) + ( + 85% * 4) (seventh round picks) ] / 6 picks

 

The above number would indicate whether Team X's picks were more or less successful than expected, given the rounds during which the picks were made.

[San Jose Bills Fan]

EA, on top of your proposed measurement system, you'd then ideally want to factor in the fact that it's easier for rookies to make the roster of a weak team as opposed to a strong team.

 

It seems like Joe Mays' formula is way too simple to have much meaning.

[ganesh]

I realize there's no such thing as a perfect statistical measure, and that most measures are variants of the "good but imperfect" theme. Nevertheless, I feel the urge to improve this particular statistical measure.

 

Suppose Team A had one pick per round in each round of the draft. Whereas Team B has one pick per round in rounds 1 - 3, and two or even three picks per round in rounds 4 - 7.

 

As you know, first and second round picks are much more likely to meaningfully contribute than are sixth or seventh rounders. Assuming Team A and Team B are equally good at drafting, Team B's success rate will look a lot worse because of all those late round picks it had.

 

The solution is something I'll call "adjusted success rate." Here is how it should be calculated:

 

1. Compute the average success rate by round. For example, let's say you wanted to find the adjusted success rate for the 2007 draft. You would begin by determining the percentage of 2007 first round picks which are still with their original teams, the percentage of second round picks, and so forth.

 

2. Compare each team's success rate by round against the average rate for the round. For example, if the average success rate for the first round is 75%, and if 100% of team X's first round players from 2007 are successful, that team's first round success rate is +25%.

 

3. Multiply the above numbers by the number of players in each round. If Team X had one sixth round pick (bust) and four seventh round picks (all successes), its -20% success rate for the sixth round should be multiplied by 1, and its +85% success rate for the seventh round should be multiplied by 4.

 

4. Add each team's numbers from step 3, and then divide by the number of players that team drafted. This is its adjusted success rate. For example, Team X's adjusted success rate would be:

 

[ +25% (first round pick) -20% (sixth round pick) + ( + 85% * 4) (seventh round picks) ] / 6 picks

 

The above number would indicate whether Team X's picks were more or less successful than expected, given the rounds during which the picks were made.

 

Everything is not just about some statistical analysis. Though numbers don't lie, they also can be interpreted at each one's convenience.

 

This is simply not sufficient. The success of a draft is felt by the impact made by the player on the team. For example, a Donte Whitner played for the Bills for five full seasons. Does it mean draft success...as he had no visible impact on the success of the team.

[Orton's Arm]

Everything is not just about some statistical analysis. Though numbers don't lie, they also can be interpreted at each one's convenience.

 

This is simply not sufficient. The success of a draft is felt by the impact made by the player on the team. For example, a Donte Whitner played for the Bills for five full seasons. Does it mean draft success...as he had no visible impact on the success of the team.

Good point.

 

I recall that one football site has created player performance measurements, with a positive performance number indicating that the player is reasonably competent. Negative numbers indicate players who are below-average and should be replaced.

 

One way of evaluating drafts--and a better way than the one I suggested, by the way--would be to take the player performance numbers from each of a given team's draftees, ignore the negative numbers, and add up the positive numbers. That way a team would get only a small amount of credit (or hopefully no credit) for drafting a Whitner, while getting a lot of credit for taking a Ngata. The worst possible score on this system would be a zero, so any player who gives you no credit would bring you that much closer to a zero score.

 

If you wanted to take things one step further than this, you could use a draft day value chart to determine the total value of each team's draft picks. Then you'd divide the total player performance number described above by the value of their draft picks. Then you'd multiply by, say, 1000, to give you a nice, reasonable number like three or four, instead of something tiny like 0.004.

 

With a system like this, a team that uses an 8th overall pick to get a mediocre player would receive a much lower score than a team which had a pick in the second half of the first round, and which came away with an equally mediocre player.

[maddenboy]

note to ASTROBOT:

 

Regarding your 2012 draft chart, you might wanna factor in Buddy's statement that you have to draft a corner every year. At least one. (See 2nd paragraph).

 

http://www.buffalobills.com/news/article-2/Rookie-CBs-provide-much-needed-depth/a6c06d89-504c-422f-a381-4087e8c47dfc

 

When I first read that, I thought it was stupid. But on further reflection, considering how many you need, how they play special teams, and that speed doesnt age well, I guess I can see it.

Edited September 17, 2011 by maddenboy

[Bill from NYC]

Good point.

 

I recall that one football site has created player performance measurements, with a positive performance number indicating that the player is reasonably competent. Negative numbers indicate players who are below-average and should be replaced.

 

One way of evaluating drafts--and a better way than the one I suggested, by the way--would be to take the player performance numbers from each of a given team's draftees, ignore the negative numbers, and add up the positive numbers. That way a team would get only a small amount of credit (or hopefully no credit) for drafting a Whitner, while getting a lot of credit for taking a Ngata. The worst possible score on this system would be a zero, so any player who gives you no credit would bring you that much closer to a zero score.

 

If you wanted to take things one step further than this, you could use a draft day value chart to determine the total value of each team's draft picks. Then you'd divide the total player performance number described above by the value of their draft picks. Then you'd multiply by, say, 1000, to give you a nice, reasonable number like three or four, instead of something tiny like 0.004.

 

With a system like this, a team that uses an 8th overall pick to get a mediocre player would receive a much lower score than a team which had a pick in the second half of the first round, and which came away with an equally mediocre player.

 

Sorry my friend. The above doesn't mean much imo. One must take into account many other factors, such as the overall quality of a draft, and the availability of players of need to a particular team. This, and even trade down offers come into play.

The Bills were said to have spurned an extra second and fourth round pick, and I believe this. Levy admitted on SIRIUS that he was offered more than a second rounder to trade down.

Here is my mathmatatical formula for the Whitner pick: It sucked, and helped to ruin this team for years.

Hopefully we are climbing out of the hole.

[Orton's Arm]

Sorry my friend. The above doesn't mean much imo. One must take into account many other factors, such as the overall quality of a draft, and the availability of players of need to a particular team. This, and even trade down offers come into play.

The Bills were said to have spurned an extra second and fourth round pick, and I believe this. Levy admitted on SIRIUS that he was offered more than a second rounder to trade down.

Here is my mathmatatical formula for the Whitner pick: It sucked, and helped to ruin this team for years.

Hopefully we are climbing out of the hole.

I fully agree with your assessment of the Whitner pick. My goal was to build a mathematical model which would illustrate the damage done by such picks. I agree that there are factors outside the model which, when taken into account, make the Whitner pick even worse than the model would indicate.

 

One could draw a comparison between the Whitner pick of 2006 and the Mike Williams pick of 2002. The 2006 draft was a better draft than the 2002 draft. That could mean that the 4th overall pick in 2002 was roughly equivalent to the 8th overall pick of 2006.

 

Mike Williams spent four years with the Bills before his release. Whitner spent five years with the Bills before the team decided to go in a different direction at safety. (Note that Nix didn't engage in serious contract talks with Whitner after the 2011 draft, even though Whitner indicated his openness to returning to Buffalo.)

 

Mike Willams was a starter for the Bills for his first three seasons, and lost his starting position in his fourth season with the team. Williams was hampered by injuries during the 2005 season, and Jason Peters played very well.

 

Donte Whitner was a starter for the Bills for his first three seasons, and lost his starting position in his fourth season with the team. George Wilson was a lot less good at SS than Jason Peters was at RT, so Whitner had less competition than Williams. Unlike Williams, Whitner returned for a fifth year with the Bills, and was able to regain his starting spot. Part of the reason for that is that while Wilson is better than Whitner in pass coverage, Whitner is better at stopping the run. The Bills had the worst run defense in the NFL during the 2010 season, so it was felt that more emphasis had to be placed on run-stopping than pass coverage when selecting the starting SS.

 

The picks used on Williams and Whitner were of roughly equal value. Both picks resulted in players who started for a few years. When I think of either player, I think, He's a guy who can get you by over the short-term, but you really want to find an upgrade for him sooner or later.

 

One difference between the two players is that Mike Williams was a joke at OG, and didn't do anything after he left Buffalo. Whitner seems likely to continue to be a starter in the NFL, albeit not a very good one. Whitner is likely to have a better career than Mike Williams, even though the two players' contributions to the Bills are very comparable.

Edited September 17, 2011 by Edwards' Arm

[ganesh]

 

One difference between the two players is that Mike Williams was a joke at OG, and didn't do anything after he left Buffalo. Whitner seems likely to continue to be a starter in the NFL, albeit not a very good one. Whitner is likely to have a better career than Mike Williams, even though the two players' contributions to the Bills are very comparable.

 

Another point that is forgotten is the coaching turnstile. I believe Mike Williams had 3 OL coaches in his 4 seasons...whereas Whitner has mainly seen the same DC for his 1st 4 seasons before Edwards replaced Perry. At that time, I believe that was another reason attributed to Williams failure to succeed in the NFL.

 

McGahee, Whitner, McCargo, Losman, Lynch, McKelvin and the list goes on...