FBG Rankings - Discussion (1 Viewer)

[Marauder]

Wimer's draft rankings finished 130th out of the 132 submitted last year. You'd have done almost as well flipping a coin to make decisions as you would have using his rankings.

And 89 out of 91 in 2012.

Honestly, Is there any good reason why Wimer is still on staff? What does he contribute to this site?

This assumes that the whole point of having all those individual rankings is to maximize accuracy. I'm rather thankful that it's not, because if it was, Dodds would just fire us all and submit one set of rankings; his own. Seriously, Dodds has one of the best track records on rankings in the entire industry. I submit rankings based on what I believe will come true, but I still recognize that Dodds' rankings are in all likelihood a better representation of what will actually happen. If you really wanted to maximize your accuracy, you should probably exclude my own rankings, and everyone else's too, and just go with what Dodds says.

The purpose of having 20 different staff rankings is to increase the number of viewpoints that are represented. Dodds alone might be more accurate, but having everyone else in gives you a better idea of what other people are thinking and that information is actionable in its own right. There is value in a diversity of viewpoints. And Mark's viewpoint is quite extreme- I'm sure he'd be the first to own up to that- but extreme viewpoints exist, and by excluding them from the discussion, by refusing to give them a seat at the table, we're essentially denying their existence. We're marginalizing the very idea of heterogeneity and boiling the entire endeavor down to a series of "best practices" revolving around a narrowly-defined range of acceptable beliefs. We're setting ourselves up as the gatekeepers of orthodoxy and labeling the dissenters as heretics.

Are Mark Wimer's rankings accurate? Historically, no. Would I ever draft by them? No chance, although Wimer himself does with some degree of success. Does any of this mean that Wimer does not add any value to the fantasy discussion? I don't think it does. And yeah, this is the reason the "exclude" button exists. You don't have to like Wimer's rankings. You don't even have to use them. It probably doesn't hurt to at least see them and consider them, though.

I'm all for diversity of opinion, that's one of the main reasons I subscribe to FBG. Waldman probably thinks outside the box as much as anybody but he backs his opinions up with well thought out analysis. I can't remember ever seeing an article by Wimer that gives well thought out reasoning for his strategies. Can you point me to some?

It's not a matter of refusing anybody a seat at the table. It is a matter of determining who to include at the limited seats at the table. How should the owners of this site determine who gets those seats? Presumably it should be either someone who has demonstrated a history of success or has consistently provided well thought out and well written analysis. Do you disagree with this? Please point me to what Wimer has done to deserve one of those seats?

 

[Anarchy99]

As for his comments that he plans to play every game this season, what was his plan in prior years

Something less than 16 last year -- which is kind of the context I read into his comment.

I have a slightly different take on things, as I think this was partly a preemptive thing that should he not be in the lineup Week 1, it was a team thing and not Gron backing out. This way he looks good to the media and the fans, and if he was still a little iffy at the start of the year, the media would have to go to the brass and not him for an answer.

Hopefully is he good enough to go a full season, he stays healthy, and he can be close to 100%.

 

[Adam Harstad]

Someone should do a draft using his rankings and see how the team turns out.

Someone does. Mark Wimer himself. He did a lot of mocks last year that were posted to the site, if you want to look back on them. This one was the first (and biggest) mock we put out last year, and Wimer actually did quite solidly, for whatever that is worth (and I recognize that, with a sample size of 1, it's not worth that much). His RB-phobic ways resulted in Demarco Murray and Eddie Lacy as his starters. Calvin and Welker anchored his WR corps, and the Roddy injury hurt him doubly (with so much draft capital invested in both White and Matt Ryan), and in his commentary he focused on wanting to corner the market on Green Bay RBs and Cincy WR2s, both of which proved to be very strong calls.

A brief look also turned up this mock and this mock. I'll leave the analysis up to the reader.

Now I'm more confused than ever. He took Cameron as the 15th TE off the board last year, and now has him ranked 20th?

Is the takeaway from this actually supposed to be that he was more valuable as a potential breakout guy with an incredibly low floor higher than it is now as a guy who's actually broken out and has a pretty high floor? I don't get it.

I wonder how much of Wimer's rankings represent how willing he is to buy at the player's current ADP. That would certainly explain the drop, as well as Gronk's "off the board entirely".

Part of me does think it would be good if there was a place for people to discuss their ranking philosophy. Last year we did a pair of articles on rookie ranking philosophy and dynasty ranking philosophy, but I don't think we did anything for redraft. As my disagreement with JWB demonstrates, there's quite a lot of room for variation there even within the relatively stricter confines of redraft rankings. Maybe I'll pitch it as an article idea.

 

[FreeBaGeL]

Someone should do a draft using his rankings and see how the team turns out.

Someone does. Mark Wimer himself. He did a lot of mocks last year that were posted to the site, if you want to look back on them. This one was the first (and biggest) mock we put out last year, and Wimer actually did quite solidly, for whatever that is worth (and I recognize that, with a sample size of 1, it's not worth that much). His RB-phobic ways resulted in Demarco Murray and Eddie Lacy as his starters. Calvin and Welker anchored his WR corps, and the Roddy injury hurt him doubly (with so much draft capital invested in both White and Matt Ryan), and in his commentary he focused on wanting to corner the market on Green Bay RBs and Cincy WR2s, both of which proved to be very strong calls.

A brief look also turned up this mock and this mock. I'll leave the analysis up to the reader.

Now I'm more confused than ever. He took Cameron as the 15th TE off the board last year, and now has him ranked 20th?

Is the takeaway from this actually supposed to be that he was more valuable as a potential breakout guy with an incredibly low floor higher than it is now as a guy who's actually broken out and has a pretty high floor? I don't get it.

I wonder how much of Wimer's rankings represent how willing he is to buy at the player's current ADP. That would certainly explain the drop, as well as Gronk's "off the board entirely".

The thought had crossed my mind, but it seems like that would be pretty bad form, no?

"Meh, there's no way I'm going to get Cameron at TE6 so I may as well just rank him TE20 to make sure he doesn't even show up in my data dominator. Who cares if all those people paying money to see my rankings want to see where I would actually rank him".

I highly doubt that's the case.

 

[Adam Harstad]

I'm all for diversity of opinion, that's one of the main reasons I subscribe to FBG. Waldman probably thinks outside the box as much as anybody but he backs his opinions up with well thought out analysis. I can't remember ever seeing an article by Wimer that gives well thought out reasoning for his strategies. Can you point me to some?


It's not a matter of refusing anybody a seat at the table. It is a matter of determining who to include at the limited seats at the table. How should the owners of this site determine who gets those seats? Presumably it should be either someone who has demonstrated a history of success or has consistently provided well thought out and well written analysis. Do you disagree with this? Please point me to what Wimer has done to deserve one of those seats?

After rankings, article writing is probably the most visible thing the staffers do around here. Wimer doesn't do a lot of it (though he does do some. See here, and here), so he often seems a bit inconspicuous outside of his rankings (which, it should be noted, are consistently some of the most heavily-commented and annotated among the entire staff).

There's a lot of other stuff that goes on behind the scenes, though. Wimer writes training camp reports for a truly prodigious number of teams (6 of them, which is 50% more than anyone else on staff), so if you've enjoyed that feature in the past, odds are great that you were benefiting heavily from Wimer's insights and analysis. I do camp reports on just one team (Denver), and that alone is so much work I couldn't begin to comprehend handling a fifth of the league. He's also in charge of the extensive injury reporting for both offense and defense, (though he leaves the analysis to our own Dr. Jene Bramel), and I know he handles some responsibilities on the business end of running the FBGs empire.

Wimer is far more worthy of his seat at the table than I am, of this I have absolutely no doubt whatsoever- though everyone who makes staff is deserving and none of them should have to feel compelled to justify their inclusion.

 

[Stinkin Ref]

Someone should do a draft using his rankings and see how the team turns out.

Someone does. Mark Wimer himself. He did a lot of mocks last year that were posted to the site, if you want to look back on them. This one was the first (and biggest) mock we put out last year, and Wimer actually did quite solidly, for whatever that is worth (and I recognize that, with a sample size of 1, it's not worth that much). His RB-phobic ways resulted in Demarco Murray and Eddie Lacy as his starters. Calvin and Welker anchored his WR corps, and the Roddy injury hurt him doubly (with so much draft capital invested in both White and Matt Ryan), and in his commentary he focused on wanting to corner the market on Green Bay RBs and Cincy WR2s, both of which proved to be very strong calls.

A brief look also turned up this mock and this mock. I'll leave the analysis up to the reader.

Now I'm more confused than ever. He took Cameron as the 15th TE off the board last year, and now has him ranked 20th?

Is the takeaway from this actually supposed to be that he was more valuable as a potential breakout guy with an incredibly low floor higher than it is now as a guy who's actually broken out and has a pretty high floor? I don't get it.

I wonder how much of Wimer's rankings represent how willing he is to buy at the player's current ADP. That would certainly explain the drop, as well as Gronk's "off the board entirely".

Part of me does think it would be good if there was a place for people to discuss their ranking philosophy. Last year we did a pair of articles on rookie ranking philosophy and dynasty ranking philosophy, but I don't think we did anything for redraft. As my disagreement with JWB demonstrates, there's quite a lot of room for variation there even within the relatively stricter confines of redraft rankings. Maybe I'll pitch it as an article idea.

correct me if I am wrong, but for this tool.....shouldn't you all be using the same mindset/representation/rubric....?

if one guy is straight up ranking on his projections and another guy is using the "I wouldn't take him at this current ADP" mentality when doing his rankings, that skews things....no?.....

 

[Adam Harstad]

The thought had crossed my mind, but it seems like that would be pretty bad form, no?

"Meh, there's no way I'm going to get Cameron at TE6 so I may as well just rank him TE20 to make sure he doesn't even show up in my data dominator. Who cares if all those people paying money to see my rankings want to see where I would actually rank him".

I highly doubt that's the case.

I would think that ranking on only a single likely outcome and ignoring all other possibilities altogether would be pretty bad form, too, but JWB is one of the guys I respect most around here and he obviously disagrees (as does Yudkin, who I also have an abundant amount of respect for). FBGs is a pretty big tent, and it seems there's plenty of room for different approaches.

As an aside, I see ranking by value at current ADP not as "I just don't want him showing up in my Draft Dominator" and more as "I want to sort the league into potential targets and guys to ignore, and then draft my targets at an appropriate time as indicated by ADP". Which, again, is not the path I'd take, but I don't see why it can't be a valid approach.

 

[Stinkin Ref]

my thoughts are that all things being considered (meaning not worrying about bye weeks or the makeup of the rest of your team, etc) that rankings should mean "this is the next TE you should take no matter what".....

 

[cheese]

I disagree. It's the same disagreement I had back then. Perhaps everyone else thinks differently about it.

To me, rankings shouldn't be about what we *know*. They should be about what we expect at the time of the ranking. If you expect Gordon not to play, don't rank him. If you expect him to play, rank him based on how you value the production you expect from him.

My expectation is that rankings based on Expected Value should handily outperform rankings based on Expectations based on EoY accuracy. So if I'm ranking based on my expectations, then isn't that paradoxically an argument that I shouldn't be ranking based on expectations, (assuming the point of my individual rankings is to provide actionable info and maximize accuracy)?

Seriously, all of the heavy-hitters in the fantasy industry rank based on EV and not singular "most likely" expectations. It seems that battle has already been fought, and the hedgers have won rather decisively.

Sure, but I think we can agree that those within the fantasy industry that fought some of these battles stand to benefit from some of the vague difference in goals that you refer to. I like any ranker who will stick his neck out, make a suggestion, and own it. There are too many guys already (not at all specific to FBG) that will rank a Foster 4th and Lacy 10th and then suggest "well I didn't actually mean pick Foster ahead of Lacy the rankings are meant to ..... etc. This isn't useful. Put out rankings you believe in. Back them up with whatever drove your opinions when challenged. Be right on some, wrong on some. See where you hit and where you missed at the end of the year and own it. That's how I think staff should approach it if the ultimate goal is to provide the most useful preseason info for the players.

 

[Adam Harstad]

correct me if I am wrong, but for this tool.....shouldn't you all be using the same mindset/representation/rubric....?

if one guy is straight up ranking on his projections and another guy is using the "I wouldn't take him at this current ADP" mentality when doing his rankings, that skews things....no?.....

For the overall rankings list? I don't think there's any problem with having several different philosophies in play. If one guy wants to rank by EV, and another by expectations, and another by highest upside, and another by highest floor, and yet another by value at current ADP, it's going to produce a pretty big hodgepodge... but any ranking tool that's providing you with 20 different sets of rankings all at once is naturally going to be prone to producing a hodgepodge. It's meant to represent a pretty diverse range of viewpoints, and in that it succeeds. That's also why it includes the "exclude" button, so you can filter out the viewpoints you don't find helpful to come up with a composite that you're comfortable with. I certainly would not feel comfortable in the slightest using the staff composite rankings as a simple "draft list", just going down the line one by one ticking players off. I use it more as a representative of the different players in my league. I know that if I'm in a league with 11 other guys, they're all going to have different opinions and philosophies, and the ranking composite gives me an idea of what an aggregate of those 11 different opinions and philosophies might look like.

For something like the Draft Dominator where it is explicitly advertising its value in telling you exactly who to draft and when, then yeah, it's pretty important for everyone to be on the same page. Of course, the Draft Dominator player values are based solely on actual projection data from Dodds, Tremblay, Henry, and Wood, and they're all definitely on the same page in terms of approach.

Edit: Perhaps I'm much more comfortable with this tension since I'm coming from a primarily dynasty background, where philosophical differences are the rule and not the exception. The idea of ranking players based on a consensus derived from a variety of different guys with radically different viewpoints just seems anathema to me. Jeff Pasquino plays dynasty with a very short-window, win-now mindset. Chad Parsons plays with a long-window, build-depth mindset. Both styles can be very successful. If you averaged out their rankings against each other, you'd wind up with a milquetoast middle-ground ranking that would likely be far, far less successful than either extreme. I would never say that you should kick out all the Pasquinos and hire only Parsons, or that you should kick out all of the Parsons and hire only Pasquinos. Instead, I'd say that you likely have some Parsons in your league, and you likely have some Pasquinos in your league, and there's a lot of value to be gained from learning what makes each of them tick- even if you should not yourself strive to be some abominable Parsons/Pasquino hybrid.

 

[Dr. Octopus]

I wonder how much of Wimer's rankings represent how willing he is to buy at the player's current ADP. That would certainly explain the drop, as well as Gronk's "off the board entirely".

His execution, if that was the case, would make very little sense. If Gronks ADP is TE3 and Cameron's TE5 - there's a pretty huge gap in ranking them as TE20 and TE28.

If he thought the ADP of TE5 was too high, would he really pass Cameron up if he somehow dropped to TE12 out of principal?

Fantasy backup level TEs are not valuable assets. Cameron and Gronk's upside alone make them far more valuable assets than say Brandon Pettigrew or Jermaine Greschem as bench depth.

 

[cheese]

correct me if I am wrong, but for this tool.....shouldn't you all be using the same mindset/representation/rubric....?

if one guy is straight up ranking on his projections and another guy is using the "I wouldn't take him at this current ADP" mentality when doing his rankings, that skews things....no?.....

For the overall rankings list? I don't think there's any problem with having several different philosophies in play. If one guy wants to rank by EV, and another by expectations, and another by highest upside, and another by highest floor, and yet another by value at current ADP, it's going to produce a pretty big hodgepodge... but any ranking tool that's providing you with 20 different sets of rankings all at once is naturally going to be prone to producing a hodgepodge. It's meant to represent a pretty diverse range of viewpoints, and in that it succeeds. That's also why it includes the "exclude" button, so you can filter out the viewpoints you don't find helpful to come up with a composite that you're comfortable with. I certainly would not feel comfortable in the slightest using the staff composite rankings as a simple "draft list", just going down the line one by one ticking players off. I use it more as a representative of the different players in my league. I know that if I'm in a league with 11 other guys, they're all going to have different opinions and philosophies, and the ranking composite gives me an idea of what an aggregate of those 11 different opinions and philosophies might look like.

For something like the Draft Dominator where it is explicitly advertising its value in telling you exactly who to draft and when, then yeah, it's pretty important for everyone to be on the same page. Of course, the Draft Dominator player values are based solely on actual projection data from Dodds, Tremblay, Henry, and Wood, and they're all definitely on the same page in terms of approach.

Edit: Perhaps I'm much more comfortable with this tension since I'm coming from a primarily dynasty background, where philosophical differences are the rule and not the exception. The idea of ranking players based on a consensus derived from a variety of different guys with radically different viewpoints just seems anathema to me. Jeff Pasquino plays dynasty with a very short-window, win-now mindset. Chad Parsons plays with a long-window, build-depth mindset. Both styles can be very successful. If you averaged out their rankings against each other, you'd wind up with a milquetoast middle-ground ranking that would likely be far, far less successful than either extreme. I would never say that you should kick out all the Pasquinos and hire only Parsons, or that you should kick out all of the Parsons and hire only Pasquinos. Instead, I'd say that you likely have some Parsons in your league, and you likely have some Pasquinos in your league, and there's a lot of value to be gained from learning what makes each of them tick- even if you should not yourself strive to be some abominable Parsons/Pasquino hybrid.

Piecing together a few of your posts... Are you suggesting that Dodds or other rankers are skipping straight to the more successful milquetoast middle-ground rankings and that is why they score better as projections? This is an honest question as I don't want to put words in your mouth.

 

[cdubz]

correct me if I am wrong, but for this tool.....shouldn't you all be using the same mindset/representation/rubric....?

if one guy is straight up ranking on his projections and another guy is using the "I wouldn't take him at this current ADP" mentality when doing his rankings, that skews things....no?.....

For the overall rankings list? I don't think there's any problem with having several different philosophies in play. If one guy wants to rank by EV, and another by expectations, and another by highest upside, and another by highest floor, and yet another by value at current ADP, it's going to produce a pretty big hodgepodge... but any ranking tool that's providing you with 20 different sets of rankings all at once is naturally going to be prone to producing a hodgepodge. It's meant to represent a pretty diverse range of viewpoints, and in that it succeeds. That's also why it includes the "exclude" button, so you can filter out the viewpoints you don't find helpful to come up with a composite that you're comfortable with. I certainly would not feel comfortable in the slightest using the staff composite rankings as a simple "draft list", just going down the line one by one ticking players off. I use it more as a representative of the different players in my league. I know that if I'm in a league with 11 other guys, they're all going to have different opinions and philosophies, and the ranking composite gives me an idea of what an aggregate of those 11 different opinions and philosophies might look like.

For something like the Draft Dominator where it is explicitly advertising its value in telling you exactly who to draft and when, then yeah, it's pretty important for everyone to be on the same page. Of course, the Draft Dominator player values are based solely on actual projection data from Dodds, Tremblay, Henry, and Wood, and they're all definitely on the same page in terms of approach.

Edit: Perhaps I'm much more comfortable with this tension since I'm coming from a primarily dynasty background, where philosophical differences are the rule and not the exception. The idea of ranking players based on a consensus derived from a variety of different guys with radically different viewpoints just seems anathema to me. Jeff Pasquino plays dynasty with a very short-window, win-now mindset. Chad Parsons plays with a long-window, build-depth mindset. Both styles can be very successful. If you averaged out their rankings against each other, you'd wind up with a milquetoast middle-ground ranking that would likely be far, far less successful than either extreme. I would never say that you should kick out all the Pasquinos and hire only Parsons, or that you should kick out all of the Parsons and hire only Pasquinos. Instead, I'd say that you likely have some Parsons in your league, and you likely have some Pasquinos in your league, and there's a lot of value to be gained from learning what makes each of them tick- even if you should not yourself strive to be some abominable Parsons/Pasquino hybrid.

Your point that as a practical matter this is largely irrelevant is well received. The exclusion function allows everyone to choose their own adventure.

However, on a theoretical level I would contend that you are going to get "better" rankings if everyone has a similar goal. Yes, aggregating 20 answers to a question is going to produce a "hodgepodge", but a "hodgepodge" of answers to the same question instead of those answering different questions.

My preference in consuming those ranking is to see the rankings as a staffer would generally take them in a draft with shark competition. That means not adjusting for ADP. This also means not a straight projections list as high volatility players are generally going to be preferable in later rounds.

What we don't know is if any of this even comes into play with Wimer's rankings. If he literally thinks that Cameron is TE21 at this point in time and would take 20 guys ahead of him, I am glad he has the stones to take that stand publicly. That is exactly what I want out of those rankings. If he has Cameron as TE21 for some vague reason of he might be roughly the 8th-12th best TE, but his ADP is too high - then that is whack.

 

[Stinkin Ref]

correct me if I am wrong, but for this tool.....shouldn't you all be using the same mindset/representation/rubric....?

if one guy is straight up ranking on his projections and another guy is using the "I wouldn't take him at this current ADP" mentality when doing his rankings, that skews things....no?.....

For the overall rankings list? I don't think there's any problem with having several different philosophies in play. If one guy wants to rank by EV, and another by expectations, and another by highest upside, and another by highest floor, and yet another by value at current ADP, it's going to produce a pretty big hodgepodge... but any ranking tool that's providing you with 20 different sets of rankings all at once is naturally going to be prone to producing a hodgepodge. It's meant to represent a pretty diverse range of viewpoints, and in that it succeeds. That's also why it includes the "exclude" button, so you can filter out the viewpoints you don't find helpful to come up with a composite that you're comfortable with. I certainly would not feel comfortable in the slightest using the staff composite rankings as a simple "draft list", just going down the line one by one ticking players off. I use it more as a representative of the different players in my league. I know that if I'm in a league with 11 other guys, they're all going to have different opinions and philosophies, and the ranking composite gives me an idea of what an aggregate of those 11 different opinions and philosophies might look like.

For something like the Draft Dominator where it is explicitly advertising its value in telling you exactly who to draft and when, then yeah, it's pretty important for everyone to be on the same page. Of course, the Draft Dominator player values are based solely on actual projection data from Dodds, Tremblay, Henry, and Wood, and they're all definitely on the same page in terms of approach.

Edit: Perhaps I'm much more comfortable with this tension since I'm coming from a primarily dynasty background, where philosophical differences are the rule and not the exception. The idea of ranking players based on a consensus derived from a variety of different guys with radically different viewpoints just seems anathema to me. Jeff Pasquino plays dynasty with a very short-window, win-now mindset. Chad Parsons plays with a long-window, build-depth mindset. Both styles can be very successful. If you averaged out their rankings against each other, you'd wind up with a milquetoast middle-ground ranking that would likely be far, far less successful than either extreme. I would never say that you should kick out all the Pasquinos and hire only Parsons, or that you should kick out all of the Parsons and hire only Pasquinos. Instead, I'd say that you likely have some Parsons in your league, and you likely have some Pasquinos in your league, and there's a lot of value to be gained from learning what makes each of them tick- even if you should not yourself strive to be some abominable Parsons/Pasquino hybrid.

dynasty and redraft are apples and oranges....we need to be careful in this discussion here not to mingle them.....

this discussion was solely about redraft rankings....and I will stand by my belief that if someone is going to provide rankings for the tool we are talking about....you should all be doing them with the same mindset/approach (this is who I think will be next best)......whether you can filter people out or not should not matter.....

 

[Monsters of the Midway]

If you don't like Wimer's rankings you can either read his rationales and confirm your opinions by discounting his, ignore his rankings, or both. I don't see an issue.

 

[Clifford]

So Logan Paulsen = SOD? Who is Logan Paulsen?

 

[ConnSKINS26]

So Logan Paulsen = SOD? Who is Logan Paulsen?

Redskins 2nd string TE.

 

[bicycle_seat_sniffer]

Wimer has been called out here for years for his unique rankings, dont think hes an expert at all

 

[Adam Harstad]

Piecing together a few of your posts... Are you suggesting that Dodds or other rankers are skipping straight to the more successful milquetoast middle-ground rankings and that is why they score better as projections? This is an honest question as I don't want to put words in your mouth.

No, I'm saying Dodds has a better accuracy than the other rankers because he's simply better at ranking than the other rankers. There's a lot of science to creating great projections, normalizing to league history, finding the trends in the data, and accurately projecting them forward. There's a lot of art to creating great projections, too; you have to anticipate changes before they happen, imagine the impact of personnel moves in chaotic systems with untold possibilities. It's a very difficult balancing act to pull off, and historically, few have been as good at it as David Dodds.

Personally, I don't use that process. It's a very, very good process, but I have neither the skill, the knowledge, nor the deft touch to pull it off. I create my rankings off of a series of much simpler heuristics. They've served me quite well over the years. If I were in a league with Dodds, I would estimate that I had a pretty good chance of outperforming him- maybe a 45% chance, even. If I were in a million leagues with Dodds, my odds of outperforming him overall would be much, much lower. Still, like I said, my heuristics are the best process for me with my own personal skills. I can't do what Dodds does, but I can do what I do, and I can do it pretty well I think.

If someone were to average my rankings with Dodds' rankings and split the difference, he'd wind up with something that was half specific projection, and half vague estimation. The resulting "middle ground", I would think, would be less accurate than either using Dodds' rankings alone, or using my rankings alone. By mixing philosophies, you wind up with a muddled mess that lacks the strengths of either approach.

To put it in culinary terms, David is making a cheesecake and I'm making barbecue ribs. David makes a fantastic cheesecake- it's actually internationally known. My ribs aren't quite as acclaimed, but they're still pretty good in their own right- I've got some regional barbecue trophies sitting on my mantle. Now imagine someone else wants to start cooking, and they know that Dodds' cheesecake is really good, and they know that my ribs are really good, so they make up a batch of Dodds' cheesecake batter, dip my ribs in it, and smoke the whole thing for 12 hours. Suddenly, you wind up with a dish that has all of the components of two award-winning concoctions, but instead of being even better than either alone, the combination is far, far worse.

If you browse around over on the FantasyPros Accuracy Rankings, I think this concept becomes pretty apparent. Using the 3-year accuracy rankings, for instance, there have been 36 expert sources that have submitted their rankings for evaluation for each of the last three seasons. Of these 36 sources, 30 have been individuals (or, in the case of the Sablich Brothers, a two-person team), and 6 more have been explicitly labeled as "Staff Rankings" or "Site Projections", implying they are an amalgamation of several different viewpoints. Of the 36 experts, the composite rankings finished 12th, 13th, 30th, 33rd, 35th, and 36th (Dodds, for what it's worth, checks in at 6th). Over the last two seasons, there have been 70 experts submitting rankings to FantasyPros for evaluation, and 10 of those 70 were explicitly labeled as "Staff Rankings" or "Site Projections". Of those ten, the highest one checked in at 16th overall, and three finished in the bottom six (Dodds, for what it's worth, checks in at 4th of 70). The average finish of the ten was 45.1 overall, noticeably below average. If there's some truth to the "wisdom of crowds" when it comes to projection accuracy, I haven't seen it. Instead, it seems to me that the individual voices outperform the consensuses from year to year. Dipping those ribs into that cheesecake batter doesn't make for a better product. Letting the cheesecake guys make their cheesecake, and the rib guys make their ribs, winds up producing a better product overall. In other words, if your only goal is accuracy, I think you're better off listening to just one guy than you are averaging a bunch of guys together.

Of course, as I said, there are other goals to rankings beyond merely accuracy. Accuracy is vitally important, for sure, but you're probably drafting in a league with 11 other players, each with his own unique viewpoint. If you can gain some sort of insight and understanding into the thoughts of your competition, that has value, too. If you just went off of David Dodds' rankings, you might never know that the crowd loves player X, or that some small percentage of people are sky-high on player Y. That's the value that I think the ranking consensus really provides. It's noisy, it's messy, it's not always going to be the most accurate, but it sure as heck does a better job of actually modeling the noisy, messy realities of actual real-world fantasy leagues than a single set of expertly-crafted projections.

 

[jurb26]

I don't know what process anybody else uses to create rankings but personally mine is fairly straight forward, yet time consuming. IMO rankings can not be created unless you are projecting the stats for every player. Well, every feasible player anyway. Once you've done that, excel becomes your best friend as it creates the rankings for you. I then add a touch of subjective risk/reward factor into certain players and perhaps move them up or down slightly based on their point finish, but not much... Never much. From there I simply reference common ADP data and if it's a league I'm familiar with personal history to how they draft or manage teams.

So I guess my response to this thread is the same as it is every year. Don't like someone's rankings? Just do your own. Quite frankly I don't know if I could manage my fantasy life by using someone else's rankings and projections. It just feels wrong.

 

[Bronco Billy]

I'm not sure why so many leap to the defense of the staff when they are criticized. They've chosen to portray themselves as experts on very subjective material and make their analysis public. The site charges for its consumption. As such, they ought to expect criticism, and in cases where their opinion appears well outside conventional rational boundaries they ought to expect rebuke.

This goes with the territory they have chosen. Claimed expertise in a subject matter like this should bring forth differing opinions. The staff are grown ups and have chosen this venue. They can defend themselves. And because it is put forth in public - and a significant amount of material released for pay - as long as people are reasonably respectful and stick to the subject matter, they ought to be able to criticize without coming under fire from those here who think the staff requires sheltering.

 

[fightingillini]

Look, Mark Wimer has been on the FBG staff for years. And his rankings have always been a little different than other staff members, so this isn't a new thing. He thinks very differently than most on staff (and on the messageboard). I respect that. If you don't like his rankings, just exclude him from the averages.

That said, I am a little puzzled by Wimer taking Gronkowski completely off his board. I assume he means that "I won't draft Gronk anywhere near his ADP". That's fair. But there has to be a round where Mark would actually feel that it's worth the risk. For me, to assess that.....say Gronk is projected for 75 rec, 1050 yds, and 12 TD if he played 16 games. That equates to 252 points in a PPR league. Say you think he will miss 4 games (not unreasonable). Then you will have to start a backup TE in his place for 4 games.....call that TE18 in a 12 teamer. Say TE18 is projected for 50 rec, 625 yds, and 5 TD, which equates to 142 pts. Gronk's projection would be 252*(12/16) + 142*(4/16) = 224.5 In other words, I would project Gronk for 10% less due to his injury risk, and probably would add a little extra on that since he may be injured during the playoffs. So if I thought Gronk was going to miss 4 games, I would probably project him for about 210 points instead of 252. Currently at 252, Gronk would be a 3rd round pick, but at 210 I would probably would value him in the late 4th/5th round. Now most likely I won't get Gronk in the 4th or 5th round, but that's fine with me.

Wimer has to think Gronk is going to miss half the year or more to take him completely off his board.

 

[fightingillini]

If someone were to average my rankings with Dodds' rankings and split the difference, he'd wind up with something that was half specific projection, and half vague estimation. The resulting "middle ground", I would think, would be less accurate than either using Dodds' rankings alone, or using my rankings alone. By mixing philosophies, you wind up with a muddled mess that lacks the strengths of either approach.

This is an interesting comment. I actually use the average of Dodds/Henry/Wood/Tremblay projections as a starting point, and I work from there. I do that so that I get a baseline of basic stats for each team. So I like to see for each team.....run to pass ratio, how each splits up the pie, and so forth. From there, I use my opinion for each team. Usually, I would generally agree with one of Dodds/Henry/Wood/Tremblay for each team, with some minor tweaks so that my rankings come out the way I want them. I usually agree with Wood the most, followed by Henry, Dodds, and Tremblay. But in many cases, the 4 guys have similar projections for a particular team. For those teams, I will for the most part use them unless I have a really different view for that team.

So I spend most of my projection time on players/teams where the 4 guys differ the most on.

I wouldn't think that this is a half vague estimation, as long as you are projecting each team separately.

 

[Adam Harstad]

If someone were to average my rankings with Dodds' rankings and split the difference, he'd wind up with something that was half specific projection, and half vague estimation. The resulting "middle ground", I would think, would be less accurate than either using Dodds' rankings alone, or using my rankings alone. By mixing philosophies, you wind up with a muddled mess that lacks the strengths of either approach.

This is an interesting comment. I actually use the average of Dodds/Henry/Wood/Tremblay projections as a starting point, and I work from there. I do that so that I get a baseline of basic stats for each team. So I like to see for each team.....run to pass ratio, how each splits up the pie, and so forth. From there, I use my opinion for each team. Usually, I would generally agree with one of Dodds/Henry/Wood/Tremblay for each team, with some minor tweaks so that my rankings come out the way I want them. I usually agree with Wood the most, followed by Henry, Dodds, and Tremblay. But in many cases, the 4 guys have similar projections for a particular team. For those teams, I will for the most part use them unless I have a really different view for that team.

So I spend most of my projection time on players/teams where the 4 guys differ the most on.

I wouldn't think that this is a half vague estimation, as long as you are projecting each team separately.

Yeah, Dodds/Wood/Tremblay/Henry are all using similar processes, so mixing and matching with those four makes more sense. I was specifically referring to mixing Dodds' rankings with mine, since mine are based on heuristics and rules of thumb ("vague estimation"), while his are based on hard-and-fast predictions ("specific projection").

 

[Don Quixote]

Looks like Wimer has moved Gronk up to #4 in his rankings.

 

[Stinkin Ref]

Look, Mark Wimer has been on the FBG staff for years. And his rankings have always been a little different than other staff members, so this isn't a new thing. He thinks very differently than most on staff (and on the messageboard). I respect that. If you don't like his rankings, just exclude him from the averages.

That said, I am a little puzzled by Wimer taking Gronkowski completely off his board. I assume he means that "I won't draft Gronk anywhere near his ADP". That's fair. But there has to be a round where Mark would actually feel that it's worth the risk. For me, to assess that.....say Gronk is projected for 75 rec, 1050 yds, and 12 TD if he played 16 games. That equates to 252 points in a PPR league. Say you think he will miss 4 games (not unreasonable). Then you will have to start a backup TE in his place for 4 games.....call that TE18 in a 12 teamer. Say TE18 is projected for 50 rec, 625 yds, and 5 TD, which equates to 142 pts. Gronk's projection would be 252*(12/16) + 142*(4/16) = 224.5 In other words, I would project Gronk for 10% less due to his injury risk, and probably would add a little extra on that since he may be injured during the playoffs. So if I thought Gronk was going to miss 4 games, I would probably project him for about 210 points instead of 252. Currently at 252, Gronk would be a 3rd round pick, but at 210 I would probably would value him in the late 4th/5th round. Now most likely I won't get Gronk in the 4th or 5th round, but that's fine with me.

Wimer has to think Gronk is going to miss half the year or more to take him completely off his board.

see I don't think it is "fair" for the rankings tool we are talking about......

when I look at the ranking tool we are talking about....I want to be able to assume that every staffer would take the next guy they have listed at that position.....

 

[Tau837]

Looks like Wimer has moved Gronk up to #4 in his rankings.

How can that be? Did Gronk already make a preseason appearance?

 

[Joe Bryant]

As an aside, I don't see much difference between ranking Gronk as, say, TE7 and taking Gronk off the board entirely. Either ranking essentially translates to "there is absolutely no way in hell this guy will ever wind up on any of my teams", although the first option allows for far more face-saving if you're wrong.

What Adam said here. It becomes something of a semantics issue at some point. If a ranker has Gronkowski at TE 7 in July as he wants to see him on the field (I believe this ranking was prior to Belichick's announcement about Gronkowski not going on PUP) then that's effectively not having him on the board. BUT, that also doesn't give you guys a true picture of where Gronkowski ranks. Mark understands what I'm looking for here now and he moved him back in.

The other angle is something I've thought for a long time but I feel the need to repeat it every year. The easiest and safest way to produce a ranking is to go straight off the Average Draft Position. Nobody bats an eye. Nobody calls you out. I don't have to spend 1 second defending a ranking here or answering the email telling me I'm stupid. Going with a straight ADP ranking is a piece of cake.

It's also not much value. We already have ADP rankings.

What I want our Staffer rankings to be is guys working hard at this, putting a lot of thought into it, and going with what they really feel. I love it when it's off the ADP. Because then people start thinking about it. Sure, you'll have the guys who yell "Player A over Player B? " But you'll also have some real discussion either here on the boards or within your own head as you think about your own rankings. And that's what we're going for.

J

 

[Soulfly3]

No offense, but what kind of discussion is Toilolo being ranked 15+ spots ahead of Cameron and Gronk going to generate?

 

[FreeBaGeL]

As an aside, I don't see much difference between ranking Gronk as, say, TE7 and taking Gronk off the board entirely. Either ranking essentially translates to "there is absolutely no way in hell this guy will ever wind up on any of my teams", although the first option allows for far more face-saving if you're wrong.

What Adam said here. It becomes something of a semantics issue at some point.

Semantics? This seems a bit odd.

For starters, it affects the guy's consensus ranking a lot more heavily. Having Gronk at TE30 pushed him down to 4th in the consensus rankings, whereas having him at TE7 would make Gronk 2nd in the consensus rankings. So what exactly is the thing dropping Gronk down to 4th in the FBG consensus rankings (which is a huge difference in the TE space and probably puts him the lower on FBG than any other website)? Is it a consensus among FBG staffers that Gronk is a worse option than Julius Thomas and Jordan Cameron? Is it a passionate feeling among a couple staffers that Gronk is overrated? Nope, it's that one guy just didn't really feel like ranking him so he threw him down near the bottom somewhere, leading to the FBG consensus rankings saying something that none of the staff actually believe.

I mean, is "eh, I have this guy a little lower than most people anyway so I'm just going to throw him in somewhere randomly" really a FBG endorsed rankings technique?

If you're doing a set of rankings for yourself and all that matters is "I'm not going to get this guy anyway" then I'd agree, it's a semantics issue. But when you're publishing your rankings publicly (for a fee, no less) I would think that people are going in with the expectation that a guy is actually, ya know, ranking the players.

 

[Tau837]

Mark understands what I'm looking for here now

Interesting comment given Mark has been submitting rankings for FBGs for a long time now. Is it now safe to assume that all of the staff members submitting rankings now share a common understanding of what is expected in their rankings?

 

[Tau837]

No offense, but what kind of discussion is Toilolo being ranked 15+ spots ahead of Cameron and Gronk going to generate?

In fairness, Mark provides a comment on his Toilolo and Cameron rankings. You may not agree, but the rationale is there. In the case of Cameron, IMO the rationale is questionable, and it goes back to the earlier discussion in the thread about "prove it" rankings.

 

[Anarchy99]

I disagree that ranking Gronk at TE7 is the same as dropping him off your rankings altogether. I could easily see a situation where that's where he winds up. If I thought he was going to miss a few games, was only playing at 80%, maybe he was asked to block more or the team ran more, the team struggled in the red zone, etc.

Even today I heard that just because he was not placed on the PUP list to start camp does not mean he is 100% and be prepared for him to be a partial or limited participation in practice for much of, if not all of, training camp. Remember, players on the PUP cannot practice at all. There is also still a chance Gronk does not suit up for Week 1. (This is not me providing an opinion, it is me forwarding an opinion of one of the Patriots beat guys.)

So I am not sure people have to have Gronk ranked 3rd or off their board. There is still some uncertainty here.

 

[Joe Bryant]

Mark understands what I'm looking for here now

Interesting comment given Mark has been submitting rankings for FBGs for a long time now. Is it now safe to assume that all of the staff members submitting rankings now share a common understanding of what is expected in their rankings?

Yes.

 

[Maurile Tremblay]

Besides, if you like rankings that aren't a hedge, then you should like Wimer's ranking of Gronk a whole lot. He's not hedging at all against the possibility that Gronk comes back and plays at full health, he's 100% committed to the "I don't think Gronk can/will stay healthy" line of reasoning.

I disagree about Wimer's Gronk ranking, as I said initially. His comment sounds like he is hedging until he sees "proof," not that he is saying he believes Gronk won't play this season. If he has him off his board because he doesn't expect him to play, then IMO that is appropriate. But that's not what his comment says.

Adam hit the mark on the head. Wimer's comment may have sounded like a hedge, but his ranking was the epitome of a non-hedge. It's exactly what you get when your ranking is based on "either he's going to play or he's not" instead of something more nuanced like "there's about an 85% chance that he'll play."

Part of doing good rankings or projections, IMO, is having the courage to stick your neck out when you're confident in your position, while also having the humility to admit that you don't have enough information to be confident about every issue that comes up.

Sometimes you just don't know whether Rob Gronkowski -- or Brett Favre, or Josh Gordon -- will play or not, and in that case all you can do is estimate the probability as best you can. In such situations, any reasonable estimate will almost never be 0% or 100%; it will be something in between. To pick 0% or 100% and just go with it is a hubris that leads to poor rankings and poor drafting, IMO.

 

[Maurile Tremblay]

This assumes that the whole point of having all those individual rankings is to maximize accuracy. I'm rather thankful that it's not, because if it was, Dodds would just fire us all and submit one set of rankings; his own. Seriously, Dodds has one of the best track records on rankings in the entire industry. I submit rankings based on what I believe will come true, but I still recognize that Dodds' rankings are in all likelihood a better representation of what will actually happen. If you really wanted to maximize your accuracy, you should probably exclude my own rankings, and everyone else's too, and just go with what Dodds says.

That doesn't follow. Dodds is the best in the biz. I'd bet on him against me. But that doesn't mean that I'd bet on him against the FBG consensus. A consensus of independent projectors will often outperform even the single most accurate projector among them over the long run.

If someone were to average my rankings with Dodds' rankings and split the difference, he'd wind up with something that was half specific projection, and half vague estimation. The resulting "middle ground", I would think, would be less accurate than either using Dodds' rankings alone, or using my rankings alone.

I don't think that's mathematically possible. To illustrate with a simple example, suppose that you and Dodds each project the number of yards that Andrew Luck will throw for this season. There is not a single possible combination of projections by you and Dodds such that the average will be less accurate than either of your individual guesses. The average may finish first or second (or tied for first), but never third.

I believe that principle holds when we're talking about ordered lists as well.

If you browse around over on the FantasyPros Accuracy Rankings, I think this concept becomes pretty apparent. Using the 3-year accuracy rankings, for instance, there have been 36 expert sources that have submitted their rankings for evaluation for each of the last three seasons. Of these 36 sources, 30 have been individuals (or, in the case of the Sablich Brothers, a two-person team), and 6 more have been explicitly labeled as "Staff Rankings" or "Site Projections", implying they are an amalgamation of several different viewpoints.

I don't think it implies they are an amalgamation; I think it implies that even if they are done by one person, nobody wants to take credit for them.

If there's some truth to the "wisdom of crowds" when it comes to projection accuracy, I haven't seen it.

Link

That's just for 2010, when the overall consensus finished #2 out of 39 experts (well, out of 40 including the consensus itself). I can't find data for the other years, but it's very likely -- if fantasy football projections are like most other kinds of projections -- that the consensus would be relatively near the top in most individual years, and very much near the top for any period of multiple years.

---

What's interesting is that when we're talking about the accuracy of a group's consensus, the accuracy of the individual projectors within the group obviously matters, but what matters just as much is the independence among those in the group. Holding their individual accuracies constant, the less their projections correlate with each other, the better. (And again, the way these things are typically measured, accuracy and independence are equally important. It's the diversity prediction theorem.)

That means that the consensus of a bunch of people who use completely different approaches is way better, holding accuracy roughly constant, than that of a bunch of people who use a similar approach.

It means that Wimer's refusal to march to the same beat as everyone else, when it comes to his effect on the accuracy of the overall consensus, is a strength rather than a weakness.

Obviously, as with the rest of us, Wimer's rankings would be more helpful if they were more accurate. But for any given level of accuracy, his rankings become more valuable, not less valuable, by virtue of departing so much from the rest of ours.

And for dynasty, a Parsons-Pasquino hybrid may be exactly the way to go.

 

[Adam Harstad]

If someone were to average my rankings with Dodds' rankings and split the difference, he'd wind up with something that was half specific projection, and half vague estimation. The resulting "middle ground", I would think, would be less accurate than either using Dodds' rankings alone, or using my rankings alone.

I don't think that's mathematically possible. To illustrate with a very simple example, suppose that you and Dodds each project the number of yards that Andrew Luck will throw for this season. There is not a single possible combination of projections by you and Dodds such that the average will be less accurate than either of your individual guesses. The average may finish first or second (or tied for first), but never third.

I believe that principle holds when we're talking about ordered lists as well.

I was under the impression that it did not still hold when minor ordinal differences resulted in outsized impacts. For instance, if the consensus opinion on Josh Gordon last year was WR30, you wouldn't have to have him 1st or 2nd in your rankings- simply having him at WR26 would be only a small edge in terms of accuracy, but would be a large edge in terms of chances of walking out of the season with a title. Winning fantasy titles requires accumulating more VBD than everyone else, and the top players contribute a dramatically outsized share of that VBD, so even small ranking advantages when it comes to the eventual top players translate into large differences. In large part, your ability to win your league stems almost entirely from your ability to be just slightly ahead of the curve on those breakout players and land them in as many leagues as possible.

I'm imagining a hypothetical where I was slightly higher than consensus on Alshon Jeffery and significantly lower than consensus on Josh Gordon, while Dodds was slightly higher than consensus on Josh Gordon and significantly lower than consensus on Alshon Jeffery. Using Dodds' rankings would ensure you wound up with Josh Gordon. Using my rankings would ensure you wound up with Alshon Jeffery. Using a combination would ensure you wound up with neither.

Of course, I know there has to be more to it than that- if I have Gordon 10 slots below consensus, then I must necessarily have all other receivers a combined 10 slots above consensus to offset, since rankings are a zero-sum game (if I move one player up, I must move everyone else down to offset). It's difficult to think through the impact of all of those shifting dominoes on the outcome, but it seems to me that the theory is sound. If I have Jeffery 5 slots higher and Gordon 10 slots lower, then I must also have someone else 5 slots higher to offset (or, alternately, I could spread that increase among multiple receivers). If Dodds had Gordon 5 slots higher and Jeffery 10 slots lower, he must likewise have someone else 5 slots higher to offset. If both Dodds and I made up that difference by moving, say, Legedu Naanee up by the requisite 5 slots, then that would have no impact on the accuracy of the consensus relative to our own individual accuracies. The consensus would share a demerit with Dodds and me for being too high on Naanee, but it would not gain an offsetting bonus for being higher on either Gordon or Jeffery. Consensus would basically say to avoid both Gordon and Jeffery in favor of Naanee, while Dodds and I would say to avoid one of the two in favor of Naanee and the other, certainly a more profitable piece of advice.

Then, also, there is this: if it is possible for a consensus ordinal ranking to be more accurate than any single one of its component parts, then must it not also be possible for a consensus ranking to be LESS accurate than any single one of its component parts? I mean, otherwise that suggests I could take one randomly generated list, average it out against another randomly generated list, and the resulting "consensus" would outperform simple random chance- a quite counterintuitive finding, if true.

And a Parsons-Pasquino hybrid may be exactly the way to go.

This one is much more easily dispelled with a simple thought experiment. If the Pasquino approach would leave you with the best team in year 1 and the worst team in year 2, while the Parsons approach would leave you with the worst team in year 1 and the best team in year 2, then averaging them together will almost certainly result in you having a middle-of-the-road team in both years, which is strictly inferior to either of the other two possible outcomes. The goal is not to outperform the average, it is to outperform the average by a greater amount than anyone else. When in search of extreme outcomes, it seems wise to avoid averaging inputs.

 

[Maurile Tremblay]

Then, also, there is this: if it is possible for a consensus ordinal ranking to be more accurate than any single one of its component parts, then must it not also be possible for a consensus ranking to be LESS accurate than any single one of its component parts? I mean, otherwise that suggests I could take one randomly generated list, average it out against another randomly generated list, and the resulting "consensus" would outperform simple random chance- a quite counterintuitive finding, if true.

A randomly generated list contains no useful information. Combine a bunch of randomly generated lists together and you still have just a randomly generated list.

If we are dealing with non-random lists, though, I think it holds. In fact, the link I pointed to earlier specifically uses mock drafts as predictors of the actual NFL Draft to illustrate the principle.

It's the diversity prediction theorem.

The principle, stated on page 209, is that the consensus list will have a lower total square of the errors than the average square of the errors from each individual list that forms the consensus.

Page 210 uses the NFL draft to illustrate, which seems rather comparable to fantasy rankings in that they are both ordered lists.

 

[Maurile Tremblay]

And a Parsons-Pasquino hybrid may be exactly the way to go.

This one is much more easily dispelled with a simple thought experiment. If the Pasquino approach would leave you with the best team in year 1 and the worst team in year 2, while the Parsons approach would leave you with the worst team in year 1 and the best team in year 2, then averaging them together will almost certainly result in you having a middle-of-the-road team in both years, which is strictly inferior to either of the other two possible outcomes. The goal is not to outperform the average, it is to outperform the average by a greater amount than anyone else. When in search of extreme outcomes, it seems wise to avoid averaging inputs.

This is going to depend on how we measure stuff, what our goals are, and on league parameters, etc.

But I tend to think that the way to maximize championships, in most league formats, is not to have the best team in year 1 and the worst team in year 2 or vice versa. Rather, it is to make the playoffs as often as possible, where anything can happen, and then try to get lucky. (The best team is seldom the league champ, in my experience.)

In your example, the best way to make the playoffs in both year 1 and year 2 may be to use the hybrid strategy.

 

[Borden]

I was listening to a podcast (CBS Fantasy Football, I think but I'm not sure) and one of the hosts was talking about how his rankings don't really reflect how he actually drafts. I'm pretty sure I've heard this on other podcasts as well.

Within the context of the conversations it seemed as though the rankings were based more on "realistic" player production numbers rather then potential upside or decline. It seemed as the rankings were geared more toward getting a higher percentage and rank with Fantasy Pros then reflecting true value.

The point I'm getting at is rankings can't and shouldn't have a distinct format that everyone uses across the board. Some people want to hit singles at .600, others want to hit home runs at .380 and most people are somewhere in between.

On top of that I think that the worse thing that could happen to a group of people ranking is to fall into group think. We've all seen (or more likely been part of it) on this board.

All that said, I know I've seen rankings that seemed off from the rest of the staff and when you go click on the blurb/write up I've felt disappointed in the reasoning or lack of content to support the ranking. Maybe a "Read more..." link would help with some of the blurbs.

 

[ggggh7]

yeah just wish FBG had more "from the gut" rankings, Rank only the guys you think will do well. avoid, ect, QB combos, Playoff scedules,

Week one thru three matchups... Rank the Defense ,

The way these ranking are done on all fantasy sites is good for only one think ADP .

 

[Ramblin Wreck]

I think one responsible thing a staffer should do is add comments to their rankings especially when it varies from the consensus ranking. A quick explanation of why you like or don't like a player is a lot more valuable than arguing over why a top 3 tight end is ranked 7th, 12th or 20th on someone's list.

 

[Bronco Billy]

As an aside, I don't see much difference between ranking Gronk as, say, TE7 and taking Gronk off the board entirely. Either ranking essentially translates to "there is absolutely no way in hell this guy will ever wind up on any of my teams", although the first option allows for far more face-saving if you're wrong.

You are a master of convoluted logic, but this may top everything. If this is seriously your position, you shouldn't be ranking. What you've just put forth is that there is no difference in the positions of, "Given the risk I wouldn't take Gronkowski before the 7th round." and "If Gronkowski were there in the 15th round I still wouldn't draft him." If you, or any other staff member, equate those two positions, which is exactly what you've done in your statement above, then you simply should not be ranking for a pay site. There's just no way around that.

.

 

[Adam Harstad]

As an aside, I don't see much difference between ranking Gronk as, say, TE7 and taking Gronk off the board entirely. Either ranking essentially translates to "there is absolutely no way in hell this guy will ever wind up on any of my teams", although the first option allows for far more face-saving if you're wrong.

You are a master of convoluted logic, but this may top everything. If this is seriously your position, you shouldn't be ranking.What you've just put forth is that there is no difference in the positions of, "Given the risk I wouldn't take Gronkowski before the 7th round." and "If Gronkowski were there in the 15th round I still wouldn't draft him." If you, or any other staff member, equate those two positions, which is exactly what you've done in your statement above, then you simply should not be ranking for a pay site. There's just no way around that.

I never said that there was NO difference, I said that there was NOT MUCH difference, which is a categorically different statement. Saying "not much difference" acknowledges that a difference exists, explicitly rejecting the position that there is no difference at all. Further, I stand behind the statement. There's a theoretical difference between the two positions, but since Gronkowski is never going to hit the 7th round, there is absolutely no practical difference whatsoever.

Take a guy who says "I will draft Gronkowski when he hits the 7th round" and another guy who said "I will never draft Rob Gronkowski. You can put that in caps. NEVER." Have them each draft 1000 MFL10s. Who do you think winds up with Gronkowski on his team more often? I'm willing to bet that you wind up with a scoreless tie. Saying that you'd theoretically take Gronk in the 7th round if he ever fell that far (all while knowing full well that he never would) amounts to a fig leaf. If you wind up being wrong after the season, it gives the appearance of being LESS wrong, even though both positions translated practically to "do not draft under any circumstances".

Now, I wouldn't ever rank like that. As has been mentioned, it screws with the consensus averages. But for all practical purposes, if someone printed out just Wimer's rankings and drafted directly from them with no deviation, it makes absolutely no practical difference whatsoever whether Wimer had Gronk as TE7 or TE77- just like it makes no difference if you go into a car dealership and say "I will not buy that car" or "I will not buy that car for more than $20". Either way, you're walking home afterwards.

 

[Stinkin Ref]

is this really that hard...?....my head is spinning reading some of that stuff....

IMO....with the tool we are talking about.....the expectation should be that if any of the staffers are on the clock in a draft I should be able to look at his list of rankings and know who he would pick next at each position....period

 

[Bronco Billy]

As an aside, I don't see much difference between ranking Gronk as, say, TE7 and taking Gronk off the board entirely. Either ranking essentially translates to "there is absolutely no way in hell this guy will ever wind up on any of my teams", although the first option allows for far more face-saving if you're wrong.

You are a master of convoluted logic, but this may top everything. If this is seriously your position, you shouldn't be ranking.What you've just put forth is that there is no difference in the positions of, "Given the risk I wouldn't take Gronkowski before the 7th round." and "If Gronkowski were there in the 15th round I still wouldn't draft him." If you, or any other staff member, equate those two positions, which is exactly what you've done in your statement above, then you simply should not be ranking for a pay site. There's just no way around that.

I never said that there was NO difference, I said that there was NOT MUCH difference, which is a categorically different statement. Saying "not much difference" acknowledges that a difference exists, explicitly rejecting the position that there is no difference at all. Further, I stand behind the statement. There's a theoretical difference between the two positions, but since Gronkowski is never going to hit the 7th round, there is absolutely no practical difference whatsoever.

Yes, I know. You love playing your semantics games. I expected this tactic from you, which is why I quoted your initial post, which stated exactly "As an aside, I don't see much difference between ranking Gronk as, say, TE7 and taking Gronk off the board entirely. "So while you play the game of being offended because you did say there was a difference, the context of your exact quote states that you allocate very little differential between a player drafted in the 7th round (a hair below where TE7 is being drafted in 12 team ppr redraft leagues per MFL) and one who is not drafted at all - your words, not mine. I chose not to get into the semantics and parse words but rather to address the content and meaning of what you posted. I disagree and contend that there is a significant difference between the value of a player drafted in the 7th round - where in many leagues you are still drafting starters or high impact backups - and letting a player fall completely off one's draft list. I further stated that if a person is incapable of recognizing this difference that consideration ought to be given as to their expertise in ranking players.

Then in your response to mine, you state unequivocally that there is no way that the player in question lasts until the 7th round anyhow - your words, not mine ("but since Gronkowski is never going to hit the 7th round, there is absolutely no practical difference whatsoever") - which further calls into question the alleged expertise of someone leaving the subject player ranked extraordinarily low or off a draft list/player ranking entirely - which was my point to begin with.

.

 

[tonydead]

No, I'm saying Dodds has a better accuracy than the other rankers because he's simply better at ranking than the other rankers. There's a lot of science to creating great projections, normalizing to league history, finding the trends in the data, and accurately projecting them forward. There's a lot of art to creating great projections, too; you have to anticipate changes before they happen, imagine the impact of personnel moves in chaotic systems with untold possibilities. It's a very difficult balancing act to pull off, and historically, few have been as good at it as David Dodds.

This is all you really need to know fellas.

 

[Ted Mullins]

IMO....with the tool we are talking about.....the expectation should be that if any of the staffers are on the clock in a draft I should be able to look at his list of rankings and know who he would pick next at each position....period

Not sure if I am off on the purpose/implication of the ranks, but this was pretty much my line of thinking as well...if the ranker would legitimately take 20 other tight ends before Cameron in a draft right now, then fine by me. I'm very skeptical though.

 

[pghrob]

As an FYI, I sent Chad Parsons an IM to see if he is willing to confirm his overall dynasty rankings. He has Davante Adams at 27 and Allen Robinson at 28, with Aaron Dobson not far behind at 32 - overall. Seems a bit high for those guys.

 

[Adam Harstad]

As an FYI, I sent Chad Parsons an IM to see if he is willing to confirm his overall dynasty rankings. He has Davante Adams at 27 and Allen Robinson at 28, with Aaron Dobson not far behind at 32 - overall. Seems a bit high for those guys.

Parsons is very youth-centric and doesn't discount much for future seasons. Notice he has both Watkins and Evans in his top 10, Cooks at 13, and Moncrief at at 22. He has Jordan Matthews at 25, though I've actually managed to top him on that one (I've got him at 24). He's also got Michael, Sankey, and Hyde among his top 12 RBs. I'd be surprised if those particular rankings you noticed were not intentional, that's just his philosophy- pay premium prices for quality youth and ride them for their entire prime.

Might seem insane to some (Tefertiller and I are probably the next most youth-centric staffers, and he's still several standard deviations beyond either of us), but I'm sure if you asked he'd share links to some of his dynasty teams. Those high rankings backfire a lot, but when they hit, they can build some insanely talented dynasty squads. And I do think that he and Pasquino are probably the most consistent rankers. They both have a super clear strategy, and I think they're both fantastic at actually following it through.