I just sent Joe Bryant an email... (1 Viewer)

[ComfortablyNumb]

I sent the following e-mail to Joe. Does it make sense to anybody else but me?

Hi Joe,

Thanks for your reply; I appreciate your perspective. I'll try again to offer some purely constructive feedback, not because I enjoy argument or debate, but because I'd like to help my favorite resource, FBG, to become even better and more useful. And because, as an engineer, I just can't help trying to fix/improve things...

Surely you remember the movie City Slickers, with Billy Crystal and Jack Palance. You may recall one of the key plot lines was Jack Palance trying to teach Billy Crystal about "The ONE THING" that matters most? Well, in fantasy football, the "One Thing" is......

Projections.

It's the basis and bottom line for everything leading up to the draft. Nothing else really matters; without good projections, even awesome tools like the VBD app or Draft Dominator are practically useless. Since nobody has a foolproof crystal ball (not even Dodds), I believe FBG should do everything possible to maximize the reliability of your projections. A few of my suggestions:

1. Stop attempting to project the # of games played by a player. It's simply ludicrous to think you can predict whether Vick, McFadden, or Austin will play 12, 14, or any specific number of games. It would be MUCH more useful to project all players as if they're playing 16 games. We know who the injury/suspension risks are and will adjust our draft strategies accordingly. Including speculation on games missed severely skews the projections and makes comparison MORE DIFFICULT.

2. Again, I will state that a consensus projection which averages the projections of several experts yields a statistically much more reliable result. It's been proven/demonstrated many times with academic studies. With all of the excellent staffers at your disposal, you could also divide the labor. Have 3-4 guys focus on QB's and TE's. A different group could focus on RB's, while a third group could handle WR's. You could let a rookie or two address Kickers and Defenses, as well. Then, MAKE these CONSENSUS Projections available for use in the VBD App and Draft Dominator.

3. Another under-utilized resource is the knowledgable membership of the Shark Pool. I know you already track Shark Pool projections for Spotlight players. Why not compile these Shark Pool consensus projections and make them available to plug into the VBD and DD, as well? Maybe you'd need to include a filtering review or algorithm to eliminate the handful of senseless projections or outliers which could skew the projections? I know I'd REALLY like to see/use some compiled Shark Pool consensus projections, and that's a free resource for you to tap.

4. If you need to free up some time/manpower from your staff to manage the increased attention on projections, perhaps you could eliminate a few of the articles published on FBG. Now, I really enjoy many of the articles, but there are just far too many of them. Who has the time to read all of it?

5. I'm not aware of any type of market research done by FBG. Do you periodically survey your paying customers about what they like or don't like or wish to improve(you probably do, but I'm just ignorant of it)?

Thanks for listening,

Bob

 

[Righetti]

 

[shadyridr]

 

[RIM]

i agree with you.

i have to use PD to have at least 5 choices (4 staff plus me) to average from on projections.

sidenote:

(and no matter how i import those projections, i can never get the PD, DD and VBD app to end up with the same result anyway. it's like their formula's are slightly different between the apps for some reason, no matter what i turn off/on (weekly weights and stuff). Too much manual work needed in the end unless i just want David's projections only.)

 

[butcher boy]

Not sure I agree on the injury prediction comment. There are certain guys I predict will miss a few games and will continue to do so until they prove they can withstand a full season. To me, it's worthless to project 16 games of production for McFadden for example, because I'm more sure than not that he won't play all 16 games. He never has. Until he plays a full season and shows he's either changed his running style, workout regimin, or whatever it is that's making him get so banged up, I'll continue to do so.

Predicting the exact number of games he'll miss is another question and is very debatable. I usually just go by the last few years and average them. I dont think about it too much, because I usually skip guys with the really big injury red flags anyway.

 

[doowain]

 

[Soulfly3]

Once the David Dodds butt sniffers have finished having their hissy fit after reading your letter, expect a headhunt on you, Numb.

Future wannabe staffers are going to have a field day defending their leader.

 

[solorca]

I agree on the injury prediction thing, but I don't agree on cutting out articles to focus more on projections. I subscribe for the articles, not the projections.

I'm sure I'm in the minority though.

 

[Leroy Hoard]

the David Dodds butt sniffers

Thanks, I was looking for a new name for my FF team.

 

[Modog814]

I agree on the injury prediction thing, but I don't agree on cutting out articles to focus more on projections. I subscribe for the articles, not the projections.I'm sure I'm in the minority though.

I'm with you. I don't really look at their projections except occasionally when I review my own and ask myself if it's really reasonable.

 

[comfortably numb]

WTF

though someone hacked into my account

 

[Otis]

TLDNR

 

[Jed]

3. Another under-utilized resource is the knowledgable membership of the Shark Pool. I know you already track Shark Pool projections for Spotlight players. Why not compile these Shark Pool consensus projections and make them available to plug into the VBD and DD, as well? Maybe you'd need to include a filtering review or algorithm to eliminate the handful of senseless projections or outliers which could skew the projections? I know I'd REALLY like to see/use some compiled Shark Pool consensus projections, and that's a free resource for you to tap.

They did this last year with the "Wisdom of Crowds" rankings - which were for the most part spot on every week. They weren't projections (just rankings), but they were very useful during the year. From what I understand that should be happening again this year (I hope).

 

[Clifford]

#4 is spot on. The content here is overwhelming just in volume. If you factor in that most articles range anywhere from 300-900 words, it's impossible to get through. I think most articles could be reduced to 150-250 words without losing anything.

 

[Q-Bert]

2. Again, I will state that a consensus projection which averages the projections of several experts yields a statistically much more reliable result. It's been proven/demonstrated many times with academic studies. With all of the excellent staffers at your disposal, you could also divide the labor. Have 3-4 guys focus on QB's and TE's. A different group could focus on RB's, while a third group could handle WR's. You could let a rookie or two address Kickers and Defenses, as well. Then, MAKE these CONSENSUS Projections available for use in the VBD App and Draft Dominator.

This is a really good idea.

 

[Soulfly3]

TLDNR

As opposed to your threads which are simply LOL

 

[BroadwayG]

What was Jo's perspective?

 

[Sarnoff]

5. I'm not aware of any type of market research done by FBG. Do you periodically survey your paying customers about what they like or don't like or wish to improve(you probably do, but I'm just ignorant of it)?

Joe & David annually ask the Shark Pool for input, suggestions, improvements, complaints, etc. Then they do nothing about it. But they do ask.

 

[Soulfly3]

5. I'm not aware of any type of market research done by FBG. Do you periodically survey your paying customers about what they like or don't like or wish to improve(you probably do, but I'm just ignorant of it)?

Joe & David annually ask the Shark Pool for input, suggestions, improvements, complaints, etc. Then they do nothing about it. But they do ask.

lol. this thread is seriously giving me a laugh.

 

[Addai's Adidas]

What I would find useful is a cheatsheet with just the white guys.

 

[Soulfly3]

What I would find useful is a cheatsheet with just the white guys.

we could even call it an "honest sheet".

 

[zed2283]

What a great way to waste the time of the FBG staff.

 

[Tbone]

The ability to weigh projections is available in the projections dominator so the data is available. You can blend the projections of Dodds, Henry, Tremblay, Woods, and even your own. I've often wanted this same option to be available in the simple VBD Excel application. You can print something from PD that's similar to the Excel application but it's not quite as condensed as the Excel application.

 

[LarryAllen'sJockstrap]

You make some interesting points...

However, I'd recomend you take the any expert's (or individual's) 2011 projections and compare them to what the final stats were.

As a quick example, I did this with Dodds top-50 RBs dated 9/5/2011 vs. season ending actuals.

- 39 or 50 projected players finished top-50 (surprised it was this high).

- Of the 39 that made the top-50 in both lists, the standard deviation in RB ranking was 11.4 spots.

- Of the 39 that made the top-50 in both lists, the standard deviation in fantasy points was 45.7 points.

What does that mean? IMO, even taking out the far outliers, the best in the businees (FBG) only gets you within about a dozen spots of where a player will actually end up, and are off about 3 points a game in either diretion on average.

Projections are fun and make us feel good when we draft players, but they are best used as a tool to guage market value, tiering and estimated usage/touches. To truly expect production = projections is a mistake.

 

[DoubleG]

WTFthough someone hacked into my account

Is your name "Bob"?

 

[RIM]

The ability to weigh projections is available in the projections dominator so the data is available. You can blend the projections of Dodds, Henry, Tremblay, Woods, and even your own. I've often wanted this same option to be available in the simple VBD Excel application. You can print something from PD that's similar to the Excel application but it's not quite as condensed as the Excel application.

I do this manually in excel every year for my final draft sheet. Take my Projections dominator results and manually insert it into David's VBD spreadsheet format.

 

[Tau837]

4. If you need to free up some time/manpower from your staff to manage the increased attention on projections, perhaps you could eliminate a few of the articles published on FBG. Now, I really enjoy many of the articles, but there are just far too many of them. Who has the time to read all of it?

On this one, I would be in favor of FBG taking a step back and reducing the articles content by 50% but focusing on delivering better quality in the remaining 50% (and on meeting deadlines for that 50%). They could use any number of methods to determine what to cut: let the staff choose, let the Shark Pool choose, use page views, or some combination.

 

[ImTheScientist]

 

[Ignoratio Elenchi]

Not sure I agree on the injury prediction comment. There are certain guys I predict will miss a few games and will continue to do so until they prove they can withstand a full season. To me, it's worthless to project 16 games of production for McFadden for example, because I'm more sure than not that he won't play all 16 games. He never has. Until he plays a full season and shows he's either changed his running style, workout regimin, or whatever it is that's making him get so banged up, I'll continue to do so.Predicting the exact number of games he'll miss is another question and is very debatable. I usually just go by the last few years and average them. I dont think about it too much, because I usually skip guys with the really big injury red flags anyway.

There's a meaningful difference between a guy who is projected to score 20 points a game for 12 games, and then miss the other four due to injury, and a guy who is projected to score 15 points a game for 16 games. Their season-total projections are the same but I'd much rather have the first guy than the second guy. Give us the per-game projections (i.e. project every player for a full 16 games) and then let me adjust for perceived injury risk. OP is correct - when you build "missed games" into the projections you're really just making them less useful.

 

[saintfool]

What was Jo's perspective?

Have a nice season, guy...

 

[ComfortablyNumb]

What was Jo's perspective?

"Joe's perspective" was his response to my earlier e-mail where he explained the challenges of logistics and coordinating multiple inputs and a lot of behind the scene work. And I'd noticed my comfortably numb alter ego years ago. Maybe one of us should have copyrighted it?My main issue here is that the primary tool that I use is the VBD Excel app which uses only Dodds' projections. I think it's their best tool, but that it could be so much better if we could plug-in other projections (preferably higher-fidelity consensus projections). And I have tried to use PD in the past but couldn't import those projections into the VBD Excel tool.

 

[ComfortablyNumb]

Not sure I agree on the injury prediction comment. There are certain guys I predict will miss a few games and will continue to do so until they prove they can withstand a full season. To me, it's worthless to project 16 games of production for McFadden for example, because I'm more sure than not that he won't play all 16 games. He never has. Until he plays a full season and shows he's either changed his running style, workout regimin, or whatever it is that's making him get so banged up, I'll continue to do so.Predicting the exact number of games he'll miss is another question and is very debatable. I usually just go by the last few years and average them. I dont think about it too much, because I usually skip guys with the really big injury red flags anyway.

There's a meaningful difference between a guy who is projected to score 20 points a game for 12 games, and then miss the other four due to injury, and a guy who is projected to score 15 points a game for 16 games. Their season-total projections are the same but I'd much rather have the first guy than the second guy. Give us the per-game projections (i.e. project every player for a full 16 games) and then let me adjust for perceived injury risk. OP is correct - when you build "missed games" into the projections you're really just making them less useful.

 

[Dinsy Ejotuz]

There's a meaningful difference between a guy who is projected to score 20 points a game for 12 games, and then miss the other four due to injury, and a guy who is projected to score 15 points a game for 16 games. Their season-total projections are the same but I'd much rather have the first guy than the second guy. Give us the per-game projections (i.e. project every player for a full 16 games) and then let me adjust for perceived injury risk. OP is correct - when you build "missed games" into the projections you're really just making them less useful.

I agree with your comment, especially when different staffers handle this differently. But at least they give you the number of games in the projections. So it's pretty easy to adjust the totals to reflect 16 games. And once you get a spreadsheet set up you can drop new projections in and the work is done. Really not too bad.

 

[remote controller]

I sent the following e-mail to Joe. Does it make sense to anybody else but me?Hi Joe, Thanks for your reply; I appreciate your perspective. I'll try again to offer some purely constructive feedback, not because I enjoy argument or debate, but because I'd like to help my favorite resource, FBG, to become even better and more useful. And because, as an engineer, I just can't help trying to fix/improve things... Surely you remember the movie City Slickers, with Billy Crystal and Jack Palance. You may recall one of the key plot lines was Jack Palance trying to teach Billy Crystal about "The ONE THING" that matters most? Well, in fantasy football, the "One Thing" is...... Projections. It's the basis and bottom line for everything leading up to the draft. Nothing else really matters; without good projections, even awesome tools like the VBD app or Draft Dominator are practically useless. Since nobody has a foolproof crystal ball (not even Dodds), I believe FBG should do everything possible to maximize the reliability of your projections. A few of my suggestions: 1. Stop attempting to project the # of games played by a player. It's simply ludicrous to think you can predict whether Vick, McFadden, or Austin will play 12, 14, or any specific number of games. It would be MUCH more useful to project all players as if they're playing 16 games. We know who the injury/suspension risks are and will adjust our draft strategies accordingly. Including speculation on games missed severely skews the projections and makes comparison MORE DIFFICULT. 2. Again, I will state that a consensus projection which averages the projections of several experts yields a statistically much more reliable result. It's been proven/demonstrated many times with academic studies. With all of the excellent staffers at your disposal, you could also divide the labor. Have 3-4 guys focus on QB's and TE's. A different group could focus on RB's, while a third group could handle WR's. You could let a rookie or two address Kickers and Defenses, as well. Then, MAKE these CONSENSUS Projections available for use in the VBD App and Draft Dominator. 3. Another under-utilized resource is the knowledgable membership of the Shark Pool. I know you already track Shark Pool projections for Spotlight players. Why not compile these Shark Pool consensus projections and make them available to plug into the VBD and DD, as well? Maybe you'd need to include a filtering review or algorithm to eliminate the handful of senseless projections or outliers which could skew the projections? I know I'd REALLY like to see/use some compiled Shark Pool consensus projections, and that's a free resource for you to tap. 4. If you need to free up some time/manpower from your staff to manage the increased attention on projections, perhaps you could eliminate a few of the articles published on FBG. Now, I really enjoy many of the articles, but there are just far too many of them. Who has the time to read all of it? 5. I'm not aware of any type of market research done by FBG. Do you periodically survey your paying customers about what they like or don't like or wish to improve(you probably do, but I'm just ignorant of it)? Thanks for listening,Bob

your name is BOB........huhu..huhuhuhuhuhu

 

[Andrew]

4. If you need to free up some time/manpower from your staff to manage the increased attention on projections, perhaps you could eliminate a few of the articles published on FBG. Now, I really enjoy many of the articles, but there are just far too many of them. Who has the time to read all of it?

I think some better organization of the articles page would help. Categories, tags, etc. Why can't there be a "dynasty" tag for articles, or a "by committee" tag, or any number of tags.One thing I'd love would be a listing of articles by author. Sometimes I'm reading an article and I want to see a previous article by the author. If I could click on their name at the top and it sent me to a page with all their articles, that'd be great.Another thing that would help would be, if you have an article that is one in a series, put a list of links to the other articles in the sidebar.I guess in general I feel like more linking would help me. Make it more like amazon than craigslist (if you want a terrible comparison off the top of my head.)

 

[Da Guru]

Projections are good fodder but in general they are basically worthless. Projections are like the commodity market..there are few sure things, you are going to be right on a couple of longshots and you are going to be wrong on a couple of longshots.

 

[kutta]

Not sure I agree on the injury prediction comment. There are certain guys I predict will miss a few games and will continue to do so until they prove they can withstand a full season. To me, it's worthless to project 16 games of production for McFadden for example, because I'm more sure than not that he won't play all 16 games. He never has. Until he plays a full season and shows he's either changed his running style, workout regimin, or whatever it is that's making him get so banged up, I'll continue to do so.Predicting the exact number of games he'll miss is another question and is very debatable. I usually just go by the last few years and average them. I dont think about it too much, because I usually skip guys with the really big injury red flags anyway.

There's a meaningful difference between a guy who is projected to score 20 points a game for 12 games, and then miss the other four due to injury, and a guy who is projected to score 15 points a game for 16 games. Their season-total projections are the same but I'd much rather have the first guy than the second guy. Give us the per-game projections (i.e. project every player for a full 16 games) and then let me adjust for perceived injury risk. OP is correct - when you build "missed games" into the projections you're really just making them less useful.

One way to do this would be to offer the projections on a points per game basis instead of a points per year basis. The user could then decide how many games he wants to project the guy plays.

 

[comfortably numb]

Do you people actually read all these articles?

How much of the site content (not including forums) do you guys use?

 

[ComfortablyNumb]

Projections are good fodder but in general they are basically worthless. Projections are like the commodity market..there are few sure things, you are going to be right on a couple of longshots and you are going to be wrong on a couple of longshots.

I know projections are never completely accurate, but to say they are worthless is to basically say draft rankings, VBD, etc is worthless, as well... If that's true, why even have FBG (other than this cool forum)?

 

[[icon]]

Once the David Dodds butt sniffers have finished having their hissy fit after reading your letter, expect a headhunt on you, Numb.

Future wannabe staffers are going to have a field day defending their leader.

Nice tone to this post.... but Dodds HAS been the most accurate Fantasy Football Projector in the industry over the last 2 years.... 2011: #5 overall (out of 84)

2010: #2 overall (out of 34)

Independent auditor Fantasypros.com actually had this to say about Dodds:

"Dodds, meanwhile, has now finished in the top 5 in four out of the five accuracy competitions we’ve held. That’s a claim no other expert can make."

OOPS, FACTS.

 

[Sarnoff]

Projections are good fodder but in general they are basically worthless. Projections are like the commodity market..there are few sure things, you are going to be right on a couple of longshots and you are going to be wrong on a couple of longshots.

I know projections are never completely accurate, but to say they are worthless is to basically say draft rankings, VBD, etc is worthless, as well... If that's true, why even have FBG (other than this cool forum)?

Bah! Who needs thought put into projections? I'd rather have FootBallGuysTV!!! And a PodCast. It's much more important to push to become a titan of social media than have worthwhile content.

 

[cheese]

Projections are good fodder but in general they are basically worthless. Projections are like the commodity market..there are few sure things, you are going to be right on a couple of longshots and you are going to be wrong on a couple of longshots.

I know projections are never completely accurate, but to say they are worthless is to basically say draft rankings, VBD, etc is worthless, as well... If that's true, why even have FBG (other than this cool forum)?

I'm here for the subscriber contest

 

[Modog814]

What was Jo's perspective?

"Joe's perspective" was his response to my earlier e-mail where he explained the challenges of logistics and coordinating multiple inputs and a lot of behind the scene work. And I'd noticed my comfortably numb alter ego years ago. Maybe one of us should have copyrighted it?My main issue here is that the primary tool that I use is the VBD Excel app which uses only Dodds' projections. I think it's their best tool, but that it could be so much better if we could plug-in other projections (preferably higher-fidelity consensus projections). And I have tried to use PD in the past but couldn't import those projections into the VBD Excel tool.

I have had no problem converting my own projections into VBD Excel format. Take some time to set up a template and it's pretty simple to convert.

 

[FF Ninja]

I don't care much for the projections, but I agree that projecting everyone for 16 games would be best. If you want, you can put an "injury factor" in one column which is a percentage representing the amount of that healthy 16 game projection you expect.

Or just give projections in a per game format. I hate to see a guy lower on the rankings just due to a projected missed game. Silly. But then again, I don't really care. I only really use the projections page to look at the historical stats. I like how RB rushes are broken out. If you look at yahoo's stats it includes QB and WR rushes. This gives a better idea of historical RB useage.

 

[Modog814]

Projections are good fodder but in general they are basically worthless. Projections are like the commodity market..there are few sure things, you are going to be right on a couple of longshots and you are going to be wrong on a couple of longshots.

I know projections are never completely accurate, but to say they are worthless is to basically say draft rankings, VBD, etc is worthless, as well... If that's true, why even have FBG (other than this cool forum)?

I'm here for the subscriber contest

This is probably 90% of the reason I come back each year. I find the articles to not really be that ground breaking and are really just formalized forum posts.

 

[rickyg]

Wait let me make are I get this: subscribers to a fantasy football site are complaining that there is TOO much content to read through?!?!

Try subscribing to draft sharks. Their content should ve short enough for you. I actually canceled my DS subscription and went all in on FBG after I kind the service. I love the content even though I can't get through it all. I pick and choose to read the articles that matter to me.

It's an incredible service. And I am not a joe Bryant butt sniffer I don't even know the guy lol

 

[joffer]

My main issue here is that the primary tool that I use is the VBD Excel app which uses only Dodds' projections. I think it's their best tool, but that it could be so much better if we could plug-in other projections (preferably higher-fidelity consensus projections). And I have tried to use PD in the past but couldn't import those projections into the VBD Excel tool.

why not just use the DD and PD apps instead of the Excel tool?

 

[Soulfly3]

Once the David Dodds butt sniffers have finished having their hissy fit after reading your letter, expect a headhunt on you, Numb.

Future wannabe staffers are going to have a field day defending their leader.

Nice tone to this post.... but Dodds HAS been the most accurate Fantasy Football Projector in the industry over the last 2 years.... 2011: #5 overall (out of 84)

2010: #2 overall (out of 34)

Independent auditor Fantasypros.com actually had this to say about Dodds:

"Dodds, meanwhile, has now finished in the top 5 in four out of the five accuracy competitions we’ve held. That’s a claim no other expert can make."

OOPS, FACTS.

I'm not knockin' Dodds. He provides a great service... many ways they/he can improve the site, but I'm knockin' the sniffers.

 

[flc735]

what i don't like is that rec yards for rb/wr/te's almost never matchup with qb passing yards

thats a great and easy way to check and make sure you are not overprojecting a particular player

also, include an 'others' stat to make these projections more accurate. basically, all the numbers from the unlisted players, like the wr 6 who goes for 68 yards by the end of the season

and ya, keep it to 16 games. i have to divide by 15 and multiply by 16 every time i look at those rankings. let us use our common sense to determine injury risk. attempting to quantify an quantifiable number only leads to less useful results

 

[Modog814]

what i don't like is that rec yards for rb/wr/te's almost never matchup with qb passing yards

thats a great and easy way to check and make sure you are not overprojecting a particular player

also, include an 'others' stat to make these projections more accurate. basically, all the numbers from the unlisted players, like the wr 6 who goes for 68 yards by the end of the seasonand ya, keep it to 16 games. i have to divide by 15 and multiply by 16 every time i look at those rankings. let us use our common sense to determine injury risk. attempting to quantify an quantifiable number only leads to less useful results

I thought they had this if you look for projections by team Houston

Too lazy to take the time to see if it adds up though.