Greg Gabriel at the National Football Post (a great website that I highly recommend) recently published an article entitled “How clubs strategize for free agency and the draft” which postulates an interesting idea that when picking at a certain position in the draft, a team can expect a certain number of players at a position to be picked ahead of them (if that sounds confusing please check out his article where he goes more indepth in the concept). For instance, if a team just won the Super Bowl and is picking last in the 1st round and is interested in drafting a quarterback, they can reasonably expect to see the 3rd or 4th best quarterback still available because on average 2-3 quarterbacks get drafted in the 1st round. If you think about it, there are only so many “can’t miss” 1st round draft picks produced every year and usually their positions are distributed rather evenly (factored in with the nature of the NFL). Obviously some years can be strong years for certain positions, like this year with defensive ends, but on average the amount of players in a certain position selected remains relatively constant.
With that in mind, the question I had is which players are statistically likely to be available at pick 28 for the Packers and which picks would make sense compared to previous years. To do that I complied a list of every draft from the 1st pick overall to the 28th pick overall from 2005 (the year Ted Thompson became the general manager) to 2011 and then computed the average number of players at a position taken and their standard deviation between each year.
|
POS |
2005 |
2006 |
2007 |
2008 |
2009 |
2010 |
2011 |
Total |
Average |
STD |
min |
max |
|
QB |
3 |
3 |
2 |
2 |
3 |
2 |
4 |
19 |
2.71 |
0.76 |
2 |
3 |
|
RB |
3 |
3 |
2 |
5 |
2 |
2 |
1 |
18 |
2.57 |
1.27 |
1 |
4 |
|
FB |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
0.00 |
0.00 |
0 |
0 |
|
WR |
6 |
1 |
4 |
0 |
4 |
2 |
3 |
20 |
2.86 |
2.04 |
1 |
5 |
|
TE |
0 |
2 |
0 |
0 |
1 |
1 |
0 |
4 |
0.57 |
0.79 |
0 |
1 |
|
C |
1 |
0 |
0 |
0 |
2 |
1 |
0 |
4 |
0.57 |
0.79 |
0 |
1 |
|
G |
0 |
1 |
0 |
1 |
0 |
1 |
1 |
4 |
0.57 |
0.53 |
0 |
1 |
|
T |
2 |
1 |
3 |
7 |
4 |
4 |
4 |
25 |
3.57 |
1.90 |
2 |
5 |
|
DT |
1 |
3 |
2 |
2 |
2 |
4 |
3 |
17 |
2.43 |
0.98 |
1 |
3 |
|
DE |
3 |
4 |
5 |
4 |
3 |
4 |
5 |
28 |
4.00 |
0.82 |
3 |
5 |
|
LB |
5 |
4 |
3 |
2 |
5 |
2 |
1 |
22 |
3.14 |
1.57 |
2 |
5 |
|
DB |
4 |
6 |
7 |
5 |
2 |
5 |
3 |
32 |
4.57 |
1.72 |
3 |
6 |
|
K |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
0.00 |
0.00 |
0 |
0 |
|
P |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
0.00 |
0.00 |
0 |
0 |
Essentially what this shows is at pick 28 how many players of a certain position are likely to have been already taken, and what is considered a “reach”. For instance, if the Packers were interested in a defensive tackle, they could reasonably expect to see the best tackle already taken by pick 28. On the flip side, on average 3 tackles are rarely ever taken by pick 28, so picking the 4th best tackle would be considered a “reach”. I’ve used the current big board found at CBSSports NFL draft page to look up what they think are the best defensive tackles this year and from that the Packers could reasonably expect to see #1 DT Michael Brockers gone by pick 28, but could be reasonably expected select from either Dontari Poe or Fletcher Cox. Devon Still, the next best defensive tackle would be considered a “reach” since on average the 4th best defensive tackle is rarely taken by pick 28.
A couple of caveats, naturally the biggest flaw in this analysis is which big board to use, I used CBSsports because they list positional rankings but there’s plenty of disagreement around the internet of who should be where. Another issue is that it doesn’t take into account some of the faster moving trends going on in the draft, for instance the running back position has taken a huge dive as the running game has continued to get marginalized, and seeing 5 running backs taken in the first 28 picks like what happen in 2008 will probably never happen again. Also there is some ambiguity in putting certain players in certain positions; for instance Aldon Smith, the 49ers 1st round pick last year is listed as a defensive end, but obviously played outside linebacker in the NFL and defensive tackle prospects like Dontari Poe or Fletcher Cox could easily be playing 3-4 defensive end in the NFL careers. Below is the complete list of players in each position that statistically make sense at pick 28.
|
Quarterback |
Ryan Tanehill |
Offensive Tackle |
Riley Reiff Jonathan Martin Mike Adams |
|
Running Back |
David Wilson Lamar Miller Doug Martin |
Defensive Tackle |
Dontari Poe Fletcher Cox |
|
Wide Receiver |
Michael Floyd Kendall Wright Rueben Randle Stephen Hill |
Defensive End |
Andre Branch Jared Crick Vinny Curry
|
|
Tight End |
Dwayne Allen |
Linebacker |
Courtney Upshaw Zach Brown Dont’a Hightower
|
|
Center |
Peter Konz |
Defensive Back |
Mark Barron Stephon Gilmore Alfonzo Dennard
|
|
Guard |
David DeCastro |
Nevertheless, it’s an interesting concept that supposedly does factor into how teams plan for the draft. Will any of these players be “the guy” for the Packers at pick 28? I guess we’ll see at the end of April.
——————
Thomas Hobbes is a staff writer for Jersey Al’s AllGreenBayPackers.com.
——————
Interesting topic, Thomas. Keep ’em coming!
I imagine that if Ted Thompson doesn’t trade up/down, we might be drafting Peter Konz or Vinny Curry.
Thanks, I should add I really don’t know if this really works, but I thought it would be interesting to apply it to the Packers situation.
I can’t see Hightower getting past either the Steelers or the Ravens. What makes you think otherwise?
I think this chart just takes into account past draft trends not really people’s opinion of which teams would break that trend if I understand it correctly. So basically it ignore teams needs and the players abilities.
Could be misleading saying players abilities. I meant only take into account players ranking within their position. Do not take into account how he compares to other player’s ability at different position. Use the past trends to determine that. Correct me if I’m wrong Thomas.
On the flip side, past draft trends do take into account teams that act weirdly, since the same teams seem to do it over and over again. For instance, everyone knew that the Oakland Raiders were going to pick someone completely random and the Jacksonville Jaguars were also a team that often picked a relatively unknown player (though usually it seemed to turn out well)
yeah but if a team acts unusually one time it doesn’t have much of an effect unless they consistantly made that unusual decision for the same position over and over again. Basically this chart is useful to show the league wide tendency to pick a particular position and guess based on that which players, via ranking within the same position, will be available. two ways this chart is flawed which you addressed one of them, are that players are ranked differently by different organizations and, one that I haven’t heard mentioned or if I did I don’t remember, it doesn’t take into account talent pooled into one postion on a particular draft year. For example this year there are tons of DT/ DE types that everyone is really excited about so it might be staggered in that direction this year. It’ll be a fun experiment to see how true the chart holds at the end of the 1st round.
Sorry I had a major brain fart. Dont’a Hightower != Vonta Burfict, though they have similar sounding first names.
Cool to see this thanks for the effort you put into designing the chart. I hope it projects accurately… It would be crazy if we could snag upshaw. Obviously probably doesn’t project that choice accurately. I don’t think we go for curry in the 1st. Curry is probably going mid second his 4.98 40 will scare teams off till then… Unless he has an incredible pro day. I could see us trading up for him but not picking him 28. I acknowledge that I defendEd his 40 time in the past but that was for those saying he wasn’t a valid pick at all anymore. It still affects my opinion of him though and should drop him to mid to late 2nd round
I feel Currys Pro Day will be so much better than the combine,taking him at 28 if DL isn’t available becomes a much smaller reach and happens.
Thomas, thanks for doing this. Why did you include pick 28, wouldn’t you want to see who’s picked 1-27 to let you know who is still left AT 28? Probably wouldn’t make a difference, just wondering.
I hope TT has a chance to pick from Cox, Poe, DeCastro, Barron, and the OT’s. And if Tannehill is there I could see a team wanting to trade back into round 1 to get him.
That’s actually how it’s setup, for instance if you look at my chart, #1 DT Michael Brockers is not on my list, because statistically he should be gone by pick 28. However Poe and Cox statistically should still be available and past trends would indicate that the #2 and #3 DTs have been picked in previous drafts so it would be considered a “reach”
Good stuff Thomas. It is always hard to look at things like this in a vaccuum, but it does make a lot of sense.
Upshaw does have a chance to be there at 28 I think. He could be a guy that the media is overhyping because of where he played.
I
Well if you have the Packers big board somewhere, I’d love to see it 😀 Like I said, I think the biggest problems with mock drafts etc is that people’s perceptions of rankings is often never even remotely close to what teams have them ranked as. I’d say the media and teams are pretty close with the top 5 players overall, but after that its a complete crap shoot.
From what I’ve seen from Upshaw (2012 BCS, 1st Alabama x LSU, 2010 BCS), much more of a read and react player than a pass rusher. Didn’t see the explosiveness, didn’t see the intensity. Heck of a run player, good instincts and sound all around. But NFL teams want a difference maker, particularly in the top 20. I’m with you, I think he falls.
If I had the choice, I would go Melvin Ingram every time.
As for the chart, that’s a terrific work, but I doubt DeCastro falls that far. A lot of analysts out there, like at NFP, saying he’ll be the 1st or 2nd OL taken, and that he’s the best OL prospect in the draft (albeit the premium on OTs)
That may not be that unlikely. In the last 10 years, there has not been more than one “true” guard drafted in the 1st round (tackles that end up as guards or are drafted to be a guard is a different issue). Furthermore the highest draft pick used on a guard was 15, and there were 3 years where no guard was picked at all.
usually their positions are “disturbed” rather evenly – ?
Sorry, I totally blame automatic spellchecker
The craftsman blames his tools…
Haha! 😀
http://www.damnyouautocorrect.com/
LOL! My iPhone does some wacky stuff, too. 🙂
Seeing as we will most likely get a 3rd and 4th round compensatory pick, I hope TT considers moving our normal 3rd/4th/5th to a team at the top of the 3rd that may want/need more picks. We could still get some quality in the early 3rd and TT would have numerous other picks to hedge the risk from trading picks away. GoPack!
That was exactly my thoughts in my last article, so I whole heartedly agree
TT probably has Reggie McKenzie’s oakland area phone number on speed dial…
Actually, I wouldn’t be all too surprised to see a draft pick trade with the Raiders, since Thompson was basically in McKenzie’s shoes in 2005 and McKenzie probably will run the show roughly the same way until he gets more draft picks and settles the salary cap in Oakland
Exactly my thoughts.. Reggie will rebuild the Raiders from top to bottom, and he’ll probably do it via the draft for the most part. Those mid round picks are gold.
Bruce Irvin should be our OLB pick in round 2
Maybe I will go further with this analysis and look into round 2, though I’m a little concerned that the standard deviation will so big that you can’t make much of a claim.
Want to see another effect of the offense-friendly NFL rules changes?
According to this chart a disportionate number of defensive players have been taken with the early picks since 2005, and even these presumably better players haven’t been able to slow offenses down.
Well if you don’t have a star quarterback, the only other option you have is to have a stellar defense that can mitigate the advantage that a star quarterback brings to the field. This makes sense with what you’re saying though, take for instance the defending Super Bowl Champion New York Giants, they only spent 1 draft pick on quarterback Eli Manning, but god knows how many picks they’ve spent on their defensive line.
Love the way you do these statistical analyses, Thomas. Don’t know if it means anything or not, but it certainly does satisfy the geek in me.
Thanks. To be honest, I don’t know if it means anything either, but it’s an interesting concept
If Upshaw or Perry were available at 28… that would be epic.
I doubt it happens, but it’d be nice.
Good overview of the potential picks at 28. I not sure which way TT will go here. That’s not unusual. My preference for #1 is an impact OLB first, followed by a DE then a C.
The D MUST be fixed and if FA is not going to be choice, then they have to u[grade the D in the Draft. A top 2-3 defense falling to dead last is UNACCEPTABLE.
There are some options at DE in FA, but not for OLB. Red Bryant & Jason Jones are interesting.Upshaw worries me because he’s too thick and slow for pass rushing OLB but too short for a DE.Perry is a little more explosive.
Don’t see either Poe or Cox available at 28. If were stuck with Crick, I’d be sick.
Well if Poe and Cox are gone, that does improve the chances of a impact OLB still be available. So really, you should be cheering every time a quarterback, offensive tackle, wide receiver or tight end is selected before the Packers.
That’s why I said OLB Prioroty #1. With Carricker available in FA, he could be TT’s next Ryan Pickett. (note: got the idea oave at Packer.com)
With Wells expected to go to FA, Priority #1 is now a Center.
I wouldn’t be so sure of that either. For one, just because Wells gets to see free agency doesn’t necessarily mean he won’t come back. Second, a starting center is probably easier to get than a starting 3-4 outside linebacker, so they might try to find a center in FA and still draft someone on defense.
No matter how things turn out with Wells, I would expect to see his (immediate or eventual) replacement taken in the first 3 rounds of the draft.
For those that don’t know statistics, his min and max can be expected to happen with an accuracy of 68%. If we add the STD to the Average twice, and subtract the STD from the Average twice, a new max and min are produced. Those new max and min can be expected to happen at an accuracy of 96%.
sorry…closer to 95%, not 96%
Frankly, I think not getting to p<.05 is probably not all that relevant because the bigger issue is that no one has the same ranking of players and even more important than that is that the media is terrible at ranking players. For instance if I doubled the STD and added it to the average I could say with 95% certainty that according to Rob Rang these players will be available
The names of players may be disputed, but the number of players taken will still be accurate.
There is no chance TT would draft Crick in the first round. Crick will be lucky to be our 3rd round pick. How anyone projects that average stiff to the first round is beyond me. GoPack!
This was the biggest issue that I brought up, you think Crick is not the 5th best DE prospect, but CBSsports apparently does. I don’t know which one of you is right (we’ll find out in 3 years or so), but which big board you go off of is likely to drastically change which players you see available
If Scott Wells leaves they should draft mike brewster!
Considering how the Packers have drafted for center during Thompson’s reign, it seems to be a rather low priority. Apparently Thompson has never drafted a “true” center while operating as the Packers GM, which is a little surprising to say the least.