Jump to content

An analytical (numbers/metrics-based) take on this year's EDGE class


Kayoh
 Share

Recommended Posts

g-dawg, you might want to click out of this thread straight away. This isn't your flavor of draft talk so you aren't going to contribute anything to the actual discussion.

That said, regarding this year's edge class, it's bad. Like, *really* bad. There just flat out isn't a lot of value to be had. Last year we had some truly freakish athletes with Vic Beasley, Danielle Hunter, Frank Clark, Preston Smith and Randy Gregory all being top tier athletes and having legit college production to back their athletic profiles up. The year before that we had Khalil Mack, Jadeveon Clowney, Anthony Barr, and even guys like Kareem Martin, Kony Ealy and Trent Murphy who all scored well as athletes and at least a majority of them with good production to back that up.

We don't really have that this year. We have one guy. A single individual who ranks as a top tier athlete AND put numbers up in college to back that athleticism up. His name is Joey Bosa, and he's not getting out of the top 10. Most likely he won't even get out of the top 5.

There are also very few top tier athletes in this class regardless of production. Outside of Bosa, the only other truly top tier athletes in this class are Virginia DE Trent Corney, and Georgia LB Leonard Floyd. Neither of those guys had production in college to back up their athletic profiles.

As far as what I'm talking about when I say "top tier" athletes, I know a bunch of you guys are at least familiar with, if not sick of hearing about, my edge formula. I've refined it to be much more accurate based on a combination of factors. It's actually gotten quite impressive, but now I'm just kinda jerkin myself off without really explaining why it works. For those of you who have no idea what I'm talking about, I've explained it a few times. Here is the first post I mentioned it in, here is another post about it, and another, and the most recent one which also looked at this year's 1st round edge prospects.

Now, as far as the refinements I've done, it's a combination of two things. First of all, I've separated the players into two different spreadsheets. The spreadsheets are basically the "pass" and "fail" spreadsheets regarding my Agility metric. On average, a player who passes my Agility filter will be approximately 90% more valuable throughout their career than a player who fails the Agility filter. I've actually done the math and the average AV per year of players who pass the filter is 3.99, but players who fail the filter only have an average AV per year of  2.10. It's pretty significant, especially since I know another method is to use a 7.25 3 cone time as the pass/fail agility filter, and I've found my filter to be more accurate. On average a player who passes my filter but fails the 7.25 method will have an AV per year of 3.06, while a player who fails my filter and passes the 7.25 filter will have an AV per year of 2.08. So I'm clearly onto something here, but I digress.

The next refinement I've made to the formula is I've separated it into rounds. So I can tell you that 1st round players who pass my Agility filter average a career AV per year of 6.27, while 1st round players who *fail* the Agility filter average a career AV per year of 3.41. That's an 84% jump in value just for passing the Agility filter, folks. That's huge. Guess who doesn't pass the Agility filter this year!

  • Emmanuel Ogbah
  • Shaq Lawson
  • Kevin Dodd
  • Noah Spence
  • Jonathan Bullard

Which is a big reason why I've been so vehemently against drafting any of those guys on this board. Ogbah is hands down the best athlete of the five and that puts him in a category where his ceiling would basically be Brian Orakpo. Now don't get me wrong, I like Brian Orakpo as much as the next guy, but he's also a pretty significant outlier in this context. He's hands down the best player to ever get drafted in the 1st round and fail my Agility filter, but after him? Adrian Clayborn, Napoleon Harris, Derrick Morgan, Jarvis Jones. It's not a pretty list of names. So if you're on the Ogbah bandwagon, you're basically hoping to the football gods that he can figure out how to be Brian Orakpo 2.0, because otherwise? Every single other option is nothing short of ugly.

Everyone else on this list though? You're looking at player comps like Erasmus James, Ebenezer Ekuban, Aaron Maybin, Anthony Spencer, Brandon Graham and Larry English. It's really, really ugly if you get drafted in the 1st round as a player who's failed the Agility filter.

Now, to move past that, there are obviously other rounds that matter. The 1st round is just a really important pick and there are a lot of people that want the Falcons to draft an edge rusher there. Personally, I'd rather we draft an off-ball LB or even a WR based on this year's class strengths, but that's an opinion for another thread. As far as later rounds, here's the kicker: Players drafted up to round 3 who pass the Agility filter will, on average, have a higher AV per year than player drafted in ANY ROUND who FAIL the Agility filter. I've done the work and it goes like this...

Players drafted in rd 1 who pass Agility filter: 6.27 AV per year
Players drafted in rd 2 who pass Agility filter: 4.45 AV per year
Players drafted in rd 3 who pass Agility filter: 3.43 AV per year
Players drafted in rd 1 who FAIL Agility filter: 3.41 AV per year

So there's that. I'd call that pretty significant. This is also why I've been so high on Bronson Kaufusi - he passes the Agility filter and is generally projected as a top-100 player. He's also highly touted by PFF, so that's another piece of the puzzle. Obviously PFF isn't the end-all be-all but what they DO offer is basically which players were really really good college football players, and if they like Kaufusi, AND he's got the athleticism, AND we know he's the son of a coach, I do the math and that's the kind of guy I want on my team. The problem is that he's a mid-tier athlete past, so his AV per year projection is only about 3.56. Don't get me wrong, that'd be good for a 3rd rounder - on average a 3rd round edge prospect will have an AV per year of 2.96, so Kaufusi is projected to out-perform that, but it's still not great.

That being said, Kaufusi isn't the only non-1st round talent to pass the Agility filter. There are a few other guys projected to go top 100 that pass it as well. Shilique Calhoun is one. He's not really an elite athlete or even a top tier athlete, he's in my mid-tier, but mid-tier guys who go in the 2nd round and pass the Agility filter still tend to out-perform their draft position with a 4.43 AV per year on average. On average a 2nd round edge defender (looking at every draft class since 1999) has a career AV per year of 3.88, so it's another slight improvement for Calhoun. Funny enough, Ogbah's projection being a top tier athlete who fails the Agility filter is 4.32 AV per year, which is right between Calhoun and Kaufusi, but those guys are 2nd and 3rd rounders respectively while Ogbah is projected to go in the mid-late 1st.

Every other option is pretty bad. The only guy left who could potentially be a blip on my radar is Charles Tapper. He's at a point where a 3 cone time could make or break him. 7.07 or better and he'd not only pass my agility filter, but he'd enter the discussion with Joey Bosa for best EDGE prospect in this class with a top-tier athleticism score as well. Those top-tier athletes drafted in the 1st round have a projected AV per year of 7.39, drafted in the 3rd round they still have a projected AV per year of 6.32, which is exceptional value for a 3rd round draft pick. He'd need to run a 3 cone of 6.62 to enter my "elite" tier, which is not only highly unlikely, but literally unprecedented for a player drafted in the 3rd round. If that 3 cone time would push him up into the 2nd round though, we're looking at an AV per year of 7.48, which is actually better than the top-tier athletes who pass and get drafted in the 1st round. That elite tier is a special class of athlete...only 5 guys have ever hit that mark, and the average AV per year regardless of round is 8.95. It's pretty ridiculous. But anyway, I'm digressing again.

At the end of the day, there are very few edge rushers in this class worth having. Joey Bosa is the main one. After him, Leonard Floyd is the only other guy who passes the agility filter, is projected as a 1st rounder, and lands in the top tier of athletes. The biggest issues with Floyd are a lack of production and the fact that the Giants will probably take him at 10 so we probably won't even get him. Regardless, the 17th overall pick is a really bad spot to need an edge rusher this year. The value just isn't there.

Moving past that, this just flat out isn't the year we should draft an edge rusher. Last year was perfect for it, this year the value doesn't show up. That's why I keep on trying to grab a late round sleeper because I know how bad the value is early and I'd rather take a flier on a really good athlete to eventually contribute to our pass rush in the 7th round than try to put my faith in a player who doesn't provide great value with one of our first four picks.

Late round potential sleepers we might consider with our 7th round pick include...

  • Virginia DE Trent Corney. Was asked to play contain, but has the athleticism to be a legitimate pass rush threat. Great against the run.
  • Central Arkansas DE Jonathan Woodard. Played a little of everything from 3 tech to OLB, but produced consistently at the FCS level.
  • Montana OLB Tyrone Holmes. He's got the kind of production you look for in an NFL prospect playing in the FCS. Really fun to watch.
  • Vanderbilt OLB Stephen Weatherly. Kind of an all-around guy who can rush and even potentially cover, plus he had a big blocked FG.
  • Miami (OH) DE Bryson Albright. Another smaller school kid with some athletic upside. Really impressive bend, but patty-cakes too much.
  • Virginia Tech DE Dadi Nicholas. Amazing burst off the line but wayyyy too light to be a full time DE in the NFL. Pack on that weight, son!

Alright guys I'm all out. I'm going to sleep for about 4 hours, so if you guys have any questions regarding anything I said here, ask them, and when i wake up in the morning I'll respond to as much as I can.

Link to comment
Share on other sites

  • Replies 59
  • Created
  • Last Reply

Top Posters In This Topic

I happen to be a person who believes that if you define enough variables and have the processing power, you can somewhat predict the future mathematically.  That said, it's hard for me to really see your formula's accuracy.  I don't know all the variables you use or how you weight them.  I can only assume you have analyzed past performers and created or improved a formula that agreed with those findings.  I haven't dug into your previous dedicated writings.  Things that matter that I wonder how you measure are:

  1. Competition level
  2. Teamate quality, role and system
  3. Injuries that hamper measureables
  4. Hard skills (hands, elusiveness, power).  You may say production but that has a dependency on at least 1 and 2.
  5. Upside.  Not the mythical hope but measuring a players performance arc (peaking or still learning craft)...your formula 'seems' to say "This is it...best they'll ever be."

I guess I could keep going on and I'm sure you've answered this before.  You're clearly passionate about it and love seeing people make stuff like this.  Hope you can (or are) trying to do more with it than sell TAFT on it. 

Link to comment
Share on other sites

3 minutes ago, Monolith2001 said:

I happen to be a person who believes that if you define enough variables and have the processing power, you can somewhat predict the future mathematically.  That said, it's hard for me to really see your formula's accuracy.  I don't know all the variables you use or how you weight them.  I can only assume you have analyzed past performers and created or improved a formula that agreed with those findings.  I haven't dug into your previous dedicated writings.  Things that matter that I wonder how you measure are:

  1. Competition level
  2. Teamate quality, role and system
  3. Injuries that hamper measureables
  4. Hard skills (hands, elusiveness, power).  You may say production but that has a dependency on at least 1 and 2.
  5. Upside.  Not the mythical hope but measuring a players performance arc (peaking or still learning craft)...your formula 'seems' to say "This is it...best they'll ever be."

I guess I could keep going on and I'm sure you've answered this before.  You're clearly passionate about it and love seeing people make stuff like this.  Hope you can (or are) trying to do more with it than sell TAFT on it. 

These numbers are all based exclusively on athleticism. Height, weight, speed, jumps, that kind of stuff.

Link to comment
Share on other sites

Pretty cool stuff. I tend to agree with the conclusion the OP comes to. The only thing that stuff can't measure is heart. Does the player have that mental toughness and desire to be the best? Instincts aren't measurable in this study either. I think Quinn said, "who cares about 40 times, does he play fast?" Game speed matters more than measurements and that's where instincts come in. If you have a high football IQ and can anticipate the plays you can play faster than your timed speed. Jerry rice wasn't that fast but you couldn't know he would be so focused and so dedicated to the game that he just beat you with pure smarts instincts and constant training. Even Freeman doesn't have great numbers/measureables but the dog in him and the dedication shine through on game day. I think Scooby Wright is that kind of player.

Link to comment
Share on other sites

Just now, FalconFanSince1970 said:

I never thought I'd see a post longer than one of The Dirty Word's, but dam if you didn't double him up.  Where are the measureables for heart, motor, instinct, skill, football acumen,  etc.?

inaccessible to the public

Link to comment
Share on other sites

3 minutes ago, Kayoh said:

These numbers are all based exclusively on athleticism. Height, weight, speed, jumps, that kind of stuff.

Oh okay.  Was thinking it was a bit more robust than that but like I said, I haven't read a lot of your previous stuff beyond reading a bit of the back and forth in other threads.  Athleticism is an important measure but not sure it is all that useful in isolation. 

Link to comment
Share on other sites

Just now, Monolith2001 said:

Oh okay.  Was thinking it was a bit more robust than that but like I said, I haven't read a lot of your previous stuff beyond reading a bit of the back and forth in other threads.  Athleticism is an important measure but not sure it is all that useful in isolation. 

For most positions it isn't, but for edge rushers it's INCREDIBLY important. Vital, even.

Link to comment
Share on other sites

30 minutes ago, Kayoh said:

inaccessible to the public

 

31 minutes ago, FalconFanSince1970 said:

I never thought I'd see a post longer than one of The Dirty Word's, but dam if you didn't double him up.  Where are the measureables for heart, motor, instinct, skill, football acumen,  etc.?

 

The things that none of us have access to b/c we are not in meetings, interviews, we haven't called former coaches, asked other players, etc...

When scouts go to games during the season, they take in a practice, see the players warmup, etc....things that never show up in the "underwear olympics" (ie. NFL Combine)

I understand there is a certain level of athleticism to be competitive at the NFL level, and to be good at a certain position....however:

Many before have had 1400+ SATs and failed out of college or werent successful....get my drift???

There are so many variables, you use what you can to make the best decisions, but sometimes, maybe a lot of times, you (in this case, TD and Quinn) use your instincts and "gut" to make decisions as well...

Edited by Matt_The Iceman_Ryan
Link to comment
Share on other sites

5 minutes ago, Matt_The Iceman_Ryan said:

 

 

The things that none of us have access to b/c we are not in meetings, interviews, we haven't called former coaches, asked other players, etc...

When scouts go to games during the season, they take in a practice, see the players warmup, etc....things that never show up in the "underwear olympics" (ie. NFL Combine)

I understand there is a certain level of athleticism to be competitive at the NFL level, and to be good at a certain position....however:

Many before have had 1400+ SATs and failed out of college or werent successful....get my drift???

There are so many variables, you use what you can to make the best decisions, but sometimes, maybe a lot of times, you (in this case, TD and Quinn) use your instincts and "gut" to make decisions as well...

You can tell a lot about a player's heart, motor, instinct, skill and football acumen just by watching dam tape. That publicly inaccessible bush it is just another way of saying all I give a sht about is athleticism, agility and my patented Kayoh score. 

Link to comment
Share on other sites

3 hours ago, Kayoh said:



The next refinement I've made to the formula is I've separated it into rounds. So I can tell you that 1st round players who pass my Agility filter average a career AV per year of 6.27, while 1st round players who *fail* the Agility filter average a career AV per year of 3.41. That's an 84% jump in value just for passing the Agility filter, folks. That's huge. Guess who doesn't pass the Agility filter this year!

  • Emmanuel Ogbah
  • Shaq Lawson
  • Kevin Dodd
  • Noah Spence
  • Jonathan Bullard

 

I am not sure why you have Bullard listed here. Like Lawson if you watch him and think he is a primary edge rusher you probably should be looking for another job, but I think you have talked about that before. Both of those players would do better on the inside and will be marginal as rushers from the outside.

Edited by Sobeit
Link to comment
Share on other sites

6 minutes ago, FalconFanSince1970 said:

You can tell a lot about a player's heart, motor, instinct, skill and football acumen just by watching dam tape. That publicly inaccessible bush it is just another way of saying all I give a sht about is athleticism, agility and my patented Kayoh score. 

I think of Jared Allen coming out of college...6'6" 265lbs...ran 4.7 and only did 16 bench reps at 225...1.68 10 yard split, etc....

Kayoh never highlights the players that dont meet his metrics, but have been successful in the NFL...

JA69 had 136 sacks in the NFL, 10 season with 7+ sacks....I will take that guy all day long over Joey Bosa (he is not as athletic as Kayoh thinks, and he is pretentious as %$@*)

Link to comment
Share on other sites

19 minutes ago, Matt_The Iceman_Ryan said:

I think of Jared Allen coming out of college...6'6" 265lbs...ran 4.7 and only did 16 bench reps at 225...1.68 10 yard split, etc....

Kayoh never highlights the players that dont meet his metrics, but have been successful in the NFL...

JA69 had 136 sacks in the NFL, 10 season with 7+ sacks....I will take that guy all day long over Joey Bosa (he is not as athletic as Kayoh thinks, and he is pretentious as %$@*)

Bingo. Dude is way too caught up in his edge scores. 

Link to comment
Share on other sites

Nice write up, Kayoh.  I believe that metrics are a vital part of assessments, but not the whole story.  Having said that, let's keep this discussion focussed on metrics.  I have three questions for you: 

1.  E. Ogbah scored very well in pSPARQ (133), which puts him in the 82nd percentile of active NFL players for his position.  Which performance measurement do you weigh more than pSPARQ?

2.  How do you feel about Alex McCallister  as a 4th round choice?

3.  Sorry for my ignorance, but what does AV mean?  Is it a relative value index?

Link to comment
Share on other sites

I don't want to speak for Kayoh here, but I don't think he's saying his EDGE score is the only thing.

There are failures that are above his EDGE cutoff, and successes below it.  There are just more successes above it, and fewer below it.

Think of it this way...Matt The Iceman mentioned SAT before.

Someone with a 1400 can fail out of college from lack of drive, topping out, etc.  Someone with a 900 can excel and learn and become much better than you expected.

But if you're going to bet on someone to succeed in college, is it the 1400 or the 900?  The odds are radically better that the 1400 is going to do fine.

And so it is with any numerical analyses.  They don't cover everything, they cover a specific measure, and show how much it is correlated.  If a formula truly shows that elite athletes have twice the "career value" of non elites...there's still tons of room for the variety between elite athletes being made up by their drive and their want to, and there's room in the variability in the non athletes that is likewise made up by those.

No one number tells the story, but it can tell enough of it that you should listen.

Link to comment
Share on other sites

Kayoh,

I applaud your efforts and find the results intriguing. There are about 4 posters on TATF that contribute anything of value, and you are one of them. 

I would, however, suggest sharing your thoughts and analysis with a legitimate publisher, rather than the morons running loose here, who can't understand the value of metrics. Nor are most sophisticated enough to understand that it metrics represent a component of the decision making process, not the final word on any particular prospect.  

For the record, I agree that there is a dearth of 1st-round-value edge rushers in this draft. I don't really want to see a first round pick used on the interior line (we have umpteen guys that can kick inside on pass downs, and between Jackson at the nose and 'Shede/Jarret at the 3-technique, we're set on run downs).  

I see our inability to cover running backs and tight ends as a greater problem than our pass rush. Our pass rush will be better with Vic having another year (and healed shoulder) under his belt, and the addition of Shelby. 

I want an off-ball linebacker in the first (should we stay there). 

I am not crazy about Lee, but I'll take him. However, Dled (I know...) mentioned to one of the local stations that Lee's value had fallen in the eyes of those at Flowery Branch. 

I am not crazy about Ragland, but I guess I'll take him. I doubt he is a mid-first-round value though. 

I would love for us to land Floyd, and I want to play him at the Mike or Will spot. I see his coverage potential as his best value. Sure, he can probably bend around the edge as well as, or better than, Vic, but I like his height/length/speed for covering tight ends and crossing routes. I'd rather have Vic/Clayborn/Jarret hunting Cam, et al., while Floyd prowls the middle as an 8-foot-tall terror, swatting balls down like an NBA center.  

That said, he'll be long gone by the time we pick. 

Accordingly, and because Jones and Cravens may both be there in the second, I think that we actually may be looking at an offensive lineman in the first. And the funny thing is, I (as a defensive-minded guy) am alright with that, given the way the board lays out. 

I can get on board with a so-so defense if it means Matt has a top-5-caliber offensive line. 

 

Edited by Falcon Ben
Link to comment
Share on other sites

1 hour ago, Matt_The Iceman_Ryan said:

I think of Jared Allen coming out of college...6'6" 265lbs...ran 4.7 and only did 16 bench reps at 225...1.68 10 yard split, etc....

Kayoh never highlights the players that dont meet his metrics, but have been successful in the NFL...

JA69 had 136 sacks in the NFL, 10 season with 7+ sacks....I will take that guy all day long over Joey Bosa (he is not as athletic as Kayoh thinks, and he is pretentious as %$@*)

Probably because the majority of his posts are about first round EDGE players. Jared Allen was a 4th rounder. And honestly, looking at the numbers, I wouldn't be shocked if Jared Allen scored very well. His agility numbers look really good anyway. 6'6 265 with a 7.11 3 cone and 4.34 shuttle?

Link to comment
Share on other sites

4 minutes ago, HEIST said:

Probably because the majority of his posts are about first round EDGE players. Jared Allen was a 4th rounder. And honestly, looking at the numbers, I wouldn't be shocked if Jared Allen scored very well. His agility numbers look really good anyway. 6'6 265 with a 7.11 3 cone and 4.34 shuttle?

Not gonna give specific values, but these would all fall in the average or below categories for Kayoh's formula...

Tamba Hali, Terrell Suggs, James Harrison, Justin Tuck...

Link to comment
Share on other sites

5 minutes ago, takeitdown said:

But if you're going to bet on someone to succeed in college, is it the 1400 or the 900?  The odds are radically better that the 1400 is going to do fine.

True but as I said above, you don't really need math to prove that.  It really is a formula to back up what people already can see.  I'm not saying the formula has no merit and I think it does demonstrate athleticism numerically.   Still I have a hard time seeing the good predictive science in a formula that only measures gross motor and size.  Skill, institutional awareness and instincts matter too much to be ignored.

Link to comment
Share on other sites

1 minute ago, Monolith2001 said:

True but as I said above, you don't really need math to prove that.  It really is a formula to back up what people already can see.  I'm not saying the formula has no merit and I think it does demonstrate athleticism numerically.   Still I have a hard time seeing the good predictive science in a formula that only measures gross motor and size.  Skill, institutional awareness and instincts matter too much to be ignored.

This is for edge rushers.  Has nothing to do with anything other than pass rushing.  For any other positions your concerns would be true.  Is the player athletic enough to get around the tackle and get to the QB.

Link to comment
Share on other sites

37 minutes ago, takeitdown said:

I don't want to speak for Kayoh here, but I don't think he's saying his EDGE score is the only thing.

There are failures that are above his EDGE cutoff, and successes below it.  There are just more successes above it, and fewer below it.

Think of it this way...Matt The Iceman mentioned SAT before.

Someone with a 1400 can fail out of college from lack of drive, topping out, etc.  Someone with a 900 can excel and learn and become much better than you expected.

But if you're going to bet on someone to succeed in college, is it the 1400 or the 900?  The odds are radically better that the 1400 is going to do fine.

And so it is with any numerical analyses.  They don't cover everything, they cover a specific measure, and show how much it is correlated.  If a formula truly shows that elite athletes have twice the "career value" of non elites...there's still tons of room for the variety between elite athletes being made up by their drive and their want to, and there's room in the variability in the non athletes that is likewise made up by those.

No one number tells the story, but it can tell enough of it that you should listen.

What I was trying to say, but more eloquently written...

I have no doubt our FO and coaching staff are putting an emphasis on metrics...but I still think an "edge" score or "agility score is not the only metric that can show success for pass rushers...

Reggie White...how well did he "bend" around the the edge? How good was his agility?

I think Kayoh doesnt account for power and length in his formula, which are also great metrics for pass rushers...

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

 Share


×
×
  • Create New...