Confused about the Projections (1 Viewer)

[Dentist]

No QB is predicted to have a 300 yard game?

no RB 100 yards on the ground?

no WR 100 yards through the air?



is there something i'm missing here?

 

[Doug Drinen]

No QB is predicted to have a 300 yard game?

no RB 100 yards on the ground?

no WR 100 yards through the air?



is there something i'm missing here?

I think if you asked Dodds, he would certainly project that some QB will throw for 300. But, when he carefully looked at each QB individually, he did not find any specific QB who he felt confident projecting for 300. David looks at each team specifically and makes the best projections he can for that team. He does not try to tweak them to match the league's typical macro patterns.

I hope that helps explain what you're seeing.

 

[Mungo Burrows]

No QB is predicted to have a 300 yard game?no RB 100 yards on the ground?no WR 100 yards through the air?is there something i'm missing here?

Weighted averages based on probabilities. Play the game 50 times and what would that player's average stats be. Or some such statistical nonsense. It's the same reason there are fractional touchdowns.They're less interested on taking wild guesses at who will have a blow-up game, than they are crunching numbers to come up with more reliable rankings over time. Whether they succeed or not is up for debate, but I'm pretty sure that's the approach.

 

[Dentist]

I appreciate the previous responses.

i'm a new subscriber so maybe i just don't understand the genius of it.

But i mean... we're pimping Arizona as thumping the niners.. right? well why can't warner get the 300 yards, and one of the AZ receivers hit 100..

this just seems very vanilla... and I respect dodds work as much as anyone... EVAH

 

[Scarpati]

But i mean... we're pimping Arizona as thumping the niners.. right? well why can't warner get the 300 yards, and one of the AZ receivers hit 100..

First, 300 yard games for QBs and 100 yard games for WRs are rare. Sure, those guys are going to get some of those games, but it's hard to say it will definitely happen this week. Even if you thought it was more likely than not that one WR would get 100 yards, you might still think it was a 35% chance for Fitzgerald and a 35% chance for Boldin.Also, if AZ gets hot early, AZ may start running the ball. They have a re-engineered running game and it wouldn't surprise me at all to see them work hard to dispell the notion that they're a one-dimensional team. Warner may very well throw for more than 150 yards in the first half, but will he keep throwing after halftime?Let's try an experiment. Each week, you (Dentist) post in this thread the name of one QB you think will get 300 yards and one WR you think will get 100 yards. Pick anybody you want. After 16 weeks, we'll see what the average number of yards is and how it compares to the projections. (I'm not trying to show you up; it's hard to do this. But it will be an interesting and instructive experiment.)

 

[Chase Stuart]

But i mean... we're pimping Arizona as thumping the niners.. right? well why can't warner get the 300 yards, and one of the AZ receivers hit 100..

First, 300 yard games for QBs and 100 yard games for WRs are rare. Sure, those guys are going to get some of those games, but it's hard to say it will definitely happen this week. Even if you thought it was more likely than not that one WR would get 100 yards, you might still think it was a 35% chance for Fitzgerald and a 35% chance for Boldin.Also, if AZ gets hot early, AZ may start running the ball. They have a re-engineered running game and it wouldn't surprise me at all to see them work hard to dispell the notion that they're a one-dimensional team. Warner may very well throw for more than 150 yards in the first half, but will he keep throwing after halftime?Let's try an experiment. Each week, you (Dentist) post in this thread the name of one QB you think will get 300 yards and one WR you think will get 100 yards. Pick anybody you want. After 16 weeks, we'll see what the average number of yards is and how it compares to the projections. (I'm not trying to show you up; it's hard to do this. But it will be an interesting and instructive experiment.)

 

[Dentist]

whoa, whoa, whoa

I'm not hating on the projections.... I'm not saying i'm better than anyone.

I know this is not easy.

However, I WILL guarantee you that there are at least 1 300 yard QB, 2-3 100 yard rushers, and probably 2-3 100 yard receivers...

Why not go out on a limb a little and really make a big prediction.

But 300 yard games by QB's and 100 yard games from WR's aren't THAT rare.

But I'll bite.. however, i'm not the expert, I just pay for the service: (if you have a dental question and want a expert opinion, PM me)

Warner 300 yards passing

LJ 100 yards rushing

Moss 100 yards receiving

 

[LHUCKS]

I'll take Warner and Chad Johnson to break the 300/100 yard barriers respectively.

 

[Scarpati]

whoa, whoa, whoaI'm not hating on the projections.... I'm not saying i'm better than anyone.I know this is not easy.

No, I know you aren't. No hard feelings here!

However, I WILL guarantee you that there are at least 1 300 yard QB, 2-3 100 yard rushers, and probably 2-3 100 yard receivers...

No doubt, but that's not the question. The question is whether anyone can predict ahead of time who those will be.

Why not go out on a limb a little and really make a big prediction.

That's the spirit!

But I'll bite.. however, i'm not the expert, I just pay for the service: (if you have a dental question and want a expert opinion, PM me)Warner 300 yards passingLJ 100 yards rushingMoss 100 yards receiving

Great! You keep predicting and I'll report on the results. We'll use as a comparison the last projections released by FBG.com on Sunday morning.By the way, everyone, I'm just going to focus on Dentist's projections. I realize that we could get a dog pile, with a whole bunch of people predicting. If we had 100 predictors, a few of them (by luck if nothing else) will hit it big. I assume Dentist is just as good a predictor as anyone else (and if nothing else, he's welcome to use the projections from FBG.com when he makes his predictions) but it's important that we have just one person making the projections in order to have a fair comparison.

 

[joffer]

individual projections vs. group projections

 

[Wheelhouse]

Before we go adding our own 300 yd QBs, 100 yd RB and 100 yd WR - why don't we create a new thread for it?

 

[Scarpati]

Moss 100 yards receiving

There are several WRs named Moss, so please specify which one you have in mind.Thanks!

 

[Dt's Mules]

Moss 100 yards receiving

There are several WRs named Moss, so please specify which one you have in mind.Thanks!

he meant randy

 

[Raiders Fan]

LJ always gets 100+ yards when he starts.. Who else can say that??

 

[secretid]

There's really no reason to go out on a limb at this point and I'd be interested in hearing why you'd like them to do it. Just because the numbers are fun to look at?

There's something to be said for going out on a limb pre-draft when you're looking for guys to break out and there's a value balance (projected pts/potential vs draft position). At this point it's pretty much who to start/who to pick-up and the safer projections are exactly what you want as they best answer those questions.

P.S. - This reminds me of those draft shows that suggest picking at least one 12 and 13 seed every year because one pretty much always advances. If I'm trying to get a perfect bracket then I might do this (probably not), but outside of that if I take a shot at a 12 seed advancing there's a really good chance I'll be wrong and then I missed that one AND the 12 seed that actually advanced. I'll pick 2 12s or no 12s based on who is actually most likely to win the game.

 

[Uruk-Hai]

This is similar to the discussion on AVT. We know some RB is going to finish at #1, but none are 100% sure to do so. Maurile does a great job of explaining this in detail; I don't know if any of the threads are still around, but the same theory applies to Dentist's question.

 

[Dentist]

randy.. sorry

 

[RoosterScott]

? The best was when they projected Cade McNown to rip it up as a top 15 QB.

 

[foxman]

I have wanted to do a fantasy league based on this. The guys on NFL live do it.

I'll take:

P.Manning for 300 (I aint sold on the Giants Defense)

L. Tomlinson for 100

M. Harrison for 100

 

[Wingnut]

Before we go adding our own 300 yd QBs, 100 yd RB and 100 yd WR - why don't we create a new thread for it?

And you havent done this yet because.............................?Hey wait, I could have done it instead of badgering YOU about it.I give myself a

 

[Bri]

But i mean... we're pimping Arizona as thumping the niners.. right? well why can't warner get the 300 yards, and one of the AZ receivers hit 100..

First, 300 yard games for QBs and 100 yard games for WRs are rare. Sure, those guys are going to get some of those games, but it's hard to say it will definitely happen this week. Even if you thought it was more likely than not that one WR would get 100 yards, you might still think it was a 35% chance for Fitzgerald and a 35% chance for Boldin.Also, if AZ gets hot early, AZ may start running the ball. They have a re-engineered running game and it wouldn't surprise me at all to see them work hard to dispell the notion that they're a one-dimensional team. Warner may very well throw for more than 150 yards in the first half, but will he keep throwing after halftime?Let's try an experiment. Each week, you (Dentist) post in this thread the name of one QB you think will get 300 yards and one WR you think will get 100 yards. Pick anybody you want. After 16 weeks, we'll see what the average number of yards is and how it compares to the projections. (I'm not trying to show you up; it's hard to do this. But it will be an interesting and instructive experiment.)

this is weak, in a way I agree with the original poster

 

[Bri]

I think if you asked Dodds, he would certainly project that some QB will throw for 300. But, when he carefully looked at each QB individually, he did not find any specific QB who he felt confident projecting for 300.

Doug,I don't understand this. Are you saying Dodds would say his projections are wrong?

 

[Chase Stuart]

I think if you asked Dodds, he would certainly project that some QB will throw for 300. But, when he carefully looked at each QB individually, he did not find any specific QB who he felt confident projecting for 300.

Doug,I don't understand this. Are you saying Dodds would say his projections are wrong?

If you throw a die, you would certainly project that some number would come up -- 1, 2, 3, 4, 5 or 6. But, when you look at each number individually, you would not find any specific number that you'd feel confident in projecting would come up.You'd project that it won't be a 1, won't be a 2, won't be a 3, won't be a 4, won't be a 5 and won't be a 6. (Using "won't be" as a proxy for more likely than not that it won't occur.) But you'd project that there would be one number that comes up.

 

[Bri]

But i mean... we're pimping Arizona as thumping the niners.. right? well why can't warner get the 300 yards, and one of the AZ receivers hit 100..

First, 300 yard games for QBs and 100 yard games for WRs are rare. Sure, those guys are going to get some of those games, but it's hard to say it will definitely happen this week. Even if you thought it was more likely than not that one WR would get 100 yards, you might still think it was a 35% chance for Fitzgerald and a 35% chance for Boldin.

huh?rare in relation to what?who is the other 30%? I don't understand that 35%*2 logic

Also, if AZ gets hot early, AZ may start running the ball. They have a re-engineered running game and it wouldn't surprise me at all to see them work hard to dispell the notion that they're a one-dimensional team. Warner may very well throw for more than 150 yards in the first half, but will he keep throwing after halftime?

Denny Green is not new. You'll be hard pressed to find a game where his Vikings or Cards pounded the ball and ignored the pass. We have a precedent to look at.

Let's try an experiment. Each week, you (Dentist) post in this thread the name of one QB you think will get 300 yards and one WR you think will get 100 yards. Pick anybody you want. After 16 weeks, we'll see what the average number of yards is and how it compares to the projections. (I'm not trying to show you up; it's hard to do this. But it will be an interesting and instructive experiment.)

He's not Dodds. If they were comparable he'd be on staff or have his own site. Dodds makes a living off of his projections and....I don't see how a comparison between the two would prove a thing

 

[Chase Stuart]

But i mean... we're pimping Arizona as thumping the niners.. right? well why can't warner get the 300 yards, and one of the AZ receivers hit 100..

First, 300 yard games for QBs and 100 yard games for WRs are rare. Sure, those guys are going to get some of those games, but it's hard to say it will definitely happen this week. Even if you thought it was more likely than not that one WR would get 100 yards, you might still think it was a 35% chance for Fitzgerald and a 35% chance for Boldin.

huh?rare in relation to what?who is the other 30%? I don't understand that 35%*2 logic

There is no other 30%. He's saying there's a good chance -- but less likely than more likely -- that Fitzgerald gets 100 yards. Or that Boldin gets 100 yards. It doesn't help anybody to make projections on a hunch that player X will have 100 yards. Or to give player X 100 yards so you're seen as "going out on a limb." The best projections are the ones that will be most accurate most often. I think Dodds' system is excellent for that.Some WR will probably get 150 yards this week, but there's no sense in projecting that. It's unlikely to happen to any one WR.

 

[Bri]

I think if you asked Dodds, he would certainly project that some QB will throw for 300. But, when he carefully looked at each QB individually, he did not find any specific QB who he felt confident projecting for 300.

Doug,I don't understand this. Are you saying Dodds would say his projections are wrong?

If you throw a die, you would certainly project that some number would come up -- 1, 2, 3, 4, 5 or 6. But, when you look at each number individually, you would not find any specific number that you'd feel confident in projecting would come up.You'd project that it won't be a 1, won't be a 2, won't be a 3, won't be a 4, won't be a 5 and won't be a 6. (Using "won't be" as a proxy for more likely than not that it won't occur.) But you'd project that there would be one number that comes up.

Note-I don't do projections preseason like alot of folks do here to really specific degrees taking in rushing average history and yards per catch history....some guys rrrreally go into detail. I'm genuinely asking here, not bashingThat said, I understand that but I don't understand how anyone(skip Dodds, doesn't need to be him specifically) can say "my projections are good" and then say "but someone will do way better than I predicted". Isn't that the nature of you guys making projections? Is there some sort of reasoning behind staying "under"? like so that a goal is reached?

Nother Q:

If someone projects 250 yards and the QB gets 200 yards. Was that a good projection?

If someone projects 250 yards and the QB gets 300 yards. Was that a good projection?

I'm guessing the latter is OK in your opinion. Why? if they're both 50 off?

 

[Chase Stuart]

I think if you asked Dodds, he would certainly project that some QB will throw for 300. But, when he carefully looked at each QB individually, he did not find any specific QB who he felt confident projecting for 300.

Doug,I don't understand this. Are you saying Dodds would say his projections are wrong?

If you throw a die, you would certainly project that some number would come up -- 1, 2, 3, 4, 5 or 6. But, when you look at each number individually, you would not find any specific number that you'd feel confident in projecting would come up.You'd project that it won't be a 1, won't be a 2, won't be a 3, won't be a 4, won't be a 5 and won't be a 6. (Using "won't be" as a proxy for more likely than not that it won't occur.) But you'd project that there would be one number that comes up.

Note-I don't do projections preseason like alot of folks do here to really specific degrees taking in rushing average history and yards per catch history....some guys rrrreally go into detail. I'm genuinely asking here, not bashingThat said, I understand that but I don't understand how anyone(skip Dodds, doesn't need to be him specifically) can say "my projections are good" and then say "but someone will do way better than I predicted". Isn't that the nature of you guys making projections? Is there some sort of reasoning behind staying "under"? like so that a goal is reached?

Nother Q:

If someone projects 250 yards and the QB gets 200 yards. Was that a good projection?

If someone projects 250 yards and the QB gets 300 yards. Was that a good projection?

I'm guessing the latter is OK in your opinion. Why? if they're both 50 off?

There's nothing about being "under". Those projections are equally good.You just notice the "under" stuff because we're talking about the "top" receivers. If we look at the "bottom" QBs, I'm sure some QB will come in this week with less than 157 yards. But Dodds' lowest projecting starting QB is at 157. But Dodds (or anyone) shouldn't project some QB to have 75 passing yards this week, because you're going to miss way more than you hit when you do things like that.

 

[Bri]

But i mean... we're pimping Arizona as thumping the niners.. right? well why can't warner get the 300 yards, and one of the AZ receivers hit 100..

First, 300 yard games for QBs and 100 yard games for WRs are rare. Sure, those guys are going to get some of those games, but it's hard to say it will definitely happen this week. Even if you thought it was more likely than not that one WR would get 100 yards, you might still think it was a 35% chance for Fitzgerald and a 35% chance for Boldin.

huh?rare in relation to what?

who is the other 30%? I don't understand that 35%*2 logic

There is no other 30%. He's saying there's a good chance -- but less likely than more likely -- that Fitzgerald gets 100 yards. Or that Boldin gets 100 yards.

well then, Scarpati, I completely disagree.Tell me what San Fran DB has a prayer of stopping them

here's a link for you

http://www.49ers.com/team/depth_chart.php

I'll hang up and listen

It doesn't help anybody to make projections on a hunch that player X will have 100 yards. Or to give player X 100 yards so you're seen as "going out on a limb." The best projections are the ones that will be most accurate most often. I think Dodds' system is excellent for that.

I would rather hear Dodds "go out on a limb". I expect him to be human and wrong sometimes so ...yeah I'd like to hear his "out on a limb" predictions

 

[The Jerk]

But i mean... we're pimping Arizona as thumping the niners.. right? well why can't warner get the 300 yards, and one of the AZ receivers hit 100..

First, 300 yard games for QBs and 100 yard games for WRs are rare. Sure, those guys are going to get some of those games, but it's hard to say it will definitely happen this week. Even if you thought it was more likely than not that one WR would get 100 yards, you might still think it was a 35% chance for Fitzgerald and a 35% chance for Boldin.

huh?rare in relation to what?who is the other 30%? I don't understand that 35%*2 logic

There is no other 30%. He's saying there's a good chance -- but less likely than more likely -- that Fitzgerald gets 100 yards. Or that Boldin gets 100 yards. It doesn't help anybody to make projections on a hunch that player X will have 100 yards. Or to give player X 100 yards so you're seen as "going out on a limb." The best projections are the ones that will be most accurate most often. I think Dodds' system is excellent for that.Some WR will probably get 150 yards this week, but there's no sense in projecting that. It's unlikely to happen to any one WR.

A different way of trying to explain this is to take the receiver with the highest projected yardage total and suggest that this player is the most likely to go over 100 yards. However, the likelihood of the receiver going over 100 is roughly the same as the likelihood that he finishes under 70 yards and definitely less than 50%. Essentially, I see the projection number as the center of a bell curve. Dodds states what he thinks is the 50th percentile mark, which means that there is equal probability that the player exceeds or falls short of that number.If the game was played 1000 times, this projected "center of the bell curve" number would be the average of all 1000 games. In many games, the receiver would be over 100, and 1 or 2 out of 1000 he may exceed even 200, but that's not at all likely. And of course in a few of those games, the receiver might get less than 20 yards due to bad weather, an injury, etc.There are some boom/bust players who may actually be more likely to finish over 100 yards despite having lower average game totals, but by and large this analogy fits most of the receivers.

 

[Bri]

You just notice the "under" stuff because we're talking about the "top" receivers. If we look at the "bottom" QBs, I'm sure some QB will come in this week with less than 157 yards. But Dodds' lowest projecting starting QB is at 157. But Dodds (or anyone) shouldn't project some QB to have 75 passing yards this week, because you're going to miss way more than you hit when you do things like that.

Why? (Assuming I agree with his thought)If Dodds thinks a QB is going to throw for 157, 75, or just 1 yard, either way I'm not starting him.

 

[Bri]

A different way of trying to explain this is to take the receiver with the highest projected yardage total and suggest that this player is the most likely to go over 100 yards. However, the likelihood of the receiver going over 100 is roughly the same as the likelihood that he finishes under 70 yards and definitely less than 50%. Essentially, I see the projection number as the center of a bell curve. Dodds states what he thinks is the 50th percentile mark, which means that there is equal probability that the player exceeds or falls short of that number.If the game was played 1000 times, this projected "center of the bell curve" number would be the average of all 1000 games. In many games, the receiver would be over 100, and 1 or 2 out of 1000 he may exceed even 200, but that's not at all likely. And of course in a few of those games, the receiver might get less than 20 yards due to bad weather, an injury, etc.There are some boom/bust players who may actually be more likely to finish over 100 yards despite having lower average game totals, but by and large this analogy fits most of the receivers.

If I agreed with this then wouldn't that make some guy named Bill a genius for saying Boldin gets 100 or more against San Fran if he gets that? How about if he predicted it again the next week for Holt against San Fran? How'd that look in your bell curve?This seems pretty elementary to me. Holt and Boldin each played SF twice last year and had over 100 in each game. Take a look, did San Fran upgrade their pass D to stop this? Can Michael Robinson play CB? "right back at ya" what reason do you have for this trend to stop?

 

[The Jerk]

A different way of trying to explain this is to take the receiver with the highest projected yardage total and suggest that this player is the most likely to go over 100 yards. However, the likelihood of the receiver going over 100 is roughly the same as the likelihood that he finishes under 70 yards and definitely less than 50%. Essentially, I see the projection number as the center of a bell curve. Dodds states what he thinks is the 50th percentile mark, which means that there is equal probability that the player exceeds or falls short of that number.If the game was played 1000 times, this projected "center of the bell curve" number would be the average of all 1000 games. In many games, the receiver would be over 100, and 1 or 2 out of 1000 he may exceed even 200, but that's not at all likely. And of course in a few of those games, the receiver might get less than 20 yards due to bad weather, an injury, etc.There are some boom/bust players who may actually be more likely to finish over 100 yards despite having lower average game totals, but by and large this analogy fits most of the receivers.

If I agreed with this then wouldn't that make some guy named Bill a genius for saying Boldin gets 100 or more against San Fran if he gets that? How about if he predicted it again the next week for Holt against San Fran? How'd that look in your bell curve?This seems pretty elementary to me. Holt and Boldin each played SF twice last year and had over 100 in each game. Take a look, did San Fran upgrade their pass D to stop this? Can Michael Robinson play CB? "right back at ya" what reason do you have for this trend to stop?

Holy crap, Bri! This isn't your personal message board. I'm responding to the original poster as much as you and just explaining how I interpret the projections. I'm not a Dodds apologist (and he doesn't need me anyway), and for all I know his method has nothing to do with my interpretation of it.Wide receivers just aren't consistent every outing. Look at Chad Johnson's receiving yards vs. the Browns the past four matchups: 37, 117, 91, 22. Averaging around 67 yards per game, which is a little lower than his per game average. But in any one game, Chad didn't finish within 20 yards of the average for all four. My guess is that most projections had Chad at 90 yards or more for every one of those games. Is it better to project higher and hit on 2 but badly miss on 2 others, or is it better to project with numbers that end up more or less in the middle? I'm not saying anyone did that either, but theoretically I'd argue that hitting closer to the average is better projecting than constantly projecting high numbers and hitting on roughly half.As for the specifics of the Arizona/SF matchup, I'd say there's probably a 70% chance that at least one of Boldin or Fitz goes over 100 yards, but the question is which one. The biggest reason I wouldn't project either to go over 100 individually is because the other could have a monster day. I also expect Arizona will try to force feed the running game especially if they have the lead in the second half as most expect.If you don't want to think about the projections the way I do, that's no problem by me. However, if you can't see ANY merit in that approach, then you're probably missing the value of the projections done in this fashion. Any number of other sites will project 100 yards for certain players more often than Dodds, but I'm not sure they are any better at producing relative differences between players.

 

[JayMan]

But i mean... we're pimping Arizona as thumping the niners.. right? well why can't warner get the 300 yards, and one of the AZ receivers hit 100..

First, 300 yard games for QBs and 100 yard games for WRs are rare. Sure, those guys are going to get some of those games, but it's hard to say it will definitely happen this week. Even if you thought it was more likely than not that one WR would get 100 yards, you might still think it was a 35% chance for Fitzgerald and a 35% chance for Boldin.

huh?rare in relation to what?

who is the other 30%? I don't understand that 35%*2 logic

There is no other 30%. He's saying there's a good chance -- but less likely than more likely -- that Fitzgerald gets 100 yards. Or that Boldin gets 100 yards. It doesn't help anybody to make projections on a hunch that player X will have 100 yards. Or to give player X 100 yards so you're seen as "going out on a limb." The best projections are the ones that will be most accurate most often. I think Dodds' system is excellent for that.

Some WR will probably get 150 yards this week, but there's no sense in projecting that. It's unlikely to happen to any one WR.

A different way of trying to explain this is to take the receiver with the highest projected yardage total and suggest that this player is the most likely to go over 100 yards. However, the likelihood of the receiver going over 100 is roughly the same as the likelihood that he finishes under 70 yards and definitely less than 50%. Essentially, I see the projection number as the center of a bell curve. Dodds states what he thinks is the 50th percentile mark, which means that there is equal probability that the player exceeds or falls short of that number.If the game was played 1000 times, this projected "center of the bell curve" number would be the average of all 1000 games. In many games, the receiver would be over 100, and 1 or 2 out of 1000 he may exceed even 200, but that's not at all likely. And of course in a few of those games, the receiver might get less than 20 yards due to bad weather, an injury, etc.

There are some boom/bust players who may actually be more likely to finish over 100 yards despite having lower average game totals, but by and large this analogy fits most of the receivers.

Here we go I respectfully disagree with you and Chase on this one... and here is why:

Last year - there was, on average:

3.8 QBs per week with 300yds game

8.1 RBs per week with 100yds game

9.8 WBs per week with 100yds game

I understand completely that predicting who these 21 players will be is extremely difficult (if not impossible to do)... but to say that we should base our projection on "if that game was played 50 times - Larry Fitzgerald would have a gaussian distribution around 85 yards" - then I predict 85 yards for him is totally wrong because this game will only be played once! (there is no Normal distribution here... it's a "one time event")

No sane person would ever predict a 10/204/2 game to a WR, and I understand that - but these will happen, just ask Plaxico Burress last year...

We know that, on average - every week 10 WRs will get at least a 100yds game... we know that these outlier games will happen... we know that we don't know who will produce them...

But to say, since I don't know who will get them... I don't predict any is totally wrong in my opinion...

To draw an analogy... we know that there are 21 gifts under the Christmas tree... we know that some kids were nicer this year than others (i.e. Boldin getting 8 - 100yds game in '05 and Fitz getting 7)... we know that Santa Claus will give these gifts on sunday... we just don't know who was nice enough to get one - so we say that none will get any?...

My 2 cents...

 

[cracKer]

Weekly projections = yearly projections/weekly SOS ratio (more goes into it, but this is the basis)

Larry Fitzgerald and Anquan Boldin are both projected right around 1250 yards for the year, meaning that they'd have to hit 8% of that this week for 100 yards. The average weekly percentage for them is 6.25% (78 yards).

Last year, the average NFL defense gave up 209.8 passing yards per game. San Francisco allowed 276.7 (24.2% higher than average). Assuming all factors remained the same, Fitz/Boldin's ratios for the 49ers matchup fall a little under 8% (7.8%, or 97 yards). In contrast, Green Bay allowed just 167.5 yards per game a year ago, which is about 20% lower than the league average. Fitz/Boldin's ratios would be at 5% (63 yards).

By the raw 2005 numbers, you should see their weekly projections fall somewhere between 63 and 97 yards, depending on the matchups. Defensive projection/personnel changes, injuries and such will all play a part in the ratio changes, so these numbers will be off, but it does show you how a receiver could not be projected for 100 yards even with the best matchup available.

 

[Marauder]

If you throw a die, you would certainly project that some number would come up -- 1, 2, 3, 4, 5 or 6. But, when you look at each number individually, you would not find any specific number that you'd feel confident in projecting would come up.

You'd project that it won't be a 1, won't be a 2, won't be a 3, won't be a 4, won't be a 5 and won't be a 6. (Using "won't be" as a proxy for more likely than not that it won't occur.) But you'd project that there would be one number that comes up.

But he's not throwing a die here. Individual rushing, receiving and passing yards are not random. It might be cosidered "rare" for the average NFL running back to rush for 100 yards but SA did it 11 times last year, LJ 10 times (with only 9 starts), Portis 9 times, Edge 9 times, Tiki 8 times. For those guys its more the norm than the exception and I'd be willing to bet that all those guys had the majority of their big games based on predictable factors (against weak rushing defenses, at home, etc.). Dodds gets paid to do projections and is presumably an expert in the field. I don't think its too much to ask for him to use his expertise to make some predictions about which of the elite players will go off in a given week. Sticking within the statistical norm is the easy way out.

 

[Uruk-Hai]

But i mean... we're pimping Arizona as thumping the niners.. right? well why can't warner get the 300 yards, and one of the AZ receivers hit 100..

First, 300 yard games for QBs and 100 yard games for WRs are rare. Sure, those guys are going to get some of those games, but it's hard to say it will definitely happen this week. Even if you thought it was more likely than not that one WR would get 100 yards, you might still think it was a 35% chance for Fitzgerald and a 35% chance for Boldin.

huh?rare in relation to what?

who is the other 30%? I don't understand that 35%*2 logic

There is no other 30%. He's saying there's a good chance -- but less likely than more likely -- that Fitzgerald gets 100 yards. Or that Boldin gets 100 yards. It doesn't help anybody to make projections on a hunch that player X will have 100 yards. Or to give player X 100 yards so you're seen as "going out on a limb." The best projections are the ones that will be most accurate most often. I think Dodds' system is excellent for that.

Some WR will probably get 150 yards this week, but there's no sense in projecting that. It's unlikely to happen to any one WR.

A different way of trying to explain this is to take the receiver with the highest projected yardage total and suggest that this player is the most likely to go over 100 yards. However, the likelihood of the receiver going over 100 is roughly the same as the likelihood that he finishes under 70 yards and definitely less than 50%. Essentially, I see the projection number as the center of a bell curve. Dodds states what he thinks is the 50th percentile mark, which means that there is equal probability that the player exceeds or falls short of that number.If the game was played 1000 times, this projected "center of the bell curve" number would be the average of all 1000 games. In many games, the receiver would be over 100, and 1 or 2 out of 1000 he may exceed even 200, but that's not at all likely. And of course in a few of those games, the receiver might get less than 20 yards due to bad weather, an injury, etc.

There are some boom/bust players who may actually be more likely to finish over 100 yards despite having lower average game totals, but by and large this analogy fits most of the receivers.

Here we go I respectfully disagree with you and Chase on this one... and here is why:

Last year - there was, on average:

3.8 QBs per week with 300yds game

8.1 RBs per week with 100yds game

9.8 WBs per week with 100yds game

I understand completely that predicting who these 21 players will be is extremely difficult (if not impossible to do)... but to say that we should base our projection on "if that game was played 50 times - Larry Fitzgerald would have a gaussian distribution around 85 yards" - then I predict 85 yards for him is totally wrong because this game will only be played once! (there is no Normal distribution here... it's a "one time event")

No sane person would ever predict a 10/204/2 game to a WR, and I understand that - but these will happen, just ask Plaxico Burress last year...

We know that, on average - every week 10 WRs will get at least a 100yds game... we know that these outlier games will happen... we know that we don't know who will produce them...

But to say, since I don't know who will get them... I don't predict any is totally wrong in my opinion...

To draw an analogy... we know that there are 21 gifts under the Christmas tree... we know that some kids were nicer this year than others (i.e. Boldin getting 8 - 100yds game in '05 and Fitz getting 7)... we know that Santa Claus will give these gifts on sunday... we just don't know who was nice enough to get one - so we say that none will get any?...

My 2 cents...

No, but how much would you be willing to bet that a specific kid gets the nicest gift? For myself, I'd be willing to bet very little on any one kid. I think that those who bet their bankrolls every Christmas on one kid aren't going to be around very long.

 

[The Jerk]

But i mean... we're pimping Arizona as thumping the niners.. right? well why can't warner get the 300 yards, and one of the AZ receivers hit 100..

First, 300 yard games for QBs and 100 yard games for WRs are rare. Sure, those guys are going to get some of those games, but it's hard to say it will definitely happen this week. Even if you thought it was more likely than not that one WR would get 100 yards, you might still think it was a 35% chance for Fitzgerald and a 35% chance for Boldin.

huh?rare in relation to what?

who is the other 30%? I don't understand that 35%*2 logic

There is no other 30%. He's saying there's a good chance -- but less likely than more likely -- that Fitzgerald gets 100 yards. Or that Boldin gets 100 yards. It doesn't help anybody to make projections on a hunch that player X will have 100 yards. Or to give player X 100 yards so you're seen as "going out on a limb." The best projections are the ones that will be most accurate most often. I think Dodds' system is excellent for that.

Some WR will probably get 150 yards this week, but there's no sense in projecting that. It's unlikely to happen to any one WR.

A different way of trying to explain this is to take the receiver with the highest projected yardage total and suggest that this player is the most likely to go over 100 yards. However, the likelihood of the receiver going over 100 is roughly the same as the likelihood that he finishes under 70 yards and definitely less than 50%. Essentially, I see the projection number as the center of a bell curve. Dodds states what he thinks is the 50th percentile mark, which means that there is equal probability that the player exceeds or falls short of that number.If the game was played 1000 times, this projected "center of the bell curve" number would be the average of all 1000 games. In many games, the receiver would be over 100, and 1 or 2 out of 1000 he may exceed even 200, but that's not at all likely. And of course in a few of those games, the receiver might get less than 20 yards due to bad weather, an injury, etc.

There are some boom/bust players who may actually be more likely to finish over 100 yards despite having lower average game totals, but by and large this analogy fits most of the receivers.

Here we go I respectfully disagree with you and Chase on this one... and here is why:

Last year - there was, on average:

3.8 QBs per week with 300yds game

8.1 RBs per week with 100yds game

9.8 WBs per week with 100yds game

I understand completely that predicting who these 21 players will be is extremely difficult (if not impossible to do)... but to say that we should base our projection on "if that game was played 50 times - Larry Fitzgerald would have a gaussian distribution around 85 yards" - then I predict 85 yards for him is totally wrong because this game will only be played once! (there is no Normal distribution here... it's a "one time event")

No sane person would ever predict a 10/204/2 game to a WR, and I understand that - but these will happen, just ask Plaxico Burress last year...

We know that, on average - every week 10 WRs will get at least a 100yds game... we know that these outlier games will happen... we know that we don't know who will produce them...

But to say, since I don't know who will get them... I don't predict any is totally wrong in my opinion...

To draw an analogy... we know that there are 21 gifts under the Christmas tree... we know that some kids were nicer this year than others (i.e. Boldin getting 8 - 100yds game in '05 and Fitz getting 7)... we know that Santa Claus will give these gifts on sunday... we just don't know who was nice enough to get one - so we say that none will get any?...

My 2 cents...

You're the mathematician, right? Why don't you run an error analysis on projections focusing on "average" weekly output vs. projections attempting to define exactly which WRs hit the 100 yard mark (i.e., exceed their season average) each week. I'm always open to learning something new. My expectation is that the latter method will have much more error, but I could be wrong. Until then, however, my need for projections is met by seeing the separation between players, not the psychic hotline telling me which typically 20-25 WR will suddenly have a big game. Like some meteorologists who call for the big storm five times before one finally hits, there will be some hits, but I expect many more misses.

 

[Bri]

A different way of trying to explain this is to take the receiver with the highest projected yardage total and suggest that this player is the most likely to go over 100 yards. However, the likelihood of the receiver going over 100 is roughly the same as the likelihood that he finishes under 70 yards and definitely less than 50%. Essentially, I see the projection number as the center of a bell curve. Dodds states what he thinks is the 50th percentile mark, which means that there is equal probability that the player exceeds or falls short of that number.If the game was played 1000 times, this projected "center of the bell curve" number would be the average of all 1000 games. In many games, the receiver would be over 100, and 1 or 2 out of 1000 he may exceed even 200, but that's not at all likely. And of course in a few of those games, the receiver might get less than 20 yards due to bad weather, an injury, etc.There are some boom/bust players who may actually be more likely to finish over 100 yards despite having lower average game totals, but by and large this analogy fits most of the receivers.

If I agreed with this then wouldn't that make some guy named Bill a genius for saying Boldin gets 100 or more against San Fran if he gets that? How about if he predicted it again the next week for Holt against San Fran? How'd that look in your bell curve?This seems pretty elementary to me. Holt and Boldin each played SF twice last year and had over 100 in each game. Take a look, did San Fran upgrade their pass D to stop this? Can Michael Robinson play CB? "right back at ya" what reason do you have for this trend to stop?

Holy crap, Bri! This isn't your personal message board. I'm responding to the original poster as much as you

so you reply to me(too) and I can't reply?personal messageboard?geesh

 

[The Jerk]

If you throw a die, you would certainly project that some number would come up -- 1, 2, 3, 4, 5 or 6. But, when you look at each number individually, you would not find any specific number that you'd feel confident in projecting would come up.

You'd project that it won't be a 1, won't be a 2, won't be a 3, won't be a 4, won't be a 5 and won't be a 6. (Using "won't be" as a proxy for more likely than not that it won't occur.) But you'd project that there would be one number that comes up.

But he's not throwing a die here. Individual rushing, receiving and passing yards are not random. It might be cosidered "rare" for the average NFL running back to rush for 100 yards but SA did it 11 times last year, LJ 10 times (with only 9 starts), Portis 9 times, Edge 9 times, Tiki 8 times. For those guys its more the norm than the exception and I'd be willing to bet that all those guys had the majority of their big games based on predictable factors (against weak rushing defenses, at home, etc.). Dodds gets paid to do projections and is presumably an expert in the field. I don't think its too much to ask for him to use his expertise to make some predictions about which of the elite players will go off in a given week. Sticking within the statistical norm is the easy way out.

Rushing yards are much more predicatbly consistent, and my recollection of 2005 is that Dodds did indeed predict 100+ yards many times for the players listed above. Receiving is more of a crapshoot.

 

[The Jerk]

A different way of trying to explain this is to take the receiver with the highest projected yardage total and suggest that this player is the most likely to go over 100 yards. However, the likelihood of the receiver going over 100 is roughly the same as the likelihood that he finishes under 70 yards and definitely less than 50%. Essentially, I see the projection number as the center of a bell curve. Dodds states what he thinks is the 50th percentile mark, which means that there is equal probability that the player exceeds or falls short of that number.If the game was played 1000 times, this projected "center of the bell curve" number would be the average of all 1000 games. In many games, the receiver would be over 100, and 1 or 2 out of 1000 he may exceed even 200, but that's not at all likely. And of course in a few of those games, the receiver might get less than 20 yards due to bad weather, an injury, etc.There are some boom/bust players who may actually be more likely to finish over 100 yards despite having lower average game totals, but by and large this analogy fits most of the receivers.

If I agreed with this then wouldn't that make some guy named Bill a genius for saying Boldin gets 100 or more against San Fran if he gets that? How about if he predicted it again the next week for Holt against San Fran? How'd that look in your bell curve?This seems pretty elementary to me. Holt and Boldin each played SF twice last year and had over 100 in each game. Take a look, did San Fran upgrade their pass D to stop this? Can Michael Robinson play CB? "right back at ya" what reason do you have for this trend to stop?

Holy crap, Bri! This isn't your personal message board. I'm responding to the original poster as much as you

so you reply to me(too) and I can't reply?personal messageboard?geesh

What part of "I'm responding to the original poster as much as you" did you not understand?For the inferentially impaired... I was suggesting that you might lighten up a little. Care to respond to my FF content as well?

 

[Chase Stuart]

If you throw a die, you would certainly project that some number would come up -- 1, 2, 3, 4, 5 or 6. But, when you look at each number individually, you would not find any specific number that you'd feel confident in projecting would come up.

You'd project that it won't be a 1, won't be a 2, won't be a 3, won't be a 4, won't be a 5 and won't be a 6. (Using "won't be" as a proxy for more likely than not that it won't occur.) But you'd project that there would be one number that comes up.

But he's not throwing a die here. Individual rushing, receiving and passing yards are not random.

That's not true. They are very much random. It's not random whether Tomlinson will beat Blaylock, but there's a significant portion of randomness in whether Tomlinson has 60, 120 or 180 rushing yards.

Dodds gets paid to do projections and is presumably an expert in the field. I don't think its too much to ask for him to use his expertise to make some predictions about which of the elite players will go off in a given week. Sticking within the statistical norm is the easy way out.

Sticking within the statistical norm is not the easy way out. It's the intelligent way. It's extremely easy to write "Kurt Warner, 400 yards."Let's say Dodds thinks the following.

Shaun Alexander has a 10% chance of rushing for 150 yards, a 10% chance of rushing for 120 yards, a 20% chance of rushing for 100 yards, a 30% chance of rushing for 80 yards, and a 30% chance of rushing for 60 yards. Assuming this is true, what would you like Dodds to project Shaun Alexander for?

 

[Chase Stuart]

There's no reason to get hung up on the number 100, either. Dodds has three RBs projected for 97 or more rushing yards.

 

[JayMan]

Weekly projections = yearly projections/weekly SOS ratio (more goes into it, but this is the basis)Larry Fitzgerald and Anquan Boldin are both projected right around 1250 yards for the year, meaning that they'd have to hit 8% of that this week for 100 yards. The average weekly percentage for them is 6.25% (78 yards).Last year, the average NFL defense gave up 209.8 passing yards per game. San Francisco allowed 276.7 (24.2% higher than average). Assuming all factors remained the same, Fitz/Boldin's ratios for the 49ers matchup fall a little under 8% (7.8%, or 97 yards). In contrast, Green Bay allowed just 167.5 yards per game a year ago, which is about 20% lower than the league average. Fitz/Boldin's ratios would be at 5% (63 yards).By the raw 2005 numbers, you should see their weekly projections fall somewhere between 63 and 97 yards, depending on the matchups. Defensive projection/personnel changes, injuries and such will all play a part in the ratio changes, so these numbers will be off, but it does show you how a receiver could not be projected for 100 yards even with the best matchup available.

I understand this thinking - and I can see that this is the only way to have "safe" projections.By "safe" projection I mean - FBG is saying to their subscribers "If you start Anquan Boldin - you should expect him to have between 63 and 97 yards depending on the matchup".Looking into '05 numbers though... you can see that Boldin had between 63 and 97 yards only 5 times in 16 weeks... this is where I would suspect FBG to crunch the numbers even more (for us to Dominate our league!) and see that Boldin had 156 / 116 / 109 / 123 / 29 yards against the 49ers the 5 games he faced them in his career... and should project him to exceed the 80 yards or so he gets on average weekly...My 2 cents...

 

[Chase Stuart]

Weekly projections = yearly projections/weekly SOS ratio (more goes into it, but this is the basis)Larry Fitzgerald and Anquan Boldin are both projected right around 1250 yards for the year, meaning that they'd have to hit 8% of that this week for 100 yards. The average weekly percentage for them is 6.25% (78 yards).Last year, the average NFL defense gave up 209.8 passing yards per game. San Francisco allowed 276.7 (24.2% higher than average). Assuming all factors remained the same, Fitz/Boldin's ratios for the 49ers matchup fall a little under 8% (7.8%, or 97 yards). In contrast, Green Bay allowed just 167.5 yards per game a year ago, which is about 20% lower than the league average. Fitz/Boldin's ratios would be at 5% (63 yards).By the raw 2005 numbers, you should see their weekly projections fall somewhere between 63 and 97 yards, depending on the matchups. Defensive projection/personnel changes, injuries and such will all play a part in the ratio changes, so these numbers will be off, but it does show you how a receiver could not be projected for 100 yards even with the best matchup available.

I understand this thinking - and I can see that this is the only way to have "safe" projections.By "safe" projection I mean - FBG is saying to their subscribers "If you start Anquan Boldin - you should expect him to have between 63 and 97 yards depending on the matchup".Looking into '05 numbers though... you can see that Boldin had between 63 and 97 yards only 5 times in 16 weeks... this is where I would suspect FBG to crunch the numbers even more (for us to Dominate our league!) and see that Boldin had 156 / 116 / 109 / 123 / 29 yards against the 49ers the 5 games he faced them in his career... and should project him to exceed the 80 yards or so he gets on average weekly...My 2 cents...

Your premise would be true only if after FBGs crunched the numbers and saw that a player's performance against Team X in recent years is a better predictor of his performance going forward than a SOS-adjusted projection outlined by cracKer. I haven't seen any proof showing that's the case, though.

 

[joffer]

always an interesting discussion. i like the way Dodds does it. crunch the numbers and give me the most likely, i can use my own hunches

 

[Bri]

Care to respond to my FF content as well?

no, this isn't your personal message board

 

[The Jerk]

Care to respond to my FF content as well?

no, this isn't your personal message board

Touche. But I value your opinion and want to see if I'm misreading what you're saying or if we simply disagree on this topic.

 

[Marauder]

That's not true. They are very much random. It's not random whether Tomlinson will beat Blaylock, but there's a significant portion of randomness in whether Tomlinson has 60, 120 or 180 rushing yards.

I don't agree with that. Tomlinson at home against the Saints is much more likley to rush for 120 than 60. On the road against Pittsburg, he's much more likely to rush for 60 than 120. I don't buy for a second that his big games are random.

 

[The Jerk]

That's not true. They are very much random. It's not random whether Tomlinson will beat Blaylock, but there's a significant portion of randomness in whether Tomlinson has 60, 120 or 180 rushing yards.

I don't agree with that. Tomlinson at home against the Saints is much more likley to rush for 120 than 60. On the road against Pittsburg, he's much more likely to rush for 60 than 120. I don't buy for a second that his big games are random.

So you can accurately predict nearly every time LT is going to have a 3+ TD game?

 

[JayMan]

No, but how much would you be willing to bet that a specific kid gets the nicest gift? For myself, I'd be willing to bet very little on any one kid. I think that those who bet their bankrolls every Christmas on one kid aren't going to be around very long.

Believe me - I'm not telling you that you should bet your FF win this weekend on which "outlier" WR will get 100 yards this week...As I mention - this is an impossible task - and I understand that the way Dodds is projecting the results is meant to be "right most of the time" (as Chase mention - if he think that Fitz has a 10% of getting 160yds / 15% of getting 130yds - and so on... the average is 90 yards)... I just find it odd that these %s don't account for anyone to have a 100yd game since we know that usually, 21 on average will...Believe me, I'm not saying these is a better way of doing projections - far from it - you have to try and be right (i.e. in the range) most of the times - just that it seems unsual to me that none are predicted to get a 100yd game...