Register a SA Forums Account here!
JOINING THE SA FORUMS WILL REMOVE THIS BIG AD, THE ANNOYING UNDERLINED ADS, AND STUPID INTERSTITIAL ADS!!!

You can: log in, read the tech support FAQ, or request your lost password. This dumb message (and those ads) will appear on every screen until you register! Get rid of this crap by registering your own SA Forums Account and joining roughly 150,000 Goons, for the one-time price of $9.95! We charge money because it costs us money per month for bills, and since we don't believe in showing ads to our users, we try to make the money back through forum registrations.
 
  • Locked thread
Forever_Peace
May 7, 2007

Shoe do do do do do do do
Shoe do do do do do do yeah
Shoe do do do do do do do
Shoe do do do do do do yeah

You rock.

Mind if I ask for some clarification on what exactly VAL is though? Is it the multiplier you'd have to apply to a replacement-level player in order to get to their projected output? A z-score compared to replacement-level player?

quote:

For all intents and purposes in our league Lacy is maybe worth a third round pick if you're lucky.

Hey Mr. "Lacy is a top 3 stud" I'm just asking you to put your money where your mouth is. :p

Adbot
ADBOT LOVES YOU

pmchem
Jan 22, 2010



What's the reasoning behind taking LBell as 1.1, missing 3 games? I mean, statistically? It's not like he's guaranteeing victory in the weeks he will play, compared to someone else you could take at 1.1.

It seems like he'd have to be much, much better than other 1.1 options in order to justify that choice.

Franks Happy Place
Mar 15, 2011

It is by weed alone I set my mind in motion. It is by the dank of Sapho that thoughts acquire speed, the lips acquire stains, stains become a warning. It is by weed alone I set my mind in motion.

pmchem posted:

What's the reasoning behind taking LBell as 1.1, missing 3 games? I mean, statistically? It's not like he's guaranteeing victory in the weeks he will play, compared to someone else you could take at 1.1.

It seems like he'd have to be much, much better than other 1.1 options in order to justify that choice.

The argument is based on a VORP calculation rather than a gross season points total. Basically the thinking goes, if you get a sufficient comparative positional advantage for 10 games by having Bell, the loss of his utility for three games is tolerable.

Leperflesh
May 17, 2007

Forever_Peace posted:

Alright, weeks ago I told somebody I would make a more sophisticated zero-rb strategy to plug into my draft simulator and run against a few thousand 2014 drafts.

I know it's a lot of work so I wouldn't blame you for not wanting to do it, but: I don't think you can draw any conclusions from any of this if you're only simulating how it worked out with a single year's drafts. We have no idea if, using this method, 2014 was typical or an outlier. If the data's available, I'd want to see it run against at least ten years of drafts, and even then I'd guess that it's highly possible that 2015 won't fit the average of the previous ten years.

Just as an example, in 2014 AP's suspension blew a gigantic hole in the top of the draft. We need to know if drafting a RB at 1.1 or 1.2 who will produce no points whatsoever is typical or very rare.

Another issue I see is that you're simply adding up total points for the year, and using that as a rank. But most fantasy formats don't start all of their drafted players every week. I'd like to see a deeper simulation, and then some accounting for the tendency to swap players in and out. Any player who got a season-ending injury should be assumed to get swapped for a free agent who would produce nonzero points in that slot the rest of the year, for example.

Are WRs more injury prone than RBs, or less? Either answer will skew the results.

Finally, I'd like to thank you again for all the work you're doing here. "Is 0RB a better strategy" is an interesting question to ask the data, but now that you've build the simulations to ask that question, you have the architecture to ask a lot of other interesting questions.

Forever_Peace
May 7, 2007

Shoe do do do do do do do
Shoe do do do do do do yeah
Shoe do do do do do do do
Shoe do do do do do do yeah

Franks Happy Place posted:

The argument is based on a VORP calculation rather than a gross season points total. Basically the thinking goes, if you get a sufficient comparative positional advantage for 10 games by having Bell, the loss of his utility for three games is tolerable.

Eh, I'm actually not convinced that losing a few games up front is so trivial. Those losses seem to impact playoff probability quite considerably. (source). I personally will not be taking Bell early this year. MAYBE end of the 1st but he's just not falling that far right now.

Forever_Peace
May 7, 2007

Shoe do do do do do do do
Shoe do do do do do do yeah
Shoe do do do do do do do
Shoe do do do do do do yeah

Leperflesh posted:

I know it's a lot of work so I wouldn't blame you for not wanting to do it, but: I don't think you can draw any conclusions from any of this if you're only simulating how it worked out with a single year's drafts. We have no idea if, using this method, 2014 was typical or an outlier. If the data's available, I'd want to see it run against at least ten years of drafts, and even then I'd guess that it's highly possible that 2015 won't fit the average of the previous ten years.

Just as an example, in 2014 AP's suspension blew a gigantic hole in the top of the draft. We need to know if drafting a RB at 1.1 or 1.2 who will produce no points whatsoever is typical or very rare.

Another issue I see is that you're simply adding up total points for the year, and using that as a rank. But most fantasy formats don't start all of their drafted players every week. I'd like to see a deeper simulation, and then some accounting for the tendency to swap players in and out. Any player who got a season-ending injury should be assumed to get swapped for a free agent who would produce nonzero points in that slot the rest of the year, for example.

Are WRs more injury prone than RBs, or less? Either answer will skew the results.

Finally, I'd like to thank you again for all the work you're doing here. "Is 0RB a better strategy" is an interesting question to ask the data, but now that you've build the simulations to ask that question, you have the architecture to ask a lot of other interesting questions.

You are completely 100% correct about all of this, of course. But I made the decision pretty early on that a reasonably "valid" in-season simulator was probably not possible for me to make on my own. Trades and WW competition in particular just depend too much on a psychology that is more individual-specific than anything else, and considering the psychology of "starts" is also pretty hard (do you play Hurns after his big week? Do you play Antone Smith after a few weeks of big games? Quick? Steve Smith? How many weeks to you keep starting Ball when he fails to perform? McCoy? Torrey Smith? These things seem like they might depend on structural differences between individuals - like risk-aversion - than random variance)

So I decided to confine my questions to "how to have a good draft", as defined by "drafting players that performed well" in an empirical sense. Can you really blame somebody for taking Moreno last year? Probably not. But by assuming that busts are essentially sampled for a normally distributed bust rate, we can still look at average differences between draft strategies.

The way my simulator handles injuries is pretty over-simplified: the "starters" are defined as the whichever players scored the most by the end of the season, so if an early-round pick busts (e.g. Ball), whichever late-going RB performs the best is declared the "starter" for measurement purposes.

I could probably measure outcomes using extrapolated performance for people who miss games (extending points per game over for that player over missed games), but I feel like that could unfairly benefit people who gamble on bellcow backs and slot receivers (who have a higher injury rate compared to other offensive skill players).

And finally, I don't have ten years of data (in particular, I need rankings, projections, AND final performance to run the simulator as intended), but I do have 2013. I should probably get around to that eventually!

sourdough
Apr 30, 2012

Spoeank posted:

Honestly I forgot about Antonio Brown when I put that together. I was shying away from Dez & DT for the #4 slot because we don't know at what point the QB situation becomes a dumpster fire in Dallas and Denver. I know DT produced with Tebow but he wasn't A1 top of the list with him.

Luck seemed a bit early but I think that stems from being influenced by a twitter argument I got into with an NFL.com guy who said Luck should be 1.01.

Also no PPR but 6 points per passing TD.

This actually changes things a decent bit with respect to Luck. 6 point passing TDs amplifies the advantage that Rodgers and Luck and other elite passing QBs have. To put it in perspective, just bumping up TD values from 4 to 6 gave Luck and Rodgers about as big a positional advantage as Gronk had last year, and just about everyone thinks Gronk is worth at least a mid 1st. Throw in the fact that QBs have such long careers, are more consistent, and are less likely to be injured, and I have no problem taking Luck with an early first in that format.

Also no PPR does bump Lacy up slightly too, though I still think 1.04 is a bit too early.

Leperflesh
May 17, 2007

Hmm.

In order to compare different drafted teams, maybe it would be useful to have a handful of broad categories, and just report how many players on each team fit those categories. Something like:
-Player massively underperformed despite being uninjured
-Player got injured early
-Player moderately underperformed or was injured late
-Player performed as expected
-Player moderately overperformed and was never injured
-Player greately overperformed

Where "underperformed" and "overperformed" are based on ADP vs. points earned over the season.

So if you then showed each team with how many players fit each category, you might have one team that went: "1 | 3 | 5 | 3 | 1 | 1" and another that went "2 | 0 | 7 | 2 | 2 | 1" and we could compare these and say hmm, the second team wound up marginally better, even if the two teams' points-totals at the end of the year wound up very close to each other.

Franks Happy Place
Mar 15, 2011

It is by weed alone I set my mind in motion. It is by the dank of Sapho that thoughts acquire speed, the lips acquire stains, stains become a warning. It is by weed alone I set my mind in motion.

Forever_Peace posted:

Eh, I'm actually not convinced that losing a few games up front is so trivial. Those losses seem to impact playoff probability quite considerably. (source). I personally will not be taking Bell early this year. MAYBE end of the 1st but he's just not falling that far right now.

But it's not an automatic loss of three games, it's just three games you have to plan a waiver strategy for, either by targeting other players who have soft matchups early in the season, or by going for Bell's handcuff in Pittsburgh in later rounds.

I'm not really sold on Bell as a 1.1 pick, mind you, but there is a logic to it. I think I'd rather have Lacy, but it really depends on whether the league is PPR or not.

pmchem
Jan 22, 2010


Franks Happy Place posted:

The argument is based on a VORP calculation rather than a gross season points total. Basically the thinking goes, if you get a sufficient comparative positional advantage for 10 games by having Bell, the loss of his utility for three games is tolerable.

I know the concept, but does anyone care to actually back that up with numbers for 9 games of Bell vs 12 games of JCharles, Lacy, ADP? Don't forget the bye week.

Forever_Peace
May 7, 2007

Shoe do do do do do do do
Shoe do do do do do do yeah
Shoe do do do do do do do
Shoe do do do do do do yeah

Leperflesh posted:

Hmm.

In order to compare different drafted teams, maybe it would be useful to have a handful of broad categories, and just report how many players on each team fit those categories. Something like:
-Player massively underperformed despite being uninjured
-Player got injured early
-Player moderately underperformed or was injured late
-Player performed as expected
-Player moderately overperformed and was never injured
-Player greately overperformed

Where "underperformed" and "overperformed" are based on ADP vs. points earned over the season.

So if you then showed each team with how many players fit each category, you might have one team that went: "1 | 3 | 5 | 3 | 1 | 1" and another that went "2 | 0 | 7 | 2 | 2 | 1" and we could compare these and say hmm, the second team wound up marginally better, even if the two teams' points-totals at the end of the year wound up very close to each other.

Yeah if that's what you want to see, it's totally doable. Incorporating stuff like injuries requires manual coding though, so you (and anybody else) are welcome to grab the 2014 projections (used to influence value-based draft strategies) and actual stats (used to calculate performance) and add whatever columns of data they think would be interesting (reuploading a draft yourself so I don't have to make it publicly editable).

It sounds like you're sort of trending towards a Likert scale of pick value here: I think the most illustrative outcome here might be one that can be quantitatively averaged (e.g. where your ranking scale as equal-sized "steps"). But If you want to make qualitative categories I can deal with that too in the manner you suggest.

Just think carefully about what question you want to address and what you want the data to tell you. If it's reasonable to code I will probably get to it eventually. =)

Forever_Peace
May 7, 2007

Shoe do do do do do do do
Shoe do do do do do do yeah
Shoe do do do do do do do
Shoe do do do do do do yeah

Franks Happy Place posted:

But it's not an automatic loss of three games, it's just three games you have to plan a waiver strategy for, either by targeting other players who have soft matchups early in the season, or by going for Bell's handcuff in Pittsburgh in later rounds.

I'm not really sold on Bell as a 1.1 pick, mind you, but there is a logic to it. I think I'd rather have Lacy, but it really depends on whether the league is PPR or not.

Yeah I understand the strategy here, but rolling a late-round RB out against everybody else's studs is a pretty distinct disadvantage, and could also have season-long ramifications if it eats up your early WW targets and available bench spots. It's a hell of a gamble, and one that I would be willing to take for the right price, but I think people might be chasing the upside enough to inflate his value beyond where it's sensible to.

Franks Happy Place
Mar 15, 2011

It is by weed alone I set my mind in motion. It is by the dank of Sapho that thoughts acquire speed, the lips acquire stains, stains become a warning. It is by weed alone I set my mind in motion.

pmchem posted:

I know the concept, but does anyone care to actually back that up with numbers for 9 games of Bell vs 12 games of JCharles, Lacy, ADP? Don't forget the bye week.

The calculation isn't 100% rearward-looking statistical, though, since it takes into account a lot of other things:

1) Age of player (likelihood to decline for age-based reasons)
2) Quality of team (aka quality of touches for the RB; number of RZ opportunities)
3) Team composition/competition (it it RBBC? is there a young up and comer pushing for touches?)
4) Team tactics (PPR rewards RBs who play four downs, some teams run it in the RZ more, etc.)
4) Consistency of track record for player, both in college and more importantly in NFL

So if you tally all of that up, the "surest bet" for a round 1 RB is a young, gifted RB in a good offense that features him in a heavy rotation, with lots of RZ chances. Which means the actual "safe" bet over guys like Peterson, Charles, ADP, Foster etc. are actually guys like Eddie Lacy, Jeremy Hill, and (suspension notwithstanding) Le'Veon Bell. They check all of the best boxes, or at least most of them.

Take Charles vs. Hill for example- they are both in a timershare, except Charles' partner is younger and considered quite talented, whereas Gio Bernard, while talented, is a) older than Hill and b) has already sort of whiffed on his starter role, so his upside is capped insofar as being a threat to Hill's role. So if I'm honestly weighting those two guys in my head, looking at team situation, coaching, age, upside, all of that, I have to say I honestly like Hill more than Charles. I like Lacy even more than Hill, frankly, because there's nearly zero threat to his touches, unlike the credible-if-capped Gio Bernard.

Which brings us to Bell: he is young, undeniably talented, in an offense that suits him, with no competition. If you have decided (for whatever reason) that you can tough out those first three weeks, he is probably a "safer" pick than some of the earlier first round RBs in a lot of ways. Honestly his health is a bigger concern to me right now than his suspension. I want to see some good carries in the pre-season before I decide to do the math on where I'd draft him vs. the cost of missing him a few games.

Beer4TheBeerGod
Aug 23, 2004
Exciting Lemon

Forever_Peace posted:

You rock.

Mind if I ask for some clarification on what exactly VAL is though? Is it the multiplier you'd have to apply to a replacement-level player in order to get to their projected output? A z-score compared to replacement-level player?

Hey Mr. "Lacy is a top 3 stud" I'm just asking you to put your money where your mouth is. :p

It's the projected score of a player (per week) relative to a baseline (the nth player in parenthesis after each position in the title block). The baseline in this case is the historical number of players needed to fill that position when you take into account injuries and bye weeks.

Benne
Sep 2, 2011

STOP DOING HEROIN

Franks Happy Place posted:

But it's not an automatic loss of three games, it's just three games you have to plan a waiver strategy for, either by targeting other players who have soft matchups early in the season, or by going for Bell's handcuff in Pittsburgh in later rounds.

I'm not really sold on Bell as a 1.1 pick, mind you, but there is a logic to it. I think I'd rather have Lacy, but it really depends on whether the league is PPR or not.

The problem with "use Bell's handcuff" is that DeAngelo Williams loving sucks

Franks Happy Place
Mar 15, 2011

It is by weed alone I set my mind in motion. It is by the dank of Sapho that thoughts acquire speed, the lips acquire stains, stains become a warning. It is by weed alone I set my mind in motion.

Benne posted:

The problem with "use Bell's handcuff" is that DeAngelo Williams loving sucks

True, but you just need replacement-level performance out of him for the math to balance out enough to make this strategy arguably viable.

Again, I'm not sure I buy this particular scenario, but the underlying logic is essentially the same as why Rob Gronkowski is a first round draft pick. Positional advantage is real.

pmchem
Jan 22, 2010


Franks Happy Place posted:

If you have decided (for whatever reason) that you can tough out those first three weeks, he is probably a "safer" pick than some of the earlier first round RBs in a lot of ways.

The issue I was asking about is how you can decide, numerically, that you can "tough out" the suspension weeks vs. just draft JCharles, ADP, or Lacy instead. You have to be making a specific decision that Bell is a certain amount more valuable in weeks 4-13 than one of those guys (not just any replacement player). That, he will add X probability over your second-favorite RB each of the weeks 4-13 so that the replacement-level loss in weeks 1-3 compared to your second-favorite RB still leads to +<EV> in terms of total games won over weeks 1-13. That means he must be _hugely_ better in +EV than your second-favorite RB, since weeks 1-3 you will be at a major disadvantage. In fact, if you get "9 games with, 3 games without" in weeks 1-13 (1 week bye, doesn't count) then he must provide an increased win probability of at least 33% of the [ win probability decrease of your replacement-level RB compared to second-best RB ].

It's a large enough difference that I doubt he can be justified as first RB off the board. You will need lucky wins in week 1-3 to be an overall winner with that strategy. But if you make the playoffs, you'd be +EV (assuming you have him ranked #1).

Spoeank
Jul 16, 2003

That's a nice set of 11 dynasty points there, it would be a shame if 3 rings were to happen with it
If only 4 teams in your league make the playoffs in a 12-team league Le'Veon Bell is suicide but those cut throat settings might just be my league.

Franks Happy Place
Mar 15, 2011

It is by weed alone I set my mind in motion. It is by the dank of Sapho that thoughts acquire speed, the lips acquire stains, stains become a warning. It is by weed alone I set my mind in motion.

pmchem posted:

The issue I was asking about is how you can decide, numerically, that you can "tough out" the suspension weeks vs. just draft JCharles, ADP, or Lacy instead.

Well for starters, I'd definitely take Lacy and probably JHill before either Bell or Charles, but I have weird opinions on RBs as most here would tell you. :tipshat:

As for the thinking behind how you'd tough out Bell's suspension, I assume it's just a Zero RB type strategy. Something stupid always happens in the first few weeks, there will be plenty of opportunity.

Spoeank
Jul 16, 2003

That's a nice set of 11 dynasty points there, it would be a shame if 3 rings were to happen with it
On a side note why am I finding Mike Evans on so many top ten lists and so many "HE SHOULD GET SO MUCH BETTER" lists. McCown/Glennon to Winston or not, outside of victimizing two of the worst secondaries in the league during a 14 day stretch of awesomeness, he was mostly good not great.


I don't get it but maybe I'm missing something.

Metapod
Mar 18, 2012

pmchem posted:

The issue I was asking about is how you can decide, numerically, that you can "tough out" the suspension weeks vs. just draft JCharles, ADP, or Lacy instead. You have to be making a specific decision that Bell is a certain amount more valuable in weeks 4-13 than one of those guys (not just any replacement player). That, he will add X probability over your second-favorite RB each of the weeks 4-13 so that the replacement-level loss in weeks 1-3 compared to your second-favorite RB still leads to +<EV> in terms of total games won over weeks 1-13. That means he must be _hugely_ better in +EV than your second-favorite RB, since weeks 1-3 you will be at a major disadvantage. In fact, if you get "9 games with, 3 games without" in weeks 1-13 (1 week bye, doesn't count) then he must provide an increased win probability of at least 33% of the [ win probability decrease of your replacement-level RB compared to second-best RB ].

It's a large enough difference that I doubt he can be justified as first RB off the board. You will need lucky wins in week 1-3 to be an overall winner with that strategy. But if you make the playoffs, you'd be +EV (assuming you have him ranked #1).

The problem is you are thinking with your brain and not your balls

VDay
Jul 2, 2003

I'm Pacman Jones!

Spoeank posted:

On a side note why am I finding Mike Evans on so many top ten lists and so many "HE SHOULD GET SO MUCH BETTER" lists. McCown/Glennon to Winston or not, outside of victimizing two of the worst secondaries in the league during a 14 day stretch of awesomeness, he was mostly good not great.


I don't get it but maybe I'm missing something.

He finished as like a top 12 WR in standard scoring and had 10+ points in something like 8 games. As a rookie with two bad QBs. Winston isn't much of an upgrade, but he should still be a bit better. If nothing else he'll still get a shitload of "gently caress it going deep" throws his way, where he's an absolute monster. If OBJ didn't have a record-setting year Mike Evans would've gotten a ton more coverage. Dude absolutely carried some teams to the playoffs with his weeks 6-12 performance.

Forever_Peace
May 7, 2007

Shoe do do do do do do do
Shoe do do do do do do yeah
Shoe do do do do do do do
Shoe do do do do do do yeah

Spoeank posted:

On a side note why am I finding Mike Evans on so many top ten lists and so many "HE SHOULD GET SO MUCH BETTER" lists. McCown/Glennon to Winston or not, outside of victimizing two of the worst secondaries in the league during a 14 day stretch of awesomeness, he was mostly good not great.


I don't get it but maybe I'm missing something.

I went searching for data on an effortpost about why Mike Evans is a stud, but came away convinced that he doesn't belong in the top 10 WR for redraft. Like yeah historically receivers have a better second year than rookie year but I'm not sure we can project the same jump for the 2014 class that accounted for such a historic proportion of passing offenses.

Spoeank
Jul 16, 2003

That's a nice set of 11 dynasty points there, it would be a shame if 3 rings were to happen with it

Forever_Peace posted:

I went searching for data on an effortpost about why Mike Evans is a stud, but came away convinced that he doesn't belong in the top 10 WR for redraft. Like yeah historically receivers have a better second year than rookie year but I'm not sure we can project the same jump for the 2014 class that accounted for such a historic proportion of passing offenses.

:hfive:

VDay posted:

He finished as like a top 12 WR in standard scoring and had 10+ points in something like 8 games. As a rookie with two bad QBs. Winston isn't much of an upgrade, but he should still be a bit better. If nothing else he'll still get a shitload of "gently caress it going deep" throws his way, where he's an absolute monster. If OBJ didn't have a record-setting year Mike Evans would've gotten a ton more coverage. Dude absolutely carried some teams to the playoffs with his weeks 6-12 performance.

His gently caress it going deep-based performance plus touchdown regression means top 15 or top 20, not top 10, to me.


Also watch the highlights from the Washington game. Holy poo poo that pass defense was pathetic (also they had Perry Riley guard him on a long TD)

sourdough
Apr 30, 2012

Forever_Peace posted:

I went searching for data on an effortpost about why Mike Evans is a stud, but came away convinced that he doesn't belong in the top 10 WR for redraft. Like yeah historically receivers have a better second year than rookie year but I'm not sure we can project the same jump for the 2014 class that accounted for such a historic proportion of passing offenses.

Isn't it the exact opposite? Everything I've seen lately has been looking at the reality of the "sophomore slump" in the past 10 or so years, and came away with the conclusion that sophomore WRs definitely do worse than in their rookie year.

Here's some analysis from 2013, and here's the table I copied that summarizes it:



That is, rookie WRs that were fantasy relevant almost universally fell in the rankings their sophomore year, plus those that finished as a top 24 WR as a rookie regressed in points per game by 21%. This is all rookie receivers that finished in the top 48 from 2000 to 2011, by the way. It's definitely a small sample size, with a few outliers, but it agrees with another analysis I saw elsewhere and can't find now :v:

Do you have a link for some analysis showing that historically receivers actually improve their sophomore year?

Edit: I found this claiming only 35% of sophomore WRs that caught 30+ passes their rookie year regress. But the few example players they show don't match the numbers from the other article, likely because the first one I posted looked at regression in points per game and the second looked at regression in raw points scored. The former seems better to me, but it's debatable. But to be fair, the WR2 group in the first link, excluding Royal, did alright!

sourdough fucked around with this message at 04:28 on Jul 27, 2015

VDay
Jul 2, 2003

I'm Pacman Jones!

Spoeank posted:

His gently caress it going deep-based performance plus touchdown regression means top 15 or top 20, not top 10, to me.

I thought he was going in the WR12-15 range which seems fine to me. If he's going as a top 10 guy then yeah I'd say that's being a bit optimistic that Winston won't bomb. Either way I was mostly trying to debunk the notion that he had a Doug Martin-esque season where he got like 75% of his production in two games. Instead he managed to supplant VJax as the go-to receiver on that team (granted, not the hardest job to win in the world), was a super consistent red zone/TD threat (compared to someone like Martavis Bryant who really slowed down after his crazy first four games), and did it all while on a horrible team with two bad QBs.

This year teams will probably focus on him more, but he'll still be a huge red zone threat and will still be on a bad Bucs team that'll probably be constantly behind in games. I'd gladly take Evans as my WR2, unless I'm picking late 1st/early 2nd and a tier 1 guy is still available.

Spoeank
Jul 16, 2003

That's a nice set of 11 dynasty points there, it would be a shame if 3 rings were to happen with it

VDay posted:

I thought he was going in the WR12-15 range which seems fine to me. If he's going as a top 10 guy then yeah I'd say that's being a bit optimistic that Winston won't bomb. Either way I was mostly trying to debunk the notion that he had a Doug Martin-esque season where he got like 75% of his production in two games. Instead he managed to supplant VJax as the go-to receiver on that team (granted, not the hardest job to win in the world), was a super consistent red zone/TD threat (compared to someone like Martavis Bryant who really slowed down after his crazy first four games), and did it all while on a horrible team with two bad QBs.

This year teams will probably focus on him more, but he'll still be a huge red zone threat and will still be on a bad Bucs team that'll probably be constantly behind in games. I'd gladly take Evans as my WR2, unless I'm picking late 1st/early 2nd and a tier 1 guy is still available.

A ton of experts (I know) have him pegged as top ten. His FantasyPros consensus is 11. This is what I was scratching my head against.

Forever_Peace
May 7, 2007

Shoe do do do do do do do
Shoe do do do do do do yeah
Shoe do do do do do do do
Shoe do do do do do do yeah

RVProfootballer posted:

Isn't it the exact opposite? Everything I've seen lately has been looking at the reality of the "sophomore slump" in the past 10 or so years, and came away with the conclusion that sophomore WRs definitely do worse than in their rookie year.

If you impose a selection filter to only get the top players every year (e.g. "rookies in the top 20"), you're going to find a good deal of regression regardless of the class or position of the players you're looking at. The regression rate for the previous year's RB1 is considerable, for example. It just so happens that he's usually the best guess we have at who will perform best the following year.

But contribution-per-age curves suggest that WRs increase their contributions until 25, then peak until 27 (source), and aging curves tell a similar story (source). While not the same thing as class, it's pretty related (and for some reason I can't find class curves at the moment).

No Butt Stuff
Jun 10, 2004

I just joined a league that's converting (slowly) to a dynasty league with salary cap.

The cap is $60 this year and goes up 2 bucks every year.

I inherited a team bereft of talent (well, not bad considering it's a 16 team league.)

This year, I get to keep Gronk and DeMarco for 10 bucks a piece.

Which means I have 40 bucks to fill out a roster, like pretty much everyone else. I thought this might not be great, and from a keeper stance it's really not, especially since those prices will go up a buck every year that I keep them, but I think that they're worth 16% of my money or whatever each. DeMarco is the one that really worries me, so I may try to flip him in a trade after the draft.

And of course we're doing an offline draft, so I need to find some beersheets and translate them to 60 dollars, then cross out everyone's keepers when I get the list this week.

poly and open-minded
Nov 22, 2006

In BOD we trust

n: had my first draft of the year yesterday and keep staring at my lineup
v: when does the season start?!

sourdough
Apr 30, 2012

Forever_Peace posted:

But contribution-per-age curves suggest that WRs increase their contributions until 25, then peak until 27 (source), and aging curves tell a similar story (source). While not the same thing as class, it's pretty related (and for some reason I can't find class curves at the moment).

Yeah, I think I can concoct a scenario where those age curves are true and there is still a sophomore slump. Rookies enter at age 20, 21, or 22. Assume younger rookies tend to do more poorly their first year, older rookies tend to do better their first year, and everyone experiences a sophomore slump that is some 20% regression or whatever. Compared to age 20 as a baseline, age 21 will mix in 21 year old sophomores doing worse and 21 year old rookies that will do better, so the net effect could be an increase in performance of 21 year olds over 20 year olds. Anyway, thanks for looking these up :) I'll keep an eye out for anything else more definitive. I just think it's a really interesting question whether probability of sophomores regression is any larger than probability of regression in general, especially this year since so many of the 2014 rookie WRs did so well.

Leperflesh
May 17, 2007

RVProfootballer posted:

Yeah, I think I can concoct a scenario where those age curves are true and there is still a sophomore slump. Rookies enter at age 20, 21, or 22. Assume younger rookies tend to do more poorly their first year, older rookies tend to do better their first year, and everyone experiences a sophomore slump that is some 20% regression or whatever. Compared to age 20 as a baseline, age 21 will mix in 21 year old sophomores doing worse and 21 year old rookies that will do better, so the net effect could be an increase in performance of 21 year olds over 20 year olds. Anyway, thanks for looking these up :) I'll keep an eye out for anything else more definitive. I just think it's a really interesting question whether probability of sophomores regression is any larger than probability of regression in general, especially this year since so many of the 2014 rookie WRs did so well.

Don't forget to account for survivor bias. Every rookie who totally washes out after one or two years won't be contributing to the stats for rookies that actually last three+ years.

Doltos
Dec 28, 2005

🤌🤌🤌
I need a name for my team that involves JPP's finger but I'm coming up blank.

VDay
Jul 2, 2003

I'm Pacman Jones!
One in the pink, one in the stink

Leperflesh
May 17, 2007

Doltos posted:

I need a name for my team that involves JPP's finger but I'm coming up blank.

The New York Firecrackers?

Metapod
Mar 18, 2012

Doltos posted:

I need a name for my team that involves JPP's finger but I'm coming up blank.

1 down 9 to go

Varg
Jan 13, 2007

A friendly face.

JPP's Fantasy Fingers... obviously

Forever_Peace
May 7, 2007

Shoe do do do do do do do
Shoe do do do do do do yeah
Shoe do do do do do do do
Shoe do do do do do do yeah

Doltos posted:

I need a name for my team that involves JPP's finger but I'm coming up blank.

"Hard Counting"


In other "Kenny Stills is criminally underrated" news, 100% (literally all) of the rankings on fantasypros consider him to be better than his ADP (of wr70).

I really just don't understand why Stills just keeps floating under the radar.

Amergin
Jan 29, 2013

THE SOUND A WET FART MAKES

Forever_Peace posted:

In other "Kenny Stills is criminally underrated" news, 100% (literally all) of the rankings on fantasypros consider him to be better than his ADP (of wr70).

I really just don't understand why Stills just keeps floating under the radar.

In my mind although he certainly has the talent, I've been a bit fearful of Miami WRs for the same reason I'm fearful of Indy WRs - I have no clue which ones to take/which ones will get the work.

I have a feeling if he shows up in the preseason (or other WRs flub their chances) his ADP will rise. I'll definitely try to snatch him late in more casual leagues but in competitive leagues he just seems like a gamble to me. :shrug:

Adbot
ADBOT LOVES YOU

Sataere
Jul 20, 2005


Step 1: Start fight
Step 2: Attack straw man
Step 3: REPEAT

Do not engage with me



So we just got our keeper rosters. We completely redid our format last year and started from scratch. It is a $300 salary cap, with salaries going up 10% the first three years. 20% the fourth year, 50% the fifth year and 100% every year forward. I was the only person in the league to realize that by letting everyone spend their money first, I could just get young rookies for under $10 and completely dominate for years going forward. Behold my keepers for next year. To keep players from always staying $1, a 10% increase on $1 is $2.

Aaron Rodger -- $55 (Why not keep him, when you look at what I have)
Jeremy Hill - $2
Kelvin Benjamin -- $15
Jordan Matthews -- $2
Coby Fleener -- $2
Sheldon Richardson -- $2
Rashad Johnson -- $2
Browns DST - $3
Adam Vinatieri -- $2
Carlos Hyde - $3
Christine Michael - $3
Eric Ebron - $3
Paul Worrilow - $2
Carson Palmer - $10 (On the fence about keeping)
Allan Robinson -- $5
Tyler Eifert - $5
Teddy Bridgewater - $2
Derek Carr - $3
Kyle Rudolph - $3


I have $175 of my $300 cap to spend and only 10 spots to fill. We start 1 QB/2 RB/3WR/TE/K/DST/DL/D/DB.

Only a handful of my players have been in the league more than three years and I control their souls for life. I might rent a stud or two, but I can't wait to load up on all the rookie running backs and wide receivers from this draft. Everyone else is in cap hell and has to make drastic cuts, so good players will be available right away. I know McCoy is getting cut and I already cut Megatron, so I'm debating which one I'll target. It might not even be worth targeting either. If I snag two of the top rookie RBs and WRs, I could be in a really interesting position for the next ten years.

  • Locked thread