What's new
Fantasy Football - Footballguys Forums

This is a sample guest message. Register a free account today to become a member! Once signed in, you'll be able to participate on this site by adding your own topics and posts, as well as connect with other members through your own private inbox!

Target Practice: TD’s and Yardage Points per Target (1 Viewer)

JGalligan

Footballguy
I’m not going to try and hard sell you on why targets are important, don’t worry. If you are doubtful that they can be helpful, or you just don't know what they are exactly, then you can read my previous article right here. I brought up some concerns in the previous version about how the fantasy points per target statistic may have been to general. Those concerns, when looked at closer, did turn out to be quite warranted. So the only other thing to do was start all over but this time with a more specific set of statistics. That also included breaking everything down by year instead of just the totals.

Those specific statistics are touchdowns per target and yardage points per target. Touchdowns per target (or TDs/Tg as it will now be known) are pretty self-explanatory. It’s the amount of TD’s a given player scores for every target they receive.

The other statistic is yardage points per target (YPts/Tg), which is the amount of fantasy points a player accrues with touchdowns (6 points per TD) subtracted. For instance, if Johnny Rocket had 100 fantasy points in 2007 and 6 TD’s, his yardage points for the year would be 64. This isn’t rocket science, people.

To start off, I broke the target stats up by year (since 2002) with the help of the Game Log Dominator. I wanted to get a good gauge of the fluctuations between YPts/Tg and TDs/Tg each year. The only cap for inclusion in each category was a minimum of 40 total targets for the season. Before we get into the analysis, here’s a nice picture of the spreadsheet for your viewing pleasure (you need to click the link as it wouldn't fit in here):

Click here for the spreadsheet picture

Assuming that you don’t have 100/20 vision and couldn’t read that spreadsheet for detail, allow me to break it down for you, as well as suggest that you perhaps see an eye doctor. And thank you, I know I have the ill Excel spreadsheet skills. But anyway, let's begin.

Here’s a breakdown for the amount of players in each year’s list that were NOT on it the year prior:

Yardage points per target

2002-2003: 18 players

2003-2004: 12 players

2004-2005: 12 players

2005-2006: 10 players

2006-2007: 17 players

Touchdowns per target

2002-2003: 17 players

2003-2004: 13 players

2004-2005: 13 players

2005-2006: 12 players

2006-2007: 14 players

I don’t think you need to be a math whiz to come to the conclusion that there’s a lot of WR’s that break into each stats Top 20 list every year. It can likely be attributed to many things. One, the WR position is the most difficult to consistently put up big numbers. This is due to a variety of factors, including but not limited to: Different QB’s, other WR’s, rookies, contract years, pure laziness, injuries, and age. Two, the WR corps on most teams can often be wide open and vary drastically from year to year. Unlike the QB and RB positions, where the starter only needs to worry about 1 or 2 possible replacements, WR’s need to worry about 3 or 4. Studs aside, it’s a very hard position to solidify a guaranteed position in. Especially in the financial landscape the NFL is experiencing right now.

You also may have noticed that both stats have relatively similar changes each year. The TDs/Tg list tends to have more veterans and solid journeyman grace its rankings whereas YPts/Tg is a haven for young, talented WR’s to break out into. The young player then matures and either makes the jump to the TD s/Tg list or finds he is talented enough to gain vaunted, both-list status. The cycle goes on and on. It makes sense, too. What kind of idiot would focus his red zone targets on a rookie unless he either a) Has hardly any quality receivers, or b) Has a Herman Moore in-his-prime-esque rookie talent? I’m not sure, but I know that I wouldn’t want him coaching my team.

On the recently broached topic of both-lists, though, let’s take a look at how many players found themselves onto both the TDs/Tg list and the YPts/Tg list each year:

Wide Receivers on both the TDs/Tg and YPts/Tg lists

2002: 8 players

2003: 12 players

2004: 11 players

2005: 12 players

2006: 13 players

2007: 11 players

Even some of the top WR’s who were consistently on at least one list each year managed to top more than 50% of their time in the league (since 2002) on both. Save for Reggie Wayne, who was on both lists every single year but one. Guy’s an absolute machine.

In the interests of putting what getting on these Top 20 lists means into perspective, here’s the average minimum stat lines by year:

YPts/Tg Minimum Stat Lines Per Year

2002: 80 targets, 93 fantasy points, 690 receiving yards, 4 TD’s

2003: 76 targets, 81 fantasy points, 690 receiving yards, 2 TD’s

2004: 89 targets, 95 fantasy points, 830 receiving yards, 2 TD’s

2005: 70 targets, 77 fantasy points, 710 receiving yards, 1 TD

2006: 70 targets, 87 fantasy points, 630 receiving yards, 4 TD’s

2007: 77 targets, 79 fantasy points, 730 receiving yards, 1 TD

TDs/Tg Minimum Stat Lines Per Year

2002: 70 targets, 71 fantasy points, 470 receiving yards, 4 TD’s

2003: 103 targets, 99 fantasy points, 630 receiving yards, 6 TD’s

2004: 87 targets, 93 fantasy points, 570 receiving yards, 6 TD’s

2005: 109 targets, 110 fantasy points, 680 receiving yards, 7 TD’s

2006: 84 targets, 86 fantasy points, 440 receiving yards, 7 TD’s

2007: 71 targets, 93 fantasy points, 630 receiving yards, 5 TD’s

Aaaand the minimum averages combined over the last six years:

YPts/Tg ‘02-‘07 Top 20 minimum stats average: 77 targets, 85 fantasy points, 730 receiving yards, 2 TD’s

TDs/Tg ’02-’07 Top 20 minimum stats averages: 87 targets, 92 fantasy points, 560 receiving yards, 6 TD’s

What’s that? Those stat lines suck you say? Well, this whole thing isn’t meant for you to somehow find the next Terrell Owens or Randy Moss. It’s meant to give a statistical low point on what you should get if you target the players correctly. It’s the minimum. On average, 11 new players will be on each respective list at the end of next year. That’s 22 possible players to research who could potentially net you a decent amount of fantasy points. They may not be a stud, but they could be a blossoming one. They may not win you a title by themselves, but they could help you in a couple of key spot starts throughout the season that prove critical to your acquisition of the Hefty Aluminum Foil Champions Trophy. See where this is going?

Unlike in the last article, where I tried to Nostradamus every exact fill-in all by myself, let’s take a look at the players who have the potential or lack thereof to get on the lists next year. But first, let’s sum up what type of players will find themselves on either list (excluding the elite WR shoo-ins):

Yardage points per target list

- Often young, inexperienced or both (rookies).

- Usually quick, nimble receivers who are good at gaining yards after the catch.

- Not usually a target of the QB in the red zone.

- Very likely to be a WR2 or WR3.

Touchdown points per target list

- Often experienced veterans and/or solid journeyman wide outs.

- Usually big-bodied possession receivers.

- Often a target in the end zone, but sometimes not the top one.

- Could be a WR1 or WR2 because of their scoring ability and experience.

Now, taking those two lists, you can more easily target WR’s who will fall into one of the two categories. It makes it easier to go and see just how many people you think will fall out of the Top 20, especially since it fluctuates year to year. So it’s not always exactly 11 exclusions.

To get your brain going, here are the 2007 Top 20’s, the mortal locks to stay on them, some possible players who may fall of them and then followed by those who could possibly gain access into the ranks next year:

2007 Top 20 in Yardage Points per Target

1 Santonio Holmes

2 Greg Jennings

3 Joey Galloway

4 Andre Johnson

5 Reggie Wayne

6 Terrell Owens

7 Antwaan Randle El

8 Donte Stallworth

9 Randy Moss

10 Chad Johnson

11 Jerricho Cotchery

12 Calvin Johnson

13 David Patten

14 Brandon Stokley

15 Justin Gage

16 Roddy White

17 Anquan Boldin

18 Donald Driver

19 Patrick Crayton

20 Dennis Northcutt

Mortal Locks To Remain On List:

Reggie Wayne

Terrell Owens

Randy Moss

Santonio Holmes**

** Interesting note: The #1 ranked player on the YPts/Tg list failed to make the list the following year in every year but one. That was Lee Evans in 2004-2005 and he finished 20th the following year (barely). What am I getting at? Well, just that he may not be a ‘mortal’ lock, but since he’s on my fantasy team I’ll throw him on there.

Possible Dropouts:

Antwaan Randle El

Justin Gage

Dennis Northcutt

Joey Galloway

Possible Non-Elite 2008 Additions:

Dwayne Bowe

Desean Jackson

Lee Evans

Reggie Brown

James Jones

Ted Ginn Jr.

Sidney Rice

2007 Top 20 in Touchdowns per Target

1 Randy Moss

2 Greg Jennings

3 Terrell Owens

4 Braylon Edwards

5 Nate Burleson

6 Santonio Holmes

7 Andre Johnson

8 Anquan Boldin

9 Patrick Crayton

10 Plaxico Burress

11 Marques Colston

12 T.J. Houshmandzadeh

13 Brandon Stokley

14 Laveranues Coles

15 Reggie Wayne

16 Hines Ward

17 Joey Galloway

18 Larry Fitzgerald

19 Jerry Porter

20 Wes Welker

Mortal Locks To Remain On List:

Randy Moss

Terrell Owens

Santonio Holmes

Plaxico Burress

Braylon Edwards

Marques Colston

Reggie Wayne

Larry Fitzgerald

Possible Dropouts:

Brandon Stokley

Hines Ward

Jerry Porter

Possible Non-Elite 2008 Additions:

Dwayne Bowe

Roddy White

Jerricho Cotchery

Roy Williams

Lee Evans

Santana Moss

Limas Sweed

The assignment my class, is to pick the 9-10 players that you think will absolutely stay on the list next year. Then, take a look at the following list and pick the difference from 20 as to who you think will break onto the list next year. And one rookie has made the YPts/Tg list in each of the last 3 years, so take that into account if you wish. Have fun and feel free to leave your findings in the replies.

 
Last edited by a moderator:
a little early to put LImas Sweed on that list no?
Eh, I wanted to throw at least one long shot on there -- especially since I didn't make actual predictions. Just threw out some possibilities.
 
Last edited by a moderator:
Interesting that Brandon Marshall is absent from both lists. Thanks for the info.
Yeah, he actually missed out on both just barely in 2007. I didn't think he would fall under the "Non-elite" category so I didn't put him under the possibilities. But I do think he'll make both lists next year.
 
The other thread got too long before I could chime in, so I'll ask my question here:

Why are you putting targets in the denominator and not the numerator? Why not Yards*Targets, or TDs*Targets?

Once again, I know I'm coming in too late to the game here, but what is the goal of using yards/target or TDs/target in Year N-1? To predict what, and whatever that is, why not use Yards*Targets or TDs*Targets?

 
The other thread got too long before I could chime in, so I'll ask my question here:

Why are you putting targets in the denominator and not the numerator? Why not Yards*Targets, or TDs*Targets?

Once again, I know I'm coming in too late to the game here, but what is the goal of using yards/target or TDs/target in Year N-1? To predict what, and whatever that is, why not use Yards*Targets or TDs*Targets?
Well to be completely honest, using TDs or YardagePts*Targets didn't even occur to me. It absolutely makes sense and it changes things significantly, but I think each method would have it's own usefulness. From what I'm seeing, TDs + Ypts*Tg would show who had the most of each. On the other hand, TDs + Ypts/Tg brings into the WR's who didn't necessarily get many targets, but managed to successfully convert them into a significant amount of fantasy points. For instance, Ypts-wise, Brandon Stokley and Justin Gage. Hopefully that made sense. But yeah, I'm not going to lie, I didn't even think of using times targets. It definitely makes sense but I'm still pretty new to the wonderfully complicated realm that is statistics. It didn't even dawn on me to use that.

Anyway, I put together the Top 20 for 2007 for each using that -- here it is:

Ypts*Tg

Chad Johnson

Randy Moss

Reggie Wayne

Larry Fitzgerald

Brandon Marshall

Braylon Edwards

T.J. Houshmandzadeh

Terrell Owens

Derrick Mason

Torry Holt

Wes Welker

Marques Colston

Roddy White

Steve Smith

Bobby Engram

Kevin Curtis

Jerricho Cotchery

Plaxico Burress

Donald Driver

Chris Chambers

TDs*Tg

Randy Moss

Braylon Edwards

Terrell Owens

T.J. Houshmandzadeh

Plaxico Burress

Larry Fitzgerald

Marques Colston

Reggie Wayne

Chad Johnson

Brandon Marshall

Wes Welker

Torry Holt

Steve Smith

Greg Jennings

Anquan Boldin

Nate Burleson

Roddy White

Derrick Mason

Kevin Curtis

Bobby Engram

They're definitely very different. I'll put something together taking a look at the stats yearly with it and post it in here when I'm done.

 
Last edited by a moderator:
The other thread got too long before I could chime in, so I'll ask my question here:Why are you putting targets in the denominator and not the numerator? Why not Yards*Targets, or TDs*Targets?Once again, I know I'm coming in too late to the game here, but what is the goal of using yards/target or TDs/target in Year N-1? To predict what, and whatever that is, why not use Yards*Targets or TDs*Targets?
Well to be completely honest, using TDs or YardagePts*Targets didn't even occur to me. It absolutely makes sense and it changes things significantly, but I think each method would have it's own usefulness. From what I'm seeing, TDs + Ypts*Tg would show who had the most of each. On the other hand, TDs + Ypts/Tg brings into the WR's who didn't necessarily get many targets, but managed to successfully convert them into a significant amount of fantasy points. For instance, Ypts-wise, Brandon Stokley and Justin Gage.
Precisely, they serve two separate purposes. JG's approach lets you highlight players who have been productive relative to the opportunities they were offered, allowing you to find breakout candidates. Chase's approach basically tells you who the most productive WRs were in absolute terms, but divides performance into yds & TDs - it's helpful for identifying guys who might bust vs. guys who will continue to dominate.Great stuff JG. Still need to re-read the analysis and look at the data.
 
Ya, the first result set is nice but magnifies the homerun hitter like Holmes and undervalues weekly consistent contributors like Holt/Marshall who get a ton of targets. In years past Ashlie Lelie would have been atop that list. The 2nd list values the wr's more fairly because it properly rewards players with a ton of targets which at the end of the day are the wr's you want on your team.

 
Banger said:
The 2nd list values the wr's more fairly because it properly rewards players with a ton of targets which at the end of the day are the wr's you want on your team.
True, but you can potentially discover those who are producing with whatever targets they are currently getting, which will eventually mean that their team will give them more targets. It's a constantly revolving cycle.
 
Michael Fox said:
JG's approach lets you highlight players who have been productive relative to the opportunities they were offered, allowing you to find breakout candidates.
True, but you can potentially discover those who are producing with whatever targets they are currently getting, which will eventually mean that their team will give them more targets. It's a constantly revolving cycle.
Both of you are assuming that a guy with 50 receptions on 70 targets is a better breakout candidate than a guy with 45 receptions on 90 targets. It's perfectly reasonable to assume the latter guy is the better breakout candidate.
 
Michael Fox said:
JG's approach lets you highlight players who have been productive relative to the opportunities they were offered, allowing you to find breakout candidates.
True, but you can potentially discover those who are producing with whatever targets they are currently getting, which will eventually mean that their team will give them more targets. It's a constantly revolving cycle.
Both of you are assuming that a guy with 50 receptions on 70 targets is a better breakout candidate than a guy with 45 receptions on 90 targets. It's perfectly reasonable to assume the latter guy is the better breakout candidate.
:goodposting: I was going to post something similar to this. And JGalligan's bolded statement above is not a valid assumption IMO.

 
Michael Fox said:
JG's approach lets you highlight players who have been productive relative to the opportunities they were offered, allowing you to find breakout candidates.
True, but you can potentially discover those who are producing with whatever targets they are currently getting, which will eventually mean that their team will give them more targets. It's a constantly revolving cycle.
Both of you are assuming that a guy with 50 receptions on 70 targets is a better breakout candidate than a guy with 45 receptions on 90 targets. It's perfectly reasonable to assume the latter guy is the better breakout candidate.
:goodposting: I was going to post something similar to this. And JGalligan's bolded statement above is not a valid assumption IMO.
Right. Fantasy owners generally think targets are a *good* thing. If you see in week 1 that your WR2 has 13 targets, you're going to be very happy, even if he only has five receptions. Certainly you'll be happier than if he has 5 receptions on 8 targets. I'm not saying that beyond a shadow of a doubt that yards*target is better than yards/target; it's basically an open question. In my view, though, I'd guess that putting targets in the numerator is better.

 
Michael Fox said:
JG's approach lets you highlight players who have been productive relative to the opportunities they were offered, allowing you to find breakout candidates.
True, but you can potentially discover those who are producing with whatever targets they are currently getting, which will eventually mean that their team will give them more targets. It's a constantly revolving cycle.
Both of you are assuming that a guy with 50 receptions on 70 targets is a better breakout candidate than a guy with 45 receptions on 90 targets. It's perfectly reasonable to assume the latter guy is the better breakout candidate.
:goodposting: I was going to post something similar to this. And JGalligan's bolded statement above is not a valid assumption IMO.
I am not sure that is true. To put it into a real world perspective, JGalligans post is somewhat describing the situation that MJD is in this year (and was in last year). A guy who has not had a ton of targets, relatively speaking, but who has done an awful lot with minimal targets.In the situation that Chase desribes it is basically saying that someone with the larger number of targets is more likely to break out. Real world situation here would be Lendale White, who had almost double the number of carries that Drew had but scored fewer fantasy points (depending on format) than did MJD.

Now I realize that carries does not equal targets, but I think it gets the point across in this situation.

Obviously there are other factors that must be taken into account. In this case, the Chris Johnson factor or the Fred Taylor factor (which have very different impact on each of these guys situations). But if just looking at the numbers, one could argue that LEndale would be up for a better year, simply because he was utilized more. One could also argue that MJD is up for the better year because he has done so much more with the limited number of carries. Either argument is valid, when the external factors are not considered.

 
JGalligan said:
I don’t think you need to be a math whiz to come to the conclusion that there’s a lot of WR’s that break into each stats Top 20 list every year. It can likely be attributed to many things. One, the WR position is the most difficult to consistently put up big numbers. This is due to a variety of factors, including but not limited to: Different QB’s, other WR’s, rookies, contract years, pure laziness, and age. Two, the WR corps on most teams can often be wide open and vary drastically from year to year. Unlike the QB and RB positions, where the starter only needs to worry about 1 or 2 possible replacements, WR’s need to worry about 3 or 4. Studs aside, it’s a very hard position to solidify a guaranteed position in. Especially in the financial landscape the NFL is experiencing right now.
Did you intentionally ignore injuries here? My guess is that at least half of those who drop out of your lists every year do so because of injury, either to themselves or to others (e.g., their starting QBs).
 
Michael Fox said:
JG's approach lets you highlight players who have been productive relative to the opportunities they were offered, allowing you to find breakout candidates.
True, but you can potentially discover those who are producing with whatever targets they are currently getting, which will eventually mean that their team will give them more targets. It's a constantly revolving cycle.
Both of you are assuming that a guy with 50 receptions on 70 targets is a better breakout candidate than a guy with 45 receptions on 90 targets. It's perfectly reasonable to assume the latter guy is the better breakout candidate.
:goodposting: I was going to post something similar to this. And JGalligan's bolded statement above is not a valid assumption IMO.
I am not sure that is true. To put it into a real world perspective, JGalligans post is somewhat describing the situation that MJD is in this year (and was in last year). A guy who has not had a ton of targets, relatively speaking, but who has done an awful lot with minimal targets.In the situation that Chase desribes it is basically saying that someone with the larger number of targets is more likely to break out. Real world situation here would be Lendale White, who had almost double the number of carries that Drew had but scored fewer fantasy points (depending on format) than did MJD.

Now I realize that carries does not equal targets, but I think it gets the point across in this situation.

Obviously there are other factors that must be taken into account. In this case, the Chris Johnson factor or the Fred Taylor factor (which have very different impact on each of these guys situations). But if just looking at the numbers, one could argue that LEndale would be up for a better year, simply because he was utilized more. One could also argue that MJD is up for the better year because he has done so much more with the limited number of carries. Either argument is valid, when the external factors are not considered.
The example you gave is exactly why it is necessary to look at each situation individually rather than relying upon any particular metric.That said, I think there are two possible conclusions from your example. One could be that given the same situations going forward (i.e., Taylor still playing well in JAX, no other good RB in TEN), Lendale would be more likely to improve. I think that would have been a true statement if not for TEN joining the team. Note that "more likely to improve" is not the same as "better" or "breaking out."

The second possible conclusion would be that MJD is more likely to regress, given that he would not stand to get as many touches (again, if the same situations held). Again, that doesn't mean Lendale would necessarily outperform MJD. The point is that the more opportunities a player gets, the more that mitigates risk associated with that player's performance, and relating that to this thread, targets are opportunities.

 
The example you gave is exactly why it is necessary to look at each situation individually rather than relying upon any particular metric.That said, I think there are two possible conclusions from your example. One could be that given the same situations going forward (i.e., Taylor still playing well in JAX, no other good RB in TEN), Lendale would be more likely to improve. I think that would have been a true statement if not for TEN joining the team. Note that "more likely to improve" is not the same as "better" or "breaking out."The second possible conclusion would be that MJD is more likely to regress, given that he would not stand to get as many touches (again, if the same situations held). Again, that doesn't mean Lendale would necessarily outperform MJD. The point is that the more opportunities a player gets, the more that mitigates risk associated with that player's performance, and relating that to this thread, targets are opportunities.
I agree completely. I actually think the two statistics show two very different things. AS you describe the most targets are probably showing you the least amount of risk associated with a player (or the more risk mitigation at the very least). Where as the other stat could potentially be used to show the most potential (biggest risk/reward sort of option). Depending on the type of person you are, either of these could be useful.I personally am a bigger fan of risk mitigation than I am of risk/reward. But many would see the other side as being more valuable.
 
Michael Fox said:
JG's approach lets you highlight players who have been productive relative to the opportunities they were offered, allowing you to find breakout candidates.
True, but you can potentially discover those who are producing with whatever targets they are currently getting, which will eventually mean that their team will give them more targets. It's a constantly revolving cycle.
Both of you are assuming that a guy with 50 receptions on 70 targets is a better breakout candidate than a guy with 45 receptions on 90 targets. It's perfectly reasonable to assume the latter guy is the better breakout candidate.
:lmao: I was going to post something similar to this. And JGalligan's bolded statement above is not a valid assumption IMO.
I am not sure that is true. To put it into a real world perspective, JGalligans post is somewhat describing the situation that MJD is in this year (and was in last year). A guy who has not had a ton of targets, relatively speaking, but who has done an awful lot with minimal targets.In the situation that Chase desribes it is basically saying that someone with the larger number of targets is more likely to break out. Real world situation here would be Lendale White, who had almost double the number of carries that Drew had but scored fewer fantasy points (depending on format) than did MJD.

Now I realize that carries does not equal targets, but I think it gets the point across in this situation.

Obviously there are other factors that must be taken into account. In this case, the Chris Johnson factor or the Fred Taylor factor (which have very different impact on each of these guys situations). But if just looking at the numbers, one could argue that LEndale would be up for a better year, simply because he was utilized more. One could also argue that MJD is up for the better year because he has done so much more with the limited number of carries. Either argument is valid, when the external factors are not considered.
I think it's very dangerous to go across positions like that, IMO. Carries are not the same thing as targets. In some ways they're similar, in that a high number indicates quality, but it's a bit more complicated than that.

If you want a real world perspective, look at Lee Evans and Ike Hilliard. Evans had 113 targets and 849 yards, or 7.5 yards per target. Hilliard had 87 targets and 722 yards, or 8.3 yards per target. Evans had a 49% catch-to-target ratio, and Hilliard had a 71% catch-to-target ratio.

Hilliard did 'an awful lot with minimal targets' last year, but that doesn't mean he's a better breakout candidate than Evans.

 
Okay, I just finished calculating some things for the multiplication method. First up, the minimum yearly stat lines for the Top 20’s:

Yearly Minimum YPts*TG Breakdown

2002: 147 targets, 920 receiving yards, 7 TD’s, 134 fantasy points

2003: 140 targets, 840 receiving yards, 3 TD’s, 102 fantasy points

2004: 132 targets, 970 receiving yards, 7 TD’s, 139 fantasy points

2005: 142 targets, 930 receiving yards, 5 TD’s, 123 fantasy points

2006: 126 targets, 1,010 receiving yards, 6 TD’s, 137 fantasy points

2007: 129 targets, 980 receiving yards, 4 TD’s, 122 fantasy points

Yearly Minimum TDs*TG Breakdown

2002: 112 targets, 780 receiving yards, 7 TD’s, 120 fantasy points

2003: 74 targets, 720 receiving yards, 9 TD’s, 126 fantasy points

2004: 114 targets, 820 receiving yards, 8 TD’s, 130 fantasy points

2005: 109 targets, 680 receiving yards, 7 TD’s, 110 fantasy points

2006: 125 targets, 990 receiving yards, 6 TD’s, 135 fantasy points

2007: 95 targets, 700 receiving yards, 9 TD’s, 124 fantasy points

Aaaand the minimum averages total:

Minimum YPts*TG stat line: 136 targets, 960 receiving yards, 5 TD’s, 126 fantasy points

Minimum TDs*TG stat line: 104 targets, 760 receiving yards, 8 TD’s, 124 fantasy points

The interesting thing there, I think, is that although there are differences between yardage and TD’s between the two, they end up nearly balancing out one another exactly in the fantasy points. And here are the yearly rankings:

2002 Ypts*Tg

1 Marvin Harrison

2 Hines Ward

3 Terrell Owens

4 Randy Moss

5 Eric Moulds

6 Peerless Price

7 Joe Horn

8 Marty Booker

9 Torry Holt

10 Plaxico Burress

11 Jerry Rice

12 Amani Toomer

13 Koren Robinson

14 Laveranues Coles

15 Rod Gardner

16 Jimmy Smith

17 Keyshawn Johnson

18 Rod Smith

19 Isaac Bruce

20 Donald Driver

2003 Ypts*Tg

1 Torry Holt

2 Randy Moss

3 Anquan Boldin

4 Chad Johnson

5 Laveranues Coles

6 Hines Ward

7 Marvin Harrison

8 Derrick Mason

9 Amani Toomer

10 Steve Smith

11 Keenan McCardell

12 Terrell Owens

13 Darrell Jackson

14 Santana Moss

15 Joe Horn

16 Chris Chambers

17 Isaac Bruce

18 Peerless Price

19 Peter Warrick

20 Andre Johnson

2004 Ypts*Tg

1 Muhsin Muhammad

2 Joe Horn

3 Chad Johnson

4 Javon Walker

5 Drew Bennett

6 Marvin Harrison

7 Torry Holt

8 Terrell Owens

9 Darrell Jackson

10 Isaac Bruce

11 Donald Driver

12 Derrick Mason

13 Reggie Wayne

14 Jimmy Smith

15 Rod Smith

16 Eric Moulds

17 Andre Johnson

18 Jerry Porter

19 Michael Clayton

20 Chris Chambers

2005 Ypts*Tg

1 Anquan Boldin

2 Steve Smith

3 Larry Fitzgerald

4 Chad Johnson

5 Torry Holt

6 Plaxico Burress

7 Chris Chambers

8 Santana Moss

9 Joey Galloway

10 Donald Driver

11 Marvin Harrison

12 Derrick Mason

13 Terry Glenn

14 Rod Smith

15 Jerry Porter

16 Jimmy Smith

17 Reggie Wayne

18 Deion Branch

19 Randy Moss

20 Antonio Bryant

2006 Ypts*Tg

1 Donald Driver

2 Chad Johnson

3 Torry Holt

4 Marvin Harrison

5 Roy Williams

6 Andre Johnson

7 Anquan Boldin

8 Steve Smith

9 Reggie Wayne

10 Terrell Owens

11 Lee Evans

12 Laveranues Coles

13 Mike Furrey

14 Javon Walker

15 Joey Galloway

16 T.J. Houshmandzadeh

17 Hines Ward

18 Plaxico Burress

19 Marques Colston

20 Darrell Jackson

2007 Ypts*Tg

1 Chad Johnson

2 Randy Moss

3 Reggie Wayne

4 Larry Fitzgerald

5 Brandon Marshall

6 Braylon Edwards

7 T.J. Houshmandzadeh

8 Terrell Owens

9 Derrick Mason

10 Torry Holt

11 Wes Welker

12 Marques Colston

13 Roddy White

14 Steve Smith

15 Bobby Engram

16 Kevin Curtis

17 Jerricho Cotchery

18 Plaxico Burress

19 Donald Driver

20 Chris Chambers

2002 TDs*Tg

1 Marvin Harrison

2 Terrell Owens

3 Hines Ward

4 Eric Moulds

5 Peerless Price

6 Randy Moss

7 Rod Gardner

8 Amani Toomer

9 Joe Horn

10 Jerry Rice

11 Donald Driver

12 Plaxico Burress

13 Marty Booker

14 Jimmy Smith

15 Isaac Bruce

16 Todd Pinkston

17 Rod Smith

18 Wayne Chrebet

19 Keyshawn Johnson

20 Travis Taylor

2003 TDs*Tg

1 Randy Moss

2 Torry Holt

3 Hines Ward

4 Chad Johnson

5 Chris Chambers

6 Marvin Harrison

7 Anquan Boldin

8 Terrell Owens

9 Joe Horn

10 Santana Moss

11 Darrell Jackson

12 Keenan McCardell

13 Derrick Mason

14 Steve Smith

15 Laveranues Coles

16 Peter Warrick

17 David Boston

18 Amani Toomer

19 Reggie Wayne

20 Javon Walker

2004 TDs*Tg

1 Muhsin Muhammad

2 Marvin Harrison

3 Terrell Owens

4 Javon Walker

5 Joe Horn

6 Drew Bennett

7 Chad Johnson

8 Reggie Wayne

9 Torry Holt

10 Donald Driver

11 Jerry Porter

12 Randy Moss

13 Darrell Jackson

14 Derrick Mason

15 Brandon Stokley

16 Chris Chambers

17 Roy Williams

18 Rod Smith

19 Nate Burleson

20 Isaac Bruce

2005 TDs*Tg

1 Chris Chambers

2 Steve Smith

3 Larry Fitzgerald

4 Marvin Harrison

5 Joey Galloway

6 Torry Holt

7 Chad Johnson

8 Hines Ward

9 Santana Moss

10 Anquan Boldin

11 Plaxico Burress

12 Randy Moss

13 Keenan McCardell

14 Donte Stallworth

15 Joe Jurevicius

16 Terry Glenn

17 T.J. Houshmandzadeh

18 Jimmy Smith

19 Amani Toomer

20 Rod Smith

2006 TDs*Tg

1 Terrell Owens

2 Torry Holt

3 Marvin Harrison

4 Donald Driver

5 Reggie Wayne

6 Plaxico Burress

7 T.J. Houshmandzadeh

8 Steve Smith

9 Darrell Jackson

10 Lee Evans

11 Chad Johnson

12 Roy Williams

13 Javon Walker

14 Joey Galloway

15 Marques Colston

16 Laveranues Coles

17 Mike Furrey

18 Andre Johnson

19 Hines Ward

20 Jerricho Cotchery

2007 TDs*Tg

1 Randy Moss

2 Braylon Edwards

3 Terrell Owens

4 T.J. Houshmandzadeh

5 Plaxico Burress

6 Larry Fitzgerald

7 Marques Colston

8 Reggie Wayne

9 Chad Johnson

10 Brandon Marshall

11 Wes Welker

12 Torry Holt

13 Steve Smith

14 Greg Jennings

15 Anquan Boldin

16 Nate Burleson

17 Roddy White

18 Derrick Mason

19 Kevin Curtis

20 Bobby Engram

It appears the * lists are a lot more consistent in terms of who stays on them from year to year. I’ll post more analysis soon but I just wanted to get this in here for everyone to look at. Thanks for the suggestion Chase!

 
Michael Fox said:
JG's approach lets you highlight players who have been productive relative to the opportunities they were offered, allowing you to find breakout candidates.
True, but you can potentially discover those who are producing with whatever targets they are currently getting, which will eventually mean that their team will give them more targets. It's a constantly revolving cycle.
Both of you are assuming that a guy with 50 receptions on 70 targets is a better breakout candidate than a guy with 45 receptions on 90 targets. It's perfectly reasonable to assume the latter guy is the better breakout candidate.
:goodposting: I was going to post something similar to this. And JGalligan's bolded statement above is not a valid assumption IMO.
I am not sure that is true. To put it into a real world perspective, JGalligans post is somewhat describing the situation that MJD is in this year (and was in last year). A guy who has not had a ton of targets, relatively speaking, but who has done an awful lot with minimal targets.In the situation that Chase desribes it is basically saying that someone with the larger number of targets is more likely to break out. Real world situation here would be Lendale White, who had almost double the number of carries that Drew had but scored fewer fantasy points (depending on format) than did MJD.

Now I realize that carries does not equal targets, but I think it gets the point across in this situation.

Obviously there are other factors that must be taken into account. In this case, the Chris Johnson factor or the Fred Taylor factor (which have very different impact on each of these guys situations). But if just looking at the numbers, one could argue that LEndale would be up for a better year, simply because he was utilized more. One could also argue that MJD is up for the better year because he has done so much more with the limited number of carries. Either argument is valid, when the external factors are not considered.
I think it's very dangerous to go across positions like that, IMO. Carries are not the same thing as targets. In some ways they're similar, in that a high number indicates quality, but it's a bit more complicated than that.

If you want a real world perspective, look at Lee Evans and Ike Hilliard. Evans had 113 targets and 849 yards, or 7.5 yards per target. Hilliard had 87 targets and 722 yards, or 8.3 yards per target. Evans had a 49% catch-to-target ratio, and Hilliard had a 71% catch-to-target ratio.

Hilliard did 'an awful lot with minimal targets' last year, but that doesn't mean he's a better breakout candidate than Evans.
I agree with everything that you've said and thank you again for suggesting the *. Although, I don't think it can hurt to take the /TG's into account when looking for breakout players. As you said, it's not the best way to go and look for them, but it does give you a nice list to begin working with. Do you think it's not worthy to use it at all? Or do you just think it's not the best way to do it?

 
Michael Fox said:
JG's approach lets you highlight players who have been productive relative to the opportunities they were offered, allowing you to find breakout candidates.
True, but you can potentially discover those who are producing with whatever targets they are currently getting, which will eventually mean that their team will give them more targets. It's a constantly revolving cycle.
Both of you are assuming that a guy with 50 receptions on 70 targets is a better breakout candidate than a guy with 45 receptions on 90 targets. It's perfectly reasonable to assume the latter guy is the better breakout candidate.
Chase - I'm not making that assumption necessarily. The analysis provides a set of facts from which you can begin to draw conclusions. It would be silly to have a narrow, rigid framework for identifying breakout candidates. This is just one approach among many.
 
Chase - I'm not making that assumption necessarily. The analysis provides a set of facts from which you can begin to draw conclusions. It would be silly to have a narrow, rigid framework for identifying breakout candidates. This is just one approach among many.
:goodposting: My thoughts exactly. Each starting point has its place, it seems to me.
 
I actually think the two statistics show two very different things. AS you describe the most targets are probably showing you the least amount of risk associated with a player (or the more risk mitigation at the very least). Where as the other stat could potentially be used to show the most potential (biggest risk/reward sort of option).
I agree with everything that you've said and thank you again for suggesting the *. Although, I don't think it can hurt to take the /TG's into account when looking for breakout players. As you said, it's not the best way to go and look for them, but it does give you a nice list to begin working with. Do you think it's not worthy to use it at all? Or do you just think it's not the best way to do it?
Once again, you guys are assuming that a guy with 50 receptions on 70 targets is a better breakout candidate than a guy with 45 receptions on 90 targets.To say that looking at yards/target can be used to show potential is to say that a high Y/T guy has more potential than a low Y/T guy, and that 50/70 is a better sign for the future than 45/90. I'm not saying it is or it isn't, but my guess is that it isn't.Targets are very tricky data. Is it meaningless to use X/target? I dunno. At this point, it's a lot of conjecture.
 
I actually think the two statistics show two very different things. AS you describe the most targets are probably showing you the least amount of risk associated with a player (or the more risk mitigation at the very least). Where as the other stat could potentially be used to show the most potential (biggest risk/reward sort of option).
I agree with everything that you've said and thank you again for suggesting the *. Although, I don't think it can hurt to take the /TG's into account when looking for breakout players. As you said, it's not the best way to go and look for them, but it does give you a nice list to begin working with. Do you think it's not worthy to use it at all? Or do you just think it's not the best way to do it?
Once again, you guys are assuming that a guy with 50 receptions on 70 targets is a better breakout candidate than a guy with 45 receptions on 90 targets.To say that looking at yards/target can be used to show potential is to say that a high Y/T guy has more potential than a low Y/T guy, and that 50/70 is a better sign for the future than 45/90. I'm not saying it is or it isn't, but my guess is that it isn't.Targets are very tricky data. Is it meaningless to use X/target? I dunno. At this point, it's a lot of conjecture.
What if we threw receptions into the equation somehow?
 
I actually think the two statistics show two very different things. AS you describe the most targets are probably showing you the least amount of risk associated with a player (or the more risk mitigation at the very least). Where as the other stat could potentially be used to show the most potential (biggest risk/reward sort of option).
I agree with everything that you've said and thank you again for suggesting the *. Although, I don't think it can hurt to take the /TG's into account when looking for breakout players. As you said, it's not the best way to go and look for them, but it does give you a nice list to begin working with. Do you think it's not worthy to use it at all? Or do you just think it's not the best way to do it?
Once again, you guys are assuming that a guy with 50 receptions on 70 targets is a better breakout candidate than a guy with 45 receptions on 90 targets.To say that looking at yards/target can be used to show potential is to say that a high Y/T guy has more potential than a low Y/T guy, and that 50/70 is a better sign for the future than 45/90. I'm not saying it is or it isn't, but my guess is that it isn't.Targets are very tricky data. Is it meaningless to use X/target? I dunno. At this point, it's a lot of conjecture.
What if we threw receptions into the equation somehow?
Or at least a reception/target ratio.
 
Chase - facts are your friend. I'd prefer to start with facts, knowing that my conclusions face the danger of being incorrect, than start with nothing but perception.

Thx to JG for giving us something to think about. That happens far too infrequently in the SP.

 
Once again, you guys are assuming that a guy with 50 receptions on 70 targets is a better breakout candidate than a guy with 45 receptions on 90 targets.To say that looking at yards/target can be used to show potential is to say that a high Y/T guy has more potential than a low Y/T guy, and that 50/70 is a better sign for the future than 45/90. I'm not saying it is or it isn't, but my guess is that it isn't.Targets are very tricky data. Is it meaningless to use X/target? I dunno. At this point, it's a lot of conjecture.
I am not actually saying that a guy with 50/70 is better than a guy with 45/90. All I am saying is that a guy with 50/70 could potentially be better than a guy with 45/90 and that it is worth looking into further.That is not a huge difference, I realize, but I think that this is a statistic to show some level of potential which could be useful in determining the value of a player.
 
I actually think the two statistics show two very different things. AS you describe the most targets are probably showing you the least amount of risk associated with a player (or the more risk mitigation at the very least). Where as the other stat could potentially be used to show the most potential (biggest risk/reward sort of option).
I agree with everything that you've said and thank you again for suggesting the *. Although, I don't think it can hurt to take the /TG's into account when looking for breakout players. As you said, it's not the best way to go and look for them, but it does give you a nice list to begin working with. Do you think it's not worthy to use it at all? Or do you just think it's not the best way to do it?
Once again, you guys are assuming that a guy with 50 receptions on 70 targets is a better breakout candidate than a guy with 45 receptions on 90 targets.To say that looking at yards/target can be used to show potential is to say that a high Y/T guy has more potential than a low Y/T guy, and that 50/70 is a better sign for the future than 45/90. I'm not saying it is or it isn't, but my guess is that it isn't.Targets are very tricky data. Is it meaningless to use X/target? I dunno. At this point, it's a lot of conjecture.
What if we threw receptions into the equation somehow?
Or at least a reception/target ratio.
Well, my first question would be what specifically are we trying to accomplish?
 
Once again, you guys are assuming that a guy with 50 receptions on 70 targets is a better breakout candidate than a guy with 45 receptions on 90 targets.To say that looking at yards/target can be used to show potential is to say that a high Y/T guy has more potential than a low Y/T guy, and that 50/70 is a better sign for the future than 45/90. I'm not saying it is or it isn't, but my guess is that it isn't.Targets are very tricky data. Is it meaningless to use X/target? I dunno. At this point, it's a lot of conjecture.
I am not actually saying that a guy with 50/70 is better than a guy with 45/90. All I am saying is that a guy with 50/70 could potentially be better than a guy with 45/90 and that it is worth looking into further.That is not a huge difference, I realize, but I think that this is a statistic to show some level of potential which could be useful in determining the value of a player.
Precisely. If somebody wants to do a study to determine which variables have the most predictive power for determining "breakout WRs", consider me interested.
 
Chase - facts are your friend. I'd prefer to start with facts, knowing that my conclusions face the danger of being incorrect, than start with nothing but perception.Thx to JG for giving us something to think about. That happens far too infrequently in the SP.
I agree that JG has done a great job here. I don't understand how wanting to start with facts should make you put targets in the denominator instead of the numerator, or vice versa.
 
I actually think the two statistics show two very different things. AS you describe the most targets are probably showing you the least amount of risk associated with a player (or the more risk mitigation at the very least). Where as the other stat could potentially be used to show the most potential (biggest risk/reward sort of option).
I agree with everything that you've said and thank you again for suggesting the *. Although, I don't think it can hurt to take the /TG's into account when looking for breakout players. As you said, it's not the best way to go and look for them, but it does give you a nice list to begin working with. Do you think it's not worthy to use it at all? Or do you just think it's not the best way to do it?
Once again, you guys are assuming that a guy with 50 receptions on 70 targets is a better breakout candidate than a guy with 45 receptions on 90 targets.To say that looking at yards/target can be used to show potential is to say that a high Y/T guy has more potential than a low Y/T guy, and that 50/70 is a better sign for the future than 45/90. I'm not saying it is or it isn't, but my guess is that it isn't.Targets are very tricky data. Is it meaningless to use X/target? I dunno. At this point, it's a lot of conjecture.
What if we threw receptions into the equation somehow?
Or at least a reception/target ratio.
IMO it is too dependent on situational factors - How is the player used (e.g., deep routes vs. short routes)? How accurate is the QB? What is the quality of offensive teammates (i.e., can the defense just focus on the player in question)? etc.Chambers is a great example of this. I posted the details in JGalligan's 110 targets thread, but IIRC he caught 47% of his targets in Miami last year, then caught 57% to 60% of his targets in San Diego, despite moving across country and joining a new offense. IMO that perfectly illustrates that receptions per target has to do with a lot more than the WR himself.
 
Once again, you guys are assuming that a guy with 50 receptions on 70 targets is a better breakout candidate than a guy with 45 receptions on 90 targets.To say that looking at yards/target can be used to show potential is to say that a high Y/T guy has more potential than a low Y/T guy, and that 50/70 is a better sign for the future than 45/90. I'm not saying it is or it isn't, but my guess is that it isn't.Targets are very tricky data. Is it meaningless to use X/target? I dunno. At this point, it's a lot of conjecture.
I am not actually saying that a guy with 50/70 is better than a guy with 45/90. All I am saying is that a guy with 50/70 could potentially be better than a guy with 45/90 and that it is worth looking into further.That is not a huge difference, I realize, but I think that this is a statistic to show some level of potential which could be useful in determining the value of a player.
What stuff are you going to use when you look into it further? I assume team situation, player talent, etc. But if all else equal, then what? Is the 50/70 guy better or the 45/90 guy better? If your answer is 'it depends', then that's a pretty good sign that to you, targets don't have a lot of value.For example, it's possible that a guy with 1000 receiving yards has more potential than a guy with 1200 receiving yards. But more often than not, the guy with more receiving yards, all else being equal, is the better guy. The question is whether more often than not, a guy with the better T*Y is better than the guy with the better Y/T.
 
Well, my first question would be what specifically are we trying to accomplish?
Chase - facts are your friend. I'd prefer to start with facts, knowing that my conclusions face the danger of being incorrect, than start with nothing but perception.Thx to JG for giving us something to think about. That happens far too infrequently in the SP.
I agree that JG has done a great job here. I don't understand how wanting to start with facts should make you put targets in the denominator instead of the numerator, or vice versa.
I'm not trying to say that the denominator way is right -- I'm just trying to make it a little more useful. Although I misunderstood your point about 45 targets/90 receptions and 50 receptions/90 targets. It just dawned on me the point you were trying to make. So uh, yeah, let's completely forget that I suggested the reception to target ratio shall we?
 
Once again, you guys are assuming that a guy with 50 receptions on 70 targets is a better breakout candidate than a guy with 45 receptions on 90 targets.To say that looking at yards/target can be used to show potential is to say that a high Y/T guy has more potential than a low Y/T guy, and that 50/70 is a better sign for the future than 45/90. I'm not saying it is or it isn't, but my guess is that it isn't.Targets are very tricky data. Is it meaningless to use X/target? I dunno. At this point, it's a lot of conjecture.
I am not actually saying that a guy with 50/70 is better than a guy with 45/90. All I am saying is that a guy with 50/70 could potentially be better than a guy with 45/90 and that it is worth looking into further.That is not a huge difference, I realize, but I think that this is a statistic to show some level of potential which could be useful in determining the value of a player.
What stuff are you going to use when you look into it further? I assume team situation, player talent, etc. But if all else equal, then what? Is the 50/70 guy better or the 45/90 guy better? If your answer is 'it depends', then that's a pretty good sign that to you, targets don't have a lot of value.For example, it's possible that a guy with 1000 receiving yards has more potential than a guy with 1200 receiving yards. But more often than not, the guy with more receiving yards, all else being equal, is the better guy. The question is whether more often than not, a guy with the better T*Y is better than the guy with the better Y/T.
It's not necessary to say for sure which one is better. You can look at each players situation heading into the new season and make an educated decision based on all of the data combined as to whether or not you think they'll improve or regress. It's not concrete and it's not set in stone. There's a lot of predictions and looking at immeasurable outside factors that are involved.So it's impossible to say which one will actually be better. But with everything taken into account, you can make a better prediction.
 
Last edited by a moderator:
I looked at all WRs from 2002-2006 that had at least 400 receiving yards in a season.

If you just use the WRs yard per game in Year N, you can get a decent prediction for how many yards per game the WR will have the next year. The R^2 is 0.34, which means that roughly 34% of the WR's yards per game in Year N+1 can be explained by their Y/G in Year N. What's the other 66%? WR improvement, injuries to other players on the offense, injuries to the WR, QB switches, coaching changes, luck, etc.

I then ran a multiple regression using yards/game in Year N and yards/game divided by targets/game (i.e., Y/T but on a per game basis) to predict Year N+1 yards per game. The R^2 only jumps to .35, which is obviously a meaningless change. The Y/T stat has a p-value of .25, which in the official statistical sense, means it is not statistically significant to predicting N+1 yards per game (and even if it were statistically significant, it's practically insignificant).

I also ran a multiple regression using yards/game in Year N and yards/game multiplied by targets per game (i.e., Y*T on a per game basis) to predict Year N+1 yards per game. The R^2 only jumps to .35, which is (still) obviously meaningless. The Y*T stat has a p-value of 0.23, which is also not statistically significant.

In sum, in the aggregate, using targets -- in the numerator or the denominator -- has no predictive value. Of course, target data may still have some predictive value in the individual sense, although I'd be skeptical.

 
JGalligan said:
Chase Stuart said:
The other thread got too long before I could chime in, so I'll ask my question here:

Why are you putting targets in the denominator and not the numerator? Why not Yards*Targets, or TDs*Targets?

Once again, I know I'm coming in too late to the game here, but what is the goal of using yards/target or TDs/target in Year N-1? To predict what, and whatever that is, why not use Yards*Targets or TDs*Targets?
Well to be completely honest, using TDs or YardagePts*Targets didn't even occur to me. It absolutely makes sense and it changes things significantly, but I think each method would have it's own usefulness. From what I'm seeing, TDs + Ypts*Tg would show who had the most of each. On the other hand, TDs + Ypts/Tg brings into the WR's who didn't necessarily get many targets, but managed to successfully convert them into a significant amount of fantasy points. For instance, Ypts-wise, Brandon Stokley and Justin Gage. Hopefully that made sense. But yeah, I'm not going to lie, I didn't even think of using times targets. It definitely makes sense but I'm still pretty new to the wonderfully complicated realm that is statistics. It didn't even dawn on me to use that.

Anyway, I put together the Top 20 for 2007 for each using that -- here it is:

Ypts*Tg

Chad Johnson

Randy Moss

Reggie Wayne

Larry Fitzgerald

Brandon Marshall

Braylon Edwards

T.J. Houshmandzadeh

Terrell Owens

Derrick Mason

Torry Holt

Wes Welker

Marques Colston

Roddy White

Steve Smith

Bobby Engram

Kevin Curtis

Jerricho Cotchery

Plaxico Burress

Donald Driver

Chris Chambers

TDs*Tg

Randy Moss

Braylon Edwards

Terrell Owens

T.J. Houshmandzadeh

Plaxico Burress

Larry Fitzgerald

Marques Colston

Reggie Wayne

Chad Johnson

Brandon Marshall

Wes Welker

Torry Holt

Steve Smith

Greg Jennings

Anquan Boldin

Nate Burleson

Roddy White

Derrick Mason

Kevin Curtis

Bobby Engram

They're definitely very different. I'll put something together taking a look at the stats yearly with it and post it in here when I'm done.
See I think this is a better list. Great work, but I think the switching numerator/denom thing makes a better list. But it is splitting hairs. very good post from top to bottom.
 
Michael Fox said:
JG's approach lets you highlight players who have been productive relative to the opportunities they were offered, allowing you to find breakout candidates.
True, but you can potentially discover those who are producing with whatever targets they are currently getting, which will eventually mean that their team will give them more targets. It's a constantly revolving cycle.
Both of you are assuming that a guy with 50 receptions on 70 targets is a better breakout candidate than a guy with 45 receptions on 90 targets. It's perfectly reasonable to assume the latter guy is the better breakout candidate.
:goodposting: I was going to post something similar to this. And JGalligan's bolded statement above is not a valid assumption IMO.
Right. Fantasy owners generally think targets are a *good* thing. If you see in week 1 that your WR2 has 13 targets, you're going to be very happy, even if he only has five receptions. Certainly you'll be happier than if he has 5 receptions on 8 targets. I'm not saying that beyond a shadow of a doubt that yards*target is better than yards/target; it's basically an open question. In my view, though, I'd guess that putting targets in the numerator is better.
agreed. I truly think the "targets" per week stas made avail by FBG's is why I am usually able to discern between the 1 week wonder, and the wk 1 FA WR that breaks out that year, even though others scored better in a particular week (but did it on 3 for 120 with 5 targets vs 5 for 88 with 12 targets-- I think most sharks would rather have the latter guy, as consistency wins championships)
 
JGalligan said:
I don’t think you need to be a math whiz to come to the conclusion that there’s a lot of WR’s that break into each stats Top 20 list every year. It can likely be attributed to many things. One, the WR position is the most difficult to consistently put up big numbers. This is due to a variety of factors, including but not limited to: Different QB’s, other WR’s, rookies, contract years, pure laziness, and age. Two, the WR corps on most teams can often be wide open and vary drastically from year to year. Unlike the QB and RB positions, where the starter only needs to worry about 1 or 2 possible replacements, WR’s need to worry about 3 or 4. Studs aside, it’s a very hard position to solidify a guaranteed position in. Especially in the financial landscape the NFL is experiencing right now.
Did you intentionally ignore injuries here? My guess is that at least half of those who drop out of your lists every year do so because of injury, either to themselves or to others (e.g., their starting QBs).
you ignore all thing out of your control. While some WR's are bigger injury risks, that is mostly a luck of the draw thing.
 
JGalligan said:
I don’t think you need to be a math whiz to come to the conclusion that there’s a lot of WR’s that break into each stats Top 20 list every year. It can likely be attributed to many things. One, the WR position is the most difficult to consistently put up big numbers. This is due to a variety of factors, including but not limited to: Different QB’s, other WR’s, rookies, contract years, pure laziness, and age. Two, the WR corps on most teams can often be wide open and vary drastically from year to year. Unlike the QB and RB positions, where the starter only needs to worry about 1 or 2 possible replacements, WR’s need to worry about 3 or 4. Studs aside, it’s a very hard position to solidify a guaranteed position in. Especially in the financial landscape the NFL is experiencing right now.
Did you intentionally ignore injuries here? My guess is that at least half of those who drop out of your lists every year do so because of injury, either to themselves or to others (e.g., their starting QBs).
you ignore all thing out of your control. While some WR's are bigger injury risks, that is mostly a luck of the draw thing.
Of course it's out of your control. But part of the premise of this post is that there is a fair amount of turnover in these lists. But if a signficant amount of that turnover is due to injury, which is not predictable, then what does that mean about using metrics like these for predictive purposes?
 
Last edited by a moderator:
JGalligan said:
I don’t think you need to be a math whiz to come to the conclusion that there’s a lot of WR’s that break into each stats Top 20 list every year. It can likely be attributed to many things. One, the WR position is the most difficult to consistently put up big numbers. This is due to a variety of factors, including but not limited to: Different QB’s, other WR’s, rookies, contract years, pure laziness, and age. Two, the WR corps on most teams can often be wide open and vary drastically from year to year. Unlike the QB and RB positions, where the starter only needs to worry about 1 or 2 possible replacements, WR’s need to worry about 3 or 4. Studs aside, it’s a very hard position to solidify a guaranteed position in. Especially in the financial landscape the NFL is experiencing right now.
Did you intentionally ignore injuries here? My guess is that at least half of those who drop out of your lists every year do so because of injury, either to themselves or to others (e.g., their starting QBs).
you ignore all thing out of your control. While some WR's are bigger injury risks, that is mostly a luck of the draw thing.
Yeah, injuries are uncontrollable, but I didn't mean to exclude them from those reasons. Injuries do play a part in the fluctuation of the lists from year to year. So it's all edited and fixed, thanks JWB.
 
Michael Fox said:
JG's approach lets you highlight players who have been productive relative to the opportunities they were offered, allowing you to find breakout candidates.
True, but you can potentially discover those who are producing with whatever targets they are currently getting, which will eventually mean that their team will give them more targets. It's a constantly revolving cycle.
Both of you are assuming that a guy with 50 receptions on 70 targets is a better breakout candidate than a guy with 45 receptions on 90 targets. It's perfectly reasonable to assume the latter guy is the better breakout candidate.
:goodposting: I was going to post something similar to this. And JGalligan's bolded statement above is not a valid assumption IMO.
I am not sure that is true. To put it into a real world perspective, JGalligans post is somewhat describing the situation that MJD is in this year (and was in last year). A guy who has not had a ton of targets, relatively speaking, but who has done an awful lot with minimal targets.In the situation that Chase desribes it is basically saying that someone with the larger number of targets is more likely to break out. Real world situation here would be Lendale White, who had almost double the number of carries that Drew had but scored fewer fantasy points (depending on format) than did MJD.

Now I realize that carries does not equal targets, but I think it gets the point across in this situation.

Obviously there are other factors that must be taken into account. In this case, the Chris Johnson factor or the Fred Taylor factor (which have very different impact on each of these guys situations). But if just looking at the numbers, one could argue that LEndale would be up for a better year, simply because he was utilized more. One could also argue that MJD is up for the better year because he has done so much more with the limited number of carries. Either argument is valid, when the external factors are not considered.
I think it's very dangerous to go across positions like that, IMO. Carries are not the same thing as targets. In some ways they're similar, in that a high number indicates quality, but it's a bit more complicated than that.

If you want a real world perspective, look at Lee Evans and Ike Hilliard. Evans had 113 targets and 849 yards, or 7.5 yards per target. Hilliard had 87 targets and 722 yards, or 8.3 yards per target. Evans had a 49% catch-to-target ratio, and Hilliard had a 71% catch-to-target ratio.

Hilliard did 'an awful lot with minimal targets' last year, but that doesn't mean he's a better breakout candidate than Evans.
I agree with everything that you've said and thank you again for suggesting the *. Although, I don't think it can hurt to take the /TG's into account when looking for breakout players. As you said, it's not the best way to go and look for them, but it does give you a nice list to begin working with. Do you think it's not worthy to use it at all? Or do you just think it's not the best way to do it?
I think it is uselfull, just not as usefull as the other list. That said, would be a great 2econd tier stat to look at.*thinking out loud* Is there ANY correlation between preseason targets by all qbs/or starting qb only that would indicate who the new "breakout target" (ie colston/cothcery) guys are gonna be so you can get em before their value gets too high?

 
I looked at all WRs from 2002-2006 that had at least 400 receiving yards in a season. If you just use the WRs yard per game in Year N, you can get a decent prediction for how many yards per game the WR will have the next year. The R^2 is 0.34, which means that roughly 34% of the WR's yards per game in Year N+1 can be explained by their Y/G in Year N. What's the other 66%? WR improvement, injuries to other players on the offense, injuries to the WR, QB switches, coaching changes, luck, etc.I then ran a multiple regression using yards/game in Year N and yards/game divided by targets/game (i.e., Y/T but on a per game basis) to predict Year N+1 yards per game. The R^2 only jumps to .35, which is obviously a meaningless change. The Y/T stat has a p-value of .25, which in the official statistical sense, means it is not statistically significant to predicting N+1 yards per game (and even if it were statistically significant, it's practically insignificant).I also ran a multiple regression using yards/game in Year N and yards/game multiplied by targets per game (i.e., Y*T on a per game basis) to predict Year N+1 yards per game. The R^2 only jumps to .35, which is (still) obviously meaningless. The Y*T stat has a p-value of 0.23, which is also not statistically significant.In sum, in the aggregate, using targets -- in the numerator or the denominator -- has no predictive value. Of course, target data may still have some predictive value in the individual sense, although I'd be skeptical.
:goodposting:
 
I looked at all WRs from 2002-2006 that had at least 400 receiving yards in a season. If you just use the WRs yard per game in Year N, you can get a decent prediction for how many yards per game the WR will have the next year. The R^2 is 0.34, which means that roughly 34% of the WR's yards per game in Year N+1 can be explained by their Y/G in Year N. What's the other 66%? WR improvement, injuries to other players on the offense, injuries to the WR, QB switches, coaching changes, luck, etc.I then ran a multiple regression using yards/game in Year N and yards/game divided by targets/game (i.e., Y/T but on a per game basis) to predict Year N+1 yards per game. The R^2 only jumps to .35, which is obviously a meaningless change. The Y/T stat has a p-value of .25, which in the official statistical sense, means it is not statistically significant to predicting N+1 yards per game (and even if it were statistically significant, it's practically insignificant).I also ran a multiple regression using yards/game in Year N and yards/game multiplied by targets per game (i.e., Y*T on a per game basis) to predict Year N+1 yards per game. The R^2 only jumps to .35, which is (still) obviously meaningless. The Y*T stat has a p-value of 0.23, which is also not statistically significant.In sum, in the aggregate, using targets -- in the numerator or the denominator -- has no predictive value. Of course, target data may still have some predictive value in the individual sense, although I'd be skeptical.
Oh wow. :goodposting: Did you do all that in Excel or do you use something different?
 
Michael Fox said:
JG's approach lets you highlight players who have been productive relative to the opportunities they were offered, allowing you to find breakout candidates.
True, but you can potentially discover those who are producing with whatever targets they are currently getting, which will eventually mean that their team will give them more targets. It's a constantly revolving cycle.
Both of you are assuming that a guy with 50 receptions on 70 targets is a better breakout candidate than a guy with 45 receptions on 90 targets. It's perfectly reasonable to assume the latter guy is the better breakout candidate.
:goodposting: I was going to post something similar to this. And JGalligan's bolded statement above is not a valid assumption IMO.
I am not sure that is true. To put it into a real world perspective, JGalligans post is somewhat describing the situation that MJD is in this year (and was in last year). A guy who has not had a ton of targets, relatively speaking, but who has done an awful lot with minimal targets.In the situation that Chase desribes it is basically saying that someone with the larger number of targets is more likely to break out. Real world situation here would be Lendale White, who had almost double the number of carries that Drew had but scored fewer fantasy points (depending on format) than did MJD.

Now I realize that carries does not equal targets, but I think it gets the point across in this situation.

Obviously there are other factors that must be taken into account. In this case, the Chris Johnson factor or the Fred Taylor factor (which have very different impact on each of these guys situations). But if just looking at the numbers, one could argue that LEndale would be up for a better year, simply because he was utilized more. One could also argue that MJD is up for the better year because he has done so much more with the limited number of carries. Either argument is valid, when the external factors are not considered.
I think it's very dangerous to go across positions like that, IMO. Carries are not the same thing as targets. In some ways they're similar, in that a high number indicates quality, but it's a bit more complicated than that.

If you want a real world perspective, look at Lee Evans and Ike Hilliard. Evans had 113 targets and 849 yards, or 7.5 yards per target. Hilliard had 87 targets and 722 yards, or 8.3 yards per target. Evans had a 49% catch-to-target ratio, and Hilliard had a 71% catch-to-target ratio.

Hilliard did 'an awful lot with minimal targets' last year, but that doesn't mean he's a better breakout candidate than Evans.
I agree with everything that you've said and thank you again for suggesting the *. Although, I don't think it can hurt to take the /TG's into account when looking for breakout players. As you said, it's not the best way to go and look for them, but it does give you a nice list to begin working with. Do you think it's not worthy to use it at all? Or do you just think it's not the best way to do it?
I think it is uselfull, just not as usefull as the other list. That said, would be a great 2econd tier stat to look at.*thinking out loud* Is there ANY correlation between preseason targets by all qbs/or starting qb only that would indicate who the new "breakout target" (ie colston/cothcery) guys are gonna be so you can get em before their value gets too high?
That's definitely a great idea to look at, but I don't think the Game Log Dominator tracks pre-season stats? I'm not sure if anywhere else does either, but it's worth a Google search. I'll go take a look.
 
I've come up with something decent here.

I looked at all WRs that had at least 400 receiving yards in any season from 2002-2006. Then I divided them into groups based on how many yards per game they had. Anyone over 70 yards per game was excluded, since they're too good to focus on. So I looked at the guys with 65.0-69.9 YPG, 60.0-64.9, 55.0-59.9..... to 25.0-29.9. I then looked at the 10 guys with the highest Y/T per game, and the 10 guys with the highest Y*T per game.

The table below shows the results.

Y/G T/G n+1 Y/G Y/G T/G n+1 Y/G65+ 67.4 6.8 39.9 68.3 9.4 52.360+ 61.7 6.5 51.1 62.9 9.0 63.555+ 57.5 5.8 43.5 58.9 8.3 46.950+ 51.8 5.6 49.9 53.2 8.3 45.445+ 46.7 5.1 41.9 47.9 6.9 55.140+ 43.6 4.6 50.7 42.9 7.2 48.435+ 37.8 4.0 35.1 37.8 6.2 29.330+ 32.4 3.5 30.5 33.4 5.7 27.025+ 28.3 3.0 27.8 27.5 4.6 26.7How does that table read? Look at the 45+ row. It states that of the WRs that averaged 45.0-49.9 YPG, the top 10 in yards per target (per game) averaged 46.7 yards per game that season and 5.1 targets per game. That's the high Y/T group. The (different) top 10 WRs in the high Y*T group averaged 47.9 yards per game, and averaged 6.9 targets per game. How did the two groups do the next season? The high Y/T group averaged 41.9 Y/G the next year; the high Y*T group averaged 55.1 Y/G the next year. So for the 45+ row, advantage Y*T over Y/T.For the most part, Y*T looks like a better predictor, at least among the WRs that averaged over 45 yards per game (that's 720 receiving yards). From 25-39.9 YPG (that's 400 to 638 receiving yards over 16 games), the Y/T group looks a little better. For example, in 2003, David Givens and Dez White both averaged 39 yards per game. White averaged 7.1 targets per game; Givens 4.1 targets per game. Givens averaged 58.3 YPG the next season, and White 23.1 YPG. Obviously that's just one example.

The sample sizes here aren't very big (10 of each group), so I'm skeptical to take any broad results. It appears that all else being equal, if you have two WRs with 800 receiving yards last season, there is strong evidence that the guy that had more targets is the better option for next year. If you have two WRs with 500 receiving yards, there's a bit of evidence to indicate that the guy that did it on fewer targets is the better option.

 
I've come up with something decent here.

I looked at all WRs that had at least 400 receiving yards in any season from 2002-2006. Then I divided them into groups based on how many yards per game they had. Anyone over 70 yards per game was excluded, since they're too good to focus on. So I looked at the guys with 65.0-69.9 YPG, 60.0-64.9, 55.0-59.9..... to 25.0-29.9. I then looked at the 10 guys with the highest Y/T per game, and the 10 guys with the highest Y*T per game.

The table below shows the results.

Y/G T/G n+1 Y/G Y/G T/G n+1 Y/G65+ 67.4 6.8 39.9 68.3 9.4 52.360+ 61.7 6.5 51.1 62.9 9.0 63.555+ 57.5 5.8 43.5 58.9 8.3 46.950+ 51.8 5.6 49.9 53.2 8.3 45.445+ 46.7 5.1 41.9 47.9 6.9 55.140+ 43.6 4.6 50.7 42.9 7.2 48.435+ 37.8 4.0 35.1 37.8 6.2 29.330+ 32.4 3.5 30.5 33.4 5.7 27.025+ 28.3 3.0 27.8 27.5 4.6 26.7How does that table read? Look at the 45+ row. It states that of the WRs that averaged 45.0-49.9 YPG, the top 10 in yards per target (per game) averaged 46.7 yards per game that season and 5.1 targets per game. That's the high Y/T group. The (different) top 10 WRs in the high Y*T group averaged 47.9 yards per game, and averaged 6.9 targets per game. How did the two groups do the next season? The high Y/T group averaged 41.9 Y/G the next year; the high Y*T group averaged 55.1 Y/G the next year. So for the 45+ row, advantage Y*T over Y/T.For the most part, Y*T looks like a better predictor, at least among the WRs that averaged over 45 yards per game (that's 720 receiving yards). From 25-39.9 YPG (that's 400 to 638 receiving yards over 16 games), the Y/T group looks a little better. For example, in 2003, David Givens and Dez White both averaged 39 yards per game. White averaged 7.1 targets per game; Givens 4.1 targets per game. Givens averaged 58.3 YPG the next season, and White 23.1 YPG. Obviously that's just one example.

The sample sizes here aren't very big (10 of each group), so I'm skeptical to take any broad results. It appears that all else being equal, if you have two WRs with 800 receiving yards last season, there is strong evidence that the guy that had more targets is the better option for next year. If you have two WRs with 500 receiving yards, there's a bit of evidence to indicate that the guy that did it on fewer targets is the better option.
It makes sense when you think about it, too. The better a player gets, the more targets he'll receive thanks to his proven skills. The younger and unproven a player is, the less targets he'll receive -- but if he's worth a look, he'll do more with less targets. A+ per usual, Chase. :goodposting:
 
I've come up with something decent here.

I looked at all WRs that had at least 400 receiving yards in any season from 2002-2006. Then I divided them into groups based on how many yards per game they had. Anyone over 70 yards per game was excluded, since they're too good to focus on. So I looked at the guys with 65.0-69.9 YPG, 60.0-64.9, 55.0-59.9..... to 25.0-29.9. I then looked at the 10 guys with the highest Y/T per game, and the 10 guys with the highest Y*T per game.

The table below shows the results.

Y/G T/G n+1 Y/G Y/G T/G n+1 Y/G65+ 67.4 6.8 39.9 68.3 9.4 52.360+ 61.7 6.5 51.1 62.9 9.0 63.555+ 57.5 5.8 43.5 58.9 8.3 46.950+ 51.8 5.6 49.9 53.2 8.3 45.445+ 46.7 5.1 41.9 47.9 6.9 55.140+ 43.6 4.6 50.7 42.9 7.2 48.435+ 37.8 4.0 35.1 37.8 6.2 29.330+ 32.4 3.5 30.5 33.4 5.7 27.025+ 28.3 3.0 27.8 27.5 4.6 26.7How does that table read? Look at the 45+ row. It states that of the WRs that averaged 45.0-49.9 YPG, the top 10 in yards per target (per game) averaged 46.7 yards per game that season and 5.1 targets per game. That's the high Y/T group. The (different) top 10 WRs in the high Y*T group averaged 47.9 yards per game, and averaged 6.9 targets per game. How did the two groups do the next season? The high Y/T group averaged 41.9 Y/G the next year; the high Y*T group averaged 55.1 Y/G the next year. So for the 45+ row, advantage Y*T over Y/T.For the most part, Y*T looks like a better predictor, at least among the WRs that averaged over 45 yards per game (that's 720 receiving yards). From 25-39.9 YPG (that's 400 to 638 receiving yards over 16 games), the Y/T group looks a little better. For example, in 2003, David Givens and Dez White both averaged 39 yards per game. White averaged 7.1 targets per game; Givens 4.1 targets per game. Givens averaged 58.3 YPG the next season, and White 23.1 YPG. Obviously that's just one example.

The sample sizes here aren't very big (10 of each group), so I'm skeptical to take any broad results. It appears that all else being equal, if you have two WRs with 800 receiving yards last season, there is strong evidence that the guy that had more targets is the better option for next year. If you have two WRs with 500 receiving yards, there's a bit of evidence to indicate that the guy that did it on fewer targets is the better option.
Cool stuff Chase. :thumbup:
 

Users who are viewing this thread

Back
Top