32 Counter Pass
Footballguy
Agree. Outstanding!Flying Elvis said:Awesome effort. Thanks!
Especially nice this year for a wr starved superflex I have with picks 8, 14 and 20. One more data point to guide me is always appreciated.
Fantasy Football - Footballguys Forums
Welcome to Our Forums. Once you've registered and logged in, you're primed to talk football, among other topics, with the sharpest and most experienced fantasy players on the internet.
Spaceshuttle (1 Viewer)
- Thread starter Interseptopus
- Start date
32 Counter Pass
Footballguy
Arcega Whiteside ran a 4.8 - 4.49. Plenty fast enough. He was as close to money as any player LY. If Shaw wasn’t so conservative AW would have had even bigger numbers. He will shoot up the draft boards with a good landing spot (Pats, Eagles, 9ers,)
cloppbeast
Footballguy
Am I the only one who doesn't find this stuff profound? Guys who produce in college are more likely to produce in the NFL. Kind of obvious, right ? It stands to reason a 34% dominator score is better than 22.0%. Quantifying how much better would be more useful.
Doucheness unintended.
Doucheness unintended.
Men-in-Cleats
Footballguy
It isn't profound that guys who produce in college are more likely to be productive pros, what is being done here is attempting to quantify that productivity in a fashion that takes into account different offensive systems, etc. What exactly is considered the line where you are considered highly productive vs. not is the question here, not that more productive is better.
Ilov80s
Footballguy
Pro days times are 0.5 sec faster than combine times so he is a little slower than that reflects but he is still plenty fast given his size.
Ilov80s
Footballguy
I think the key is the age aspect. For example, one might look at Hakeem Butler's numbers and dominator and come to the conclusion he was a more dominant college player than OBJ. He had more yards and a higher dominator. The issue is we are comparing a 22 year old senior season with a 21 year old junior season. A better comparison between the 2 would be comparing what Butler did as a junior to what OBJ did as a junior or trying to wrap our heads around the idea that at age 22 Butler was beating up on the Big 12 and OBJ was a Pro Bowler in the NFL. This is one of the reasons I really like Harry over Butler. Their last year stats look similar but Harry was a 3rd year 20 year old and Butler was a 4th year 22 year old.
Exactly. 800 yards on Rutgers is really impressive because they only threw for 1600 yards last year. 800 yards for Washington State is nothing since they threw for 4800 yards.
Catbird
Footballguy
Great tool to have in the box. I'm in agreement with those that think you need to weight dominator against the team's talent, at least in extreme cases, but breakout and dominator (both at breakout and last college year) seem useful as warnings about someone who is not productive at an young age. With the clear correlation though, and if I can integrate that as a piece together with athleticism, what I can see from film and what others better trained to look for it can see from film (and what ZWK's formula says) at least I'll feel like I touched all the bases.
Sorry if I missed it. I tried to read back through but didn't find a link. Can one get breakout age and dominator percentages already quantified for the 2019s?
Sorry if I missed it. I tried to read back through but didn't find a link. Can one get breakout age and dominator percentages already quantified for the 2019s?
steelers1080
Footballguy
Yeah, the basic premise is that if you perform at a high level at a younger age, that typically signifies success in the NFL. A good recent example was Juju, he was great when he was 19, and even though he had a down junior year, he's been great in the NFL (obviously landing spot also matters). Calvin Ridley it remains to be seen how successful an NFL career he has, he was dominant, but at an older age. And it's more expected that a player who's 22 - 23 should perform better against players that are 18 - 19 because they have more experience.
This is like the combine, it shouldn't be the only thing you look at, but it can help solidify rankings.
cloppbeast
Footballguy
The conclusion is probably sound, no issue avoiding guys with older breakout age. But the threshold thing though doesnt tell us much. These 2 things being correlated, bad results on the far end are expected. So you can set your threshold wherever you want to rule out whichever players you want. Or you can do the same thing with any correlated data: 40 time, height, weight, arm length...... Although I admit breakout age and dominator should have better correlation.
We cant ignore the NFL draft as the best predictor of fantasy points. It encompasses all this data and also unquantified human elements. No doubt efficiency exists though. Breakout age maybe is not considered as much as it should be. It would be interesting to see a comparison of players with similar draft capital and disparing breakout age. Youd also have to include the failures, which is another problem I have with this analysis. But I digress.
Last edited by a moderator:
cloppbeast
Footballguy
Can someone explain how the percentile for breakout age is figured up? Is it based on all college receivers or those drafted into the NFL? Does it include UDFAs?
Also, suppose the receiver never gets the 20% dominator score.
Also, suppose the receiver never gets the 20% dominator score.
Last edited by a moderator:
Ilov80s
Footballguy
I believe it is based on all those who get entered into the the system which would include drafted players and UDFA. For example guys like KD Cannon and Andy Jones are in there. It’s not taking into account every college player. A 5th senior with 400 yards and no pro prospects aren’t included. As for never reaching 20th percentile, Martavis Bryant is an example. He just doesn’t have a breakout age listed. His best season at Clemson was only 18.5%. That’s a very rare case where he was a pro prospect but was sharing the field with Watkins, Humphries, Mike Williams, Jordan Leggett and Charone Peake. That’s a total of 6 NFL pass catchers on a team and is a good reason to make Bryant a rare exception when looking at breakout and dominator...although Bryant’s career has just been ok and he’s never been better than a WR3 so maybe the warning signs were right here.
Humphires is even worse when we look at dominator and has no breakout age. Also he’s never been better than a low end WR3. It will be interesting to see if he does anything of note with the Titans.
Last edited by a moderator:
cloppbeast
Footballguy
I'm not debating the probability of doing something vs probability of not doing it. They are the same thing. Hit rate + miss rate = 1. Well, they dont have the same value, but knowing one means you know the other.
I am debating whether we have here an accurate predictor for not a top 12 finish. We don't. You can't draw conclusions about a population (receiver prospects) from a not random sample (top 12) without knowing the distributions of both. Do you know how many receiver prospects drafted in the last 10 years meet your criteria?
My stats teacher once brought up this study which showed something like 70% of car accidents happen within 5 miles of the drivers home. The author of an article in the paper falsely concluded you are at more risk of an accident close to home. Not true. You are more likely to wreck your car close to home because you are more likely to be driving there.
It's easy to find the fallacy here because you can assume the distribution of distance from the home, that you are more often close. In our case though, we cant assume the distribution of drafted propects meeting the criteria. To find out would take a while and some effort performing a data entry for more than 200 prospects. Who wants to do that in their free time? Nobody.
Last edited by a moderator:
While true, nobody is claiming an actual percentage of probability here. We can't say this method determines the probability of a miss at x%, therefore, if a WR meets the defined criteria, he has an (100-x)% of being a WR1
Think of this a different way. This is not trying to be applied to all prospects to determine an exact hit/miss probability. This is not trying to be incredibly specific with "all WRs not meeting criteria have a X% chance of missing".
What it is trying to do is indicate. Through other tools, you have determined your set of rankings. You have two prospects you have fairly even and are trying to decide between. Player A meets the criteria and Player B does not. Knowing that 80% of the WR1s over the past 10 years met the criteria and 20% did not. Which player do you rank higher, assuming all else is even?
Note the intent isn't to use the criteria to sort all prospects. It it simply being used to say player B has less of a chance of being a WR1 based on the last 10 years of WR1 finishers. Not that he definitely will miss or can't be hit. It is just trying to indicate vs determine outcomes.
kittenmittens
Footballguy
Awesome thread. Dominator and breakout age are two of the more predictive indicators according to the analytics.
I'm trending towards mode and more being out on Butler. I almost never draft 23 year old rookies. Hoping the other top WRs who are younger fall. I'm much more excited about Harry and Brown and to some degree Metcalf.
I'm trending towards mode and more being out on Butler. I almost never draft 23 year old rookies. Hoping the other top WRs who are younger fall. I'm much more excited about Harry and Brown and to some degree Metcalf.
Ilov80s
Footballguy
I think JJAW should be near that grouping as well. He’s got the dominator and breakout age. He has prototypical size and while he’s not a great athlete, he at least has average speed.
ZWK
Footballguy
Player Profiler does have 2017 as Butler's breakout season. They have his breakout age as 21.3, which was his age on 9/1/2017, so they're counting 2017 as his breakout season. They say that's a 30th percentile breakout age (worse than average).
Dominator is based on market share of passing yards (which was 19.6% for Butler in 2017) and market share of passing TDs (which was 25.9% for Butler in 2017). I think that Dominator is just the average of those two numbers (which was 22.8% for Butler in 2017), and it has to be at least 20% for them to count it as a breakout season.
Dominator is based on market share of passing yards (which was 19.6% for Butler in 2017) and market share of passing TDs (which was 25.9% for Butler in 2017). I think that Dominator is just the average of those two numbers (which was 22.8% for Butler in 2017), and it has to be at least 20% for them to count it as a breakout season.
Ilov80s
Footballguy
Last edited by a moderator:
Biabreakable
Footballguy
Why did you thank ZWK for pointing out that he did break out at age 21 in 2017?
The only thing you need to get past here is your error.
kittenmittens
Footballguy
Definitely a player i am targeting in the 2nd round, especially with a good landing spot.
Stop the slap fight stuff back and forth between you guys. Getting pretty tiresome. Focus 100% on football without shots at each other. That goes for everyone.easy there. I was thanking him for clarifying because I thought he broke out in 2018 and wanted to give him some leeway since he had a good 2017, but seemed to just miss a breakout based on what cloppbeast said in the Butler thread. So technically not my error.
no reason to get your undies in a bigger bundle than they usually are. I look forward to you riding clopp on his error in the Butler thread
kittenmittens
Footballguy
Draft capital is probably the 3rd leg of this table.
As @menobrown mentioned in the Jacobs thread, draft day invites might give us a preview. Apparently the 2 WRs invited to the draft are DK Metcalf and Hollywood Brown. I would have figured the other Brown. Maybe NFL teams are seeing DJax 2.0 in Marquise?
If Hollywood goes in the mid or late first round, does that do anything to your draft board?
As @menobrown mentioned in the Jacobs thread, draft day invites might give us a preview. Apparently the 2 WRs invited to the draft are DK Metcalf and Hollywood Brown. I would have figured the other Brown. Maybe NFL teams are seeing DJax 2.0 in Marquise?
If Hollywood goes in the mid or late first round, does that do anything to your draft board?
cloppbeast
Footballguy
Wasn't aware. Thanks.
I'm still not sure there's no selection bias going on here anyway. So I wont put as much emphasis on BOA as everybody else around here. Not passing on Butler or downgrading him if an NFL team makes him a 1st round pick.
Last edited by a moderator:
Ilov80s
Footballguy
Crazy that he might go that high given the foot injury and the crazy low weight. He's basically JJ Nelson with a foot injury.
I like Hollywood but not at much as NFL teams so in his case it won't. IMO NFL teams put a higher value on speed and ability to stretch the defense then fantasy football fans so that bumps up NFL stock of a guy like Brown. Probably Metcalf for that matter. I don't get fantasy points for my WR opening up the field for everyone else.
Also in general I don't put nearly as much emphasis on WR draft capital as RB. It matters but the degree compared to RB's diminishes greatly. And that reasoning for me is as simple as most of the time only one RB is on the field at once and if a team commits a higher round pick on a RB his chances of being that one RB on the field and relatively quickly goes up considerably higher IMO. If a WR has talent he can usually push his way into multiple WR sets at least and work up from their. RB's are just so cheap to sign and easy to acquire that in their case if a team is going to commit a high pick to a RB it usually stands to reason they have a plan for the RB. Sometimes, like with Ronald Jones, a player is just not up to it and sometimes even a high pick can get blocked (Penny) but odds suggest a RB chosen in first two rounds is getting usable work early.
Biabreakable
Footballguy
He was talking about this because you identified Butler as not breaking out in 2017 which was inaccurate. You acknowledged this and even thanked ZWK for the clairification then proceeded to say you were still concerned about the break out age. That makes no sense.
kittenmittens
Footballguy
Looking at mock drafts is getting quite depressing. We are getting close enough to the draft now where they are getting more accurate... I keep seeing Washington, Tennessee, and Baltimore drafting first round WRs.
All 3 of those destinations is basically a do not draft situation for me.... Although maybe I could handle Washington...
It's amazing how the wrong team drafting a player can essentially destroy that players entire career.
All 3 of those destinations is basically a do not draft situation for me.... Although maybe I could handle Washington...
It's amazing how the wrong team drafting a player can essentially destroy that players entire career.
Ilov80s
Footballguy
I agree but things can change fast in the NFL. When Corey Davis landed with the Titans it looked a dream scenario pairing him with a young star QB in the making. Two years later and Mariota looks like a bust. Conversely, I was a bit worried about JuJu in Pittsburgh. Martavis was coming back, Brown and Bell account for 280 targets and Ben was talking about retiring. Two years and two good fantasy seasons later and it's the Juju-Ben show in Pittsburgh.
Men-in-Cleats
Footballguy
Mock drafts almost always overdraft RB's and WR's instead of the less sexy picks and once the draft rolls around you see more OL, DL, CB taken early while projected 1st round RB and WR will end up going in the 2nd or 3rd.
Ilov80s
Footballguy
Yeah I have played around with the draft networks mock draft machine, it’s fine I till you get into round 4 or so and then 80% of the recommended players are RBs and WRs.
cloppbeast
Footballguy
Ilov80s
Footballguy
No but Lil’J was never really a legit draft prospect despite hit dominator and age.
cloppbeast
Footballguy
cloppbeast
Footballguy
No longer wanting to hijack the Parris Campbell thread with talk of specificity and sensitivity, moved here @Dr. Dan
With regards to sensitivity, since we aren't screening for turbuculos or reading MRIs here, it's a bit incomplete. We dont intend to rule out, but to make a choice.
Sensitivity and specificity are two sides of a coin, both need considered in this case. Ruling out a high ba/low dominator due to sensitivity shown this study, in favor of low ba/high dominator but without knowing the specificity of misses who also meet the criteria - it's just not a logically sound choice.
It could well be the case you're correct, I admit. But you havent conclusively proven anything here as it is incomplete. Making insults at me wont change the fact.
With regards to sensitivity, since we aren't screening for turbuculos or reading MRIs here, it's a bit incomplete. We dont intend to rule out, but to make a choice.
Sensitivity and specificity are two sides of a coin, both need considered in this case. Ruling out a high ba/low dominator due to sensitivity shown this study, in favor of low ba/high dominator but without knowing the specificity of misses who also meet the criteria - it's just not a logically sound choice.
It could well be the case you're correct, I admit. But you havent conclusively proven anything here as it is incomplete. Making insults at me wont change the fact.
ConnSKINS26
Footballguy
I'm not saying this is true for everyone...but some would say that the most rational way to make a good choice, with the best processes leading to the best results over time, would be to rule out the bad choices until you have a smaller pool, raising the likelihood of making a good choice.
cloppbeast
Footballguy
Analysing a process is necessity in determining whether the subsequent choice will lead to good results. Right?
Last edited by a moderator:
cloppbeast
Footballguy
All but 3 of the wrs who've had a top 12 season meet the criteria of being black. Thats like 95% or something. Is this a trend? We should avoid selecting a white guy as he is likely a miss?
TheWinz
Footballguy
Remember the old days, before the internet, when you bought a 3 month old FF magazine, drafted directly from their cheat sheet (likely taking a QB in rd 1), finding out your star RB had retired, or that you started a WR for 3 weeks before you found out he was on IR? Your commish would have to flip through the newspaper on Monday to read the box scores. I kinda miss those days, don't you?
cloppbeast
Footballguy
This is exactly what I've been saying the whole time about BA and Dominator rating. Its called selection bias.
Top wrs 95% meet black criteria - all NFL wrs 95% meet black criteria
Top wrs 71% meet your criteria - all NFL wrs ?% meets your criteria.
cloppbeast
Footballguy
You dont have to buy Country Crock from the WalMart, you can churn your own butter on your front porch. If that's what you want to do.
The good ol days.
cloppbeast
Footballguy
ConnSKINS26
Footballguy
I know you think you're dunking on the opposition left and right in here, but you aren't.
I'm about as unbiased an observer as you're gonna get on an opinionated forum like this. I'm arguing FOR Campbell in that thread, and don't really utilize BA/DR in my dynasty roster construction unless it's subconsciously. I've started taking it into consideration in regard to devy prospects to predict trade values of guys, because I know my leagues, and that's pretty much the extent of it right now.
But I find it fascinating and potentially helpful. My mind isn't closed to it. As an observer, rationally I can see the merits of these metrics.
All that to say...your posts in here aren't making much sense right now. Dan's are compelling.
Concept Coop
Footballguy
Have any breakout/dom% studies controlled for draft spot? Is the NFL behind here? Is there truly an inefficiency to exploit?
A few years ago a poster did a study that found that NFL teams were under-drafting small school WRs. For a few years at least, there was value in drafting small school WRs ahead of big school WRs drafted in the same range. (But the NFL quickly corrected for that.)
Was the NFL ever behind on these metrics? If so, have they caught up?
I think what clop is saying - is that correlation a isn’t predictive model in and of itself.
A few years ago a poster did a study that found that NFL teams were under-drafting small school WRs. For a few years at least, there was value in drafting small school WRs ahead of big school WRs drafted in the same range. (But the NFL quickly corrected for that.)
Was the NFL ever behind on these metrics? If so, have they caught up?
I think what clop is saying - is that correlation a isn’t predictive model in and of itself.
Last edited by a moderator:
cloppbeast
Footballguy
maybe I can't articulate my point as well as Dr. Dan. I can unequivocally say he's wrong, he hasn't found what he thinks he has. The link I posted above about Survival Bias could help explain.I brought up the race of wr1s as an absurd conclusion to be made using the same logic. Dr. Dan sort of gets it by bringing up that most wide receivers in the NFL are black; therefore most wr1s will also be black. But then fails to see the same flaw in his own conclusion. The only difference: we know most NFL receivers are black, we don't know how many meet his BA/Dom% criteria.
To be clear though, I'm not saying BA and Dom% aren't metrics to exploit. Only pointing out nothing here proves it, or further information is required.
Last edited by a moderator:
cloppbeast
Footballguy
cloppbeast
Footballguy
Well Dr. Dan, this has been fun. But it's starting to feel like an argument with my wife, so I will retire. Catch you on the flip side.
