It just means that 80% of the players that hit had good dominator ratings and early breakout ages(not surprising). Not that 80% of the players that checked those boxes make the grade. So breakout age and dominator rating in of itself does not have a predictive value. I'm pretty sure that the majority of all the college players that checked these boxes don't even get drafted.
We're all sleeping on Michael Pittman
- Hottoddies
- Player of the Year
- Posts: 2313
- Joined: Mon Jan 20, 2014 5:29 pm
Re: We're all sleeping on Michael Pittman
"Smart people learn from everything and everyone, average people from their experiences, stupid people already have all the answers." - Socrates
Re: We're all sleeping on Michael Pittman
So what is the mathematical reasoning for why anyone should care about Pittman’s breakout age?Hottoddies wrote: ↑Mon Jun 29, 2020 7:15 pmIt just means that 80% of the players that hit had good dominator ratings and early breakout ages(not surprising). Not that 80% of the players that checked those boxes make the grade. So breakout age and dominator rating in of itself does not have a predictive value. I'm pretty sure that the majority of all the college players that checked these boxes don't even get drafted.
12 Team .5 PPR - 1 QB, 2 RB, 2 WR, 1 TE, 1 WR/RB 20 man rosters, 5 man taxi est. 2018
QB: Dak, Pickett
RB: CMC, Taylor, Gibson, Dillon, Akers, Penny
WR: JJeff, Hollywood, Olave, Toney, Aiyuk, Jeudy, C. Davis, Boyd, C. Samuel,
TE: Njoku, Gesicki
2023 picks: 1.x, 2.x, 3.x, 4.x
QB: Dak, Pickett
RB: CMC, Taylor, Gibson, Dillon, Akers, Penny
WR: JJeff, Hollywood, Olave, Toney, Aiyuk, Jeudy, C. Davis, Boyd, C. Samuel,
TE: Njoku, Gesicki
2023 picks: 1.x, 2.x, 3.x, 4.x
-
- Combine Attendee
- Posts: 44
- Joined: Fri Jun 26, 2020 6:07 am
Re: We're all sleeping on Michael Pittman
I'm not a true believer in breakout age, but I will admit that this isn't correct. The breakout age truthers will tell you that you have the highest statistical chance at predicting an elite player at the NFL level, when you take into account their 'dominator rating'. The dominator rating takes into account the breakout age and team target share. Since breakout age is a main aggregate in the formula, it's extremely important to them. They suggest that the higher the dominator rating, the younger and higher the target share a player has, is directly correlative to their predictive NFL success.Hottoddies wrote: ↑Mon Jun 29, 2020 7:15 pmIt just means that 80% of the players that hit had good dominator ratings and early breakout ages(not surprising). Not that 80% of the players that checked those boxes make the grade. So breakout age and dominator rating in of itself does not have a predictive value. I'm pretty sure that the majority of all the college players that checked these boxes don't even get drafted.
I'm personally not a slave to the breakout age, but it's something interesting to take into account in order to break ties, IMO.
Re: We're all sleeping on Michael Pittman
Dominator rating does not factor in breakout age. It is simply the percentage of the total offense that player accounted for. A "break out" is hitting a certain DR threshold. What age they hit it at is BA.AkaliWarrior wrote: ↑Mon Jun 29, 2020 7:41 pm I'm not a true believer in breakout age, but I will admit that this isn't correct. The breakout age truthers will tell you that you have the highest statistical chance at predicting an elite player at the NFL level, when you take into account their 'dominator rating'. The dominator rating takes into account the breakout age and team target share. Since breakout age is a main aggregate in the formula, it's extremely important to them. They suggest that the higher the dominator rating, the younger and higher the target share a player has, is directly correlative to their predictive NFL success.
I'm personally not a slave to the breakout age, but it's something interesting to take into account in order to break ties, IMO.
This stuff is literally no different than playing blackjack. Just because you play blackjack by the book doesn't mean you're going to win the hand you're currently dealt. But if you play it by the book, every single hand, it increases your chances of winning in the long run. Unfortunately what it seems like you're looking for (which in blackjack would be the equivalent of counting cards I guess), does not exist in football. If you still don't understand then I don't know what more to tell you. Lol
12 Team SF IDP
QB - Mahomes, Lawrence, Mac
RB - Etienne, Walker, Rhamondre, Javonte, Akers, CEH
WR - AJB, Waddle, Higgins, Gabe, Toney
TE - Andrews, ISJ
DL - Quinnen, Highsmith, Josh Allen
LB - Lloyd, JOK, Gay, Bush
DB - Winfield, Love, Delpit
14 Team 1QB IDP
QB - Fields, Jimmy G
RB - Henry, Kamara, Swift, Herbert
WR - Adams, Juju, Jakobi, Theilen
TE - Hock, Everett
DL - Garrett, Buckner, Sweat
LB - White, David, Okereke, Jack
DB - Budda, CJGJ, Dugger, Byard
QB - Mahomes, Lawrence, Mac
RB - Etienne, Walker, Rhamondre, Javonte, Akers, CEH
WR - AJB, Waddle, Higgins, Gabe, Toney
TE - Andrews, ISJ
DL - Quinnen, Highsmith, Josh Allen
LB - Lloyd, JOK, Gay, Bush
DB - Winfield, Love, Delpit
14 Team 1QB IDP
QB - Fields, Jimmy G
RB - Henry, Kamara, Swift, Herbert
WR - Adams, Juju, Jakobi, Theilen
TE - Hock, Everett
DL - Garrett, Buckner, Sweat
LB - White, David, Okereke, Jack
DB - Budda, CJGJ, Dugger, Byard
-
- MVP
- Posts: 2732
- Joined: Sun Jul 06, 2014 5:33 pm
Re: We're all sleeping on Michael Pittman
OhCruelestRanter wrote: ↑Mon Jun 29, 2020 6:22 pmWe’re approaching the point where we just have to admit that we can explain it to you, but we can’t understand it for you.
And we’re there. I tried in good faith. This guy apparently wants proof that there’s a mathematical certainty that you can apply to breakout age.
Here’s his argument another way:
Us: Men who smoke are 23 times more likely to develop lung cancer.
PR0v3: isn’t this just a “the coin flipped 8 times thing?” How do you know that’s the real probability?
Us: no, this is data from many years and a large population
PR0v3: WHERE’S TEH MATH THAT SAYS I SHOULDNT SMOKE?
Sometimes people just aren’t going to understand things, and that’s fine.
COOGAN IS A CHEATER AND A THIEF
Re: We're all sleeping on Michael Pittman
Bruh. I’m just asking for the damn study so I can read it. Where the hell is it? It’s like you haven’t read any post I’ve made. Where the hell is the damn study so I can read it for myself and not have to just take your word for it? I can find 100 million peer reviewed studies online about the damages of smoking, can’t find a single one about breakout age.OhCruelestRanter wrote: ↑Mon Jun 29, 2020 8:14 pmOhCruelestRanter wrote: ↑Mon Jun 29, 2020 6:22 pmWe’re approaching the point where we just have to admit that we can explain it to you, but we can’t understand it for you.And we’re there. I tried in good faith. This guy apparently wants proof that there’s a mathematical certainty that you can apply to breakout age.
Here’s his argument another way:
Us: Men who smoke are 23 times more likely to develop lung cancer.
PR0v3: isn’t this just a “the coin flipped 8 times thing?” How do you know that’s the real probability?
Us: no, this is data from many years and a large population
PR0v3: WHERE’S TEH MATH THAT SAYS I SHOULDNT SMOKE?
Sometimes people just aren’t going to understand things, and that’s fine.
12 Team .5 PPR - 1 QB, 2 RB, 2 WR, 1 TE, 1 WR/RB 20 man rosters, 5 man taxi est. 2018
QB: Dak, Pickett
RB: CMC, Taylor, Gibson, Dillon, Akers, Penny
WR: JJeff, Hollywood, Olave, Toney, Aiyuk, Jeudy, C. Davis, Boyd, C. Samuel,
TE: Njoku, Gesicki
2023 picks: 1.x, 2.x, 3.x, 4.x
QB: Dak, Pickett
RB: CMC, Taylor, Gibson, Dillon, Akers, Penny
WR: JJeff, Hollywood, Olave, Toney, Aiyuk, Jeudy, C. Davis, Boyd, C. Samuel,
TE: Njoku, Gesicki
2023 picks: 1.x, 2.x, 3.x, 4.x
-
- MVP
- Posts: 2732
- Joined: Sun Jul 06, 2014 5:33 pm
Re: We're all sleeping on Michael Pittman
You want a peer reviewed study now? Like in the Journal of Dynasty Fantasy Football?PR0v3 wrote: ↑Mon Jun 29, 2020 8:22 pmBruh. I’m just asking for the damn study so I can read it. Where the hell is it? It’s like you haven’t read any post I’ve made. Where the hell is the damn study so I can read it for myself and not have to just take your word for it? I can find 100 million peer reviewed studies online about the damages of smoking, can’t find a single one about breakout age.OhCruelestRanter wrote: ↑Mon Jun 29, 2020 8:14 pmOhCruelestRanter wrote: ↑Mon Jun 29, 2020 6:22 pmWe’re approaching the point where we just have to admit that we can explain it to you, but we can’t understand it for you.And we’re there. I tried in good faith. This guy apparently wants proof that there’s a mathematical certainty that you can apply to breakout age.
Here’s his argument another way:
Us: Men who smoke are 23 times more likely to develop lung cancer.
PR0v3: isn’t this just a “the coin flipped 8 times thing?” How do you know that’s the real probability?
Us: no, this is data from many years and a large population
PR0v3: WHERE’S TEH MATH THAT SAYS I SHOULDNT SMOKE?
Sometimes people just aren’t going to understand things, and that’s fine.
COOGAN IS A CHEATER AND A THIEF
Re: We're all sleeping on Michael Pittman
So we’re dressing up in our statistician costume and playing make believe? Nothing backed by legitimate process? That’s cool if that’s the way it is, let’s just stop pretending like any of it actually matters.OhCruelestRanter wrote: ↑Mon Jun 29, 2020 8:27 pmYou want a peer reviewed study now? Like in the Journal of Dynasty Fantasy Football?PR0v3 wrote: ↑Mon Jun 29, 2020 8:22 pmBruh. I’m just asking for the damn study so I can read it. Where the hell is it? It’s like you haven’t read any post I’ve made. Where the hell is the damn study so I can read it for myself and not have to just take your word for it? I can find 100 million peer reviewed studies online about the damages of smoking, can’t find a single one about breakout age.OhCruelestRanter wrote: ↑Mon Jun 29, 2020 8:14 pm And we’re there. I tried in good faith. This guy apparently wants proof that there’s a mathematical certainty that you can apply to breakout age.
Here’s his argument another way:
Us: Men who smoke are 23 times more likely to develop lung cancer.
PR0v3: isn’t this just a “the coin flipped 8 times thing?” How do you know that’s the real probability?
Us: no, this is data from many years and a large population
PR0v3: WHERE’S TEH MATH THAT SAYS I SHOULDNT SMOKE?
Sometimes people just aren’t going to understand things, and that’s fine.
12 Team .5 PPR - 1 QB, 2 RB, 2 WR, 1 TE, 1 WR/RB 20 man rosters, 5 man taxi est. 2018
QB: Dak, Pickett
RB: CMC, Taylor, Gibson, Dillon, Akers, Penny
WR: JJeff, Hollywood, Olave, Toney, Aiyuk, Jeudy, C. Davis, Boyd, C. Samuel,
TE: Njoku, Gesicki
2023 picks: 1.x, 2.x, 3.x, 4.x
QB: Dak, Pickett
RB: CMC, Taylor, Gibson, Dillon, Akers, Penny
WR: JJeff, Hollywood, Olave, Toney, Aiyuk, Jeudy, C. Davis, Boyd, C. Samuel,
TE: Njoku, Gesicki
2023 picks: 1.x, 2.x, 3.x, 4.x
-
- Starter
- Posts: 675
- Joined: Thu Aug 16, 2018 10:40 am
Re: We're all sleeping on Michael Pittman
Planting my flag that Pittman will be a stud despite a low percentage profile.
12 Team Super-Flex TEP Dynasty
22 CHAMP
(2019 & 2020 Runner-Up)
1 QB, 3 RB, 4 WR, Flex, SF, 2 TE
3 DL, 3 LB, 3 DB, 1 Flex IDP
QB- Justin Herbert, Baker Mayfield, Deshaun Watson, Marcus Mariota, Gardner Minshew, Sam Ehlinger
RB- D’Andre Swift, Nick Chubb, Javonte Williams, Cam Akers, Kyren Williams, Jerome Ford, D’Ernest Johnson, Darrell Henderson
WR- DK Metcalf, DJ Moore, Chris Godwin, Jameson Williams, Gabe Davis, Hollywood Brown, Josh Reynolds, Calvin Austin, Laviska Shenault, Kyle Phillips
TE-Mark Andrews, Greg Dulcich, Isaiah Likely, Donald Parham, Tyler Conklin
DL-Danielle Hunter, Nick Bosa, Deforest Buckner, Jonathon Allen
LB- Micah Parsons, Jack Sanborn, Logan Wilson, Cole Holcomb, Nakobe Dean, Brian Asamoah
DB-Jeremy Chinn, Jalen Pitre, Jalen Thompson, Kam Curl, Darrick Forrest
K-Tyler Bass
22 CHAMP
(2019 & 2020 Runner-Up)
1 QB, 3 RB, 4 WR, Flex, SF, 2 TE
3 DL, 3 LB, 3 DB, 1 Flex IDP
QB- Justin Herbert, Baker Mayfield, Deshaun Watson, Marcus Mariota, Gardner Minshew, Sam Ehlinger
RB- D’Andre Swift, Nick Chubb, Javonte Williams, Cam Akers, Kyren Williams, Jerome Ford, D’Ernest Johnson, Darrell Henderson
WR- DK Metcalf, DJ Moore, Chris Godwin, Jameson Williams, Gabe Davis, Hollywood Brown, Josh Reynolds, Calvin Austin, Laviska Shenault, Kyle Phillips
TE-Mark Andrews, Greg Dulcich, Isaiah Likely, Donald Parham, Tyler Conklin
DL-Danielle Hunter, Nick Bosa, Deforest Buckner, Jonathon Allen
LB- Micah Parsons, Jack Sanborn, Logan Wilson, Cole Holcomb, Nakobe Dean, Brian Asamoah
DB-Jeremy Chinn, Jalen Pitre, Jalen Thompson, Kam Curl, Darrick Forrest
K-Tyler Bass
-
- All Pro
- Posts: 1720
- Joined: Sat Aug 22, 2015 8:16 pm
Re: We're all sleeping on Michael Pittman
Sorry if I messed up the quotes. Anyway, regarding the bold part - I imagine you mean this for a startup but for a rookie draft do you usually trade your picks or keep them?OhCruelestRanter wrote: ↑Mon Jun 29, 2020 11:13 amMost importantly, what you learn from doing this stuff is that the chances of these guys breaking out aren’t as different as we’d like to think, and frequently the answer is to just trade back and get more picks, and then use the late round picks to supplement your rosters with undervalued veterans.PR0v3 wrote: ↑Mon Jun 29, 2020 9:59 amHere’s my take away from reading that article:ericanadian wrote: ↑Mon Jun 29, 2020 4:05 am
Here is a good walkthrough.
https://thedraftnetwork.com/articles/20 ... tor-rating
The person took breakout age hit rates and then used those hit rates to give a % chance that Bryan Edwards or Tyler Johnson hit.
That’s not how stats work.
Is this really all that people are doing? Just counting 10 coin flips and taking whatever the results are to guess the next coin flip? Why are people placing so much value on this stuff?
-
- Combine Attendee
- Posts: 44
- Joined: Fri Jun 26, 2020 6:07 am
-
- Combine Attendee
- Posts: 44
- Joined: Fri Jun 26, 2020 6:07 am
Re: We're all sleeping on Michael Pittman
This is the single best article I've come across to explain breakout age, why it matters, and links to the "studies" in which they came to the conclusions they did.PR0v3 wrote: ↑Mon Jun 29, 2020 8:40 pmSo we’re dressing up in our statistician costume and playing make believe? Nothing backed by legitimate process? That’s cool if that’s the way it is, let’s just stop pretending like any of it actually matters.OhCruelestRanter wrote: ↑Mon Jun 29, 2020 8:27 pmYou want a peer reviewed study now? Like in the Journal of Dynasty Fantasy Football?PR0v3 wrote: ↑Mon Jun 29, 2020 8:22 pm
Bruh. I’m just asking for the damn study so I can read it. Where the hell is it? It’s like you haven’t read any post I’ve made. Where the hell is the damn study so I can read it for myself and not have to just take your word for it? I can find 100 million peer reviewed studies online about the damages of smoking, can’t find a single one about breakout age.
https://thedraftnetwork.com/articles/20 ... tor-rating
Re: We're all sleeping on Michael Pittman
The thing is, I don't feel like he has a low % profile. I just think him being drafted as WR6 this year is about right. In other years he could be a top 3 guyShankopotamus wrote: ↑Mon Jun 29, 2020 9:10 pm Planting my flag that Pittman will be a stud despite a low percentage profile.
-
- Legend
- Posts: 6519
- Joined: Mon Jul 02, 2012 12:13 pm
Re: We're all sleeping on Michael Pittman
The biggest criticism I have of the dominator rating/breakout age/early declare grouping as they are currently used is that every single one of them piggy back on draft position. The problem with this is that draft position is not a static dataset, nor is it an exact stand in for talent level. You’re building a correlation on a correlation and throwing around exact likelihoods of success.
At best, you’re identifying inefficiencies in the NFL talent evaluation process. The problem with banking on something like that is that the NFL decision making process is always developing as is their player development strategies. What didn’t work one year, may have been corrected in a future year. I don’t have the data on hand, but are we seeing teams draft less guys with a late breakout age? They may have weeded out some of the problems that were causing the inefficiency and then your model is worthless and its back to draft position. I believe Prov is looking for some sort of proof that these stats link directly to success and not just a higher success rate at any given draft position.
Even for what it offers on its own, I don’t like the illusions of precision. Breakout age uses the age of the guy at the beginning of a given season where he “breaks out”. How did we come to this as a conclusion? Why is it acceptable to project a full season for a guy that misses games (DJ Moore being the most recent example I can think of), but not carry over games between seasons? We’re basically admitting that a full season of data is not necessary, and yet this hasn’t taken the next logical step of taking blocks of say ten consecutive games (even across seasons) instead of single seasons. Taking Pittman as an example, his breakout age would drop almost a full year if you took the last five to seven games of his 2018 season and carried it into 2019. Cutoffs like this are arbitrary and really undermine the statistic.
At best, you’re identifying inefficiencies in the NFL talent evaluation process. The problem with banking on something like that is that the NFL decision making process is always developing as is their player development strategies. What didn’t work one year, may have been corrected in a future year. I don’t have the data on hand, but are we seeing teams draft less guys with a late breakout age? They may have weeded out some of the problems that were causing the inefficiency and then your model is worthless and its back to draft position. I believe Prov is looking for some sort of proof that these stats link directly to success and not just a higher success rate at any given draft position.
Even for what it offers on its own, I don’t like the illusions of precision. Breakout age uses the age of the guy at the beginning of a given season where he “breaks out”. How did we come to this as a conclusion? Why is it acceptable to project a full season for a guy that misses games (DJ Moore being the most recent example I can think of), but not carry over games between seasons? We’re basically admitting that a full season of data is not necessary, and yet this hasn’t taken the next logical step of taking blocks of say ten consecutive games (even across seasons) instead of single seasons. Taking Pittman as an example, his breakout age would drop almost a full year if you took the last five to seven games of his 2018 season and carried it into 2019. Cutoffs like this are arbitrary and really undermine the statistic.
All I Der Is Win - 16 Team IDP League (Pass TD 6pts)
QB - Stafford, Stroud, Tune
RB - Swift, Hall, Penny, Bigsby, Ford
WR - Pittman, Olave, Di. Johnson, G. Wilson, J. Williams, Metchie, Robinson, M. Wilson
TE - Okonkwo, Schoonmaker
LB - Brooks, R. Smith, Phillips
DL - Crosby, Allen, Simmons
DB - D. James, Baker, Delpit
K - Just a stupid kicker
QB - Stafford, Stroud, Tune
RB - Swift, Hall, Penny, Bigsby, Ford
WR - Pittman, Olave, Di. Johnson, G. Wilson, J. Williams, Metchie, Robinson, M. Wilson
TE - Okonkwo, Schoonmaker
LB - Brooks, R. Smith, Phillips
DL - Crosby, Allen, Simmons
DB - D. James, Baker, Delpit
K - Just a stupid kicker
-
- GOAT
- Posts: 14258
- Joined: Sat Sep 27, 2014 6:06 pm
Re: We're all sleeping on Michael Pittman
Good post. I understand the logic behind breakout and dominator, and they are useful stats for fantasy purposes. However, I think pigeonholing yourself into either gives you a really limited picture of things, even if you can still hit on prospects.ericanadian wrote: ↑Tue Jun 30, 2020 5:53 am The biggest criticism I have of the dominator rating/breakout age/early declare grouping as they are currently used is that every single one of them piggy back on draft position. The problem with this is that draft position is not a static dataset, nor is it an exact stand in for talent level. You’re building a correlation on a correlation and throwing around exact likelihoods of success.
At best, you’re identifying inefficiencies in the NFL talent evaluation process. The problem with banking on something like that is that the NFL decision making process is always developing as is their player development strategies. What didn’t work one year, may have been corrected in a future year. I don’t have the data on hand, but are we seeing teams draft less guys with a late breakout age? They may have weeded out some of the problems that were causing the inefficiency and then your model is worthless and its back to draft position. I believe Prov is looking for some sort of proof that these stats link directly to success and not just a higher success rate at any given draft position.
Even for what it offers on its own, I don’t like the illusions of precision. Breakout age uses the age of the guy at the beginning of a given season where he “breaks out”. How did we come to this as a conclusion? Why is it acceptable to project a full season for a guy that misses games (DJ Moore being the most recent example I can think of), but not carry over games between seasons? We’re basically admitting that a full season of data is not necessary, and yet this hasn’t taken the next logical step of taking blocks of say ten consecutive games (even across seasons) instead of single seasons. Taking Pittman as an example, his breakout age would drop almost a full year if you took the last five to seven games of his 2018 season and carried it into 2019. Cutoffs like this are arbitrary and really undermine the statistic.
Valuing players who produce at a high level at an early age is practical, but there's so much grey area before that.
Who is online
Users browsing this forum: Ahrefs [Bot] and 48 guests