Tiering

[Let Da Big Dog Eat 42]

Here are the season point projections using my league scoring and FF Today projections. Where would you draw lines for your tiers? It seems like a pretty smooth curve with no real obviously breakpoints.

 

It's not that I don't understand the concept. It seems kind of arbitrary when you look at the raw stats. And, I don't agree with all the projections. So, I don't use tiers. I prefer to make picks based on what runs are going on in the draft and what I need.

 

Steve Smith CAR 139.3

Torry Holt STL 135.4

Larry Fitzgerald ARI 130.3

Chad Johnson CIN 124

Marvin Harrison IND 117

Hines Ward PIT 116.2

Donald Driver GB 114

Terrell Owens DAL 111

Chris Chambers MIA 107.6

Darrell Jackson SEA 105.3

Randy Moss OAK 103.7

Lee Evans BUF 103.5

Anquan Boldin ARI 102.8

Plaxico Burress NYG 99.8

Reggie Wayne IND 99.7

Deion Branch NE 97.5

Eddie Kennison KC 97.3

Matt Jones JAC 96.8

Rod Smith DEN 96.1

Joey Galloway TB 94.1

Terry Glenn DAL 90.5

Santana Moss WAS 90.3

T.J. Houshmandzadeh CIN 86.4

[VaTerp 0]

In my initial Wr rankings I have an elite tier of 8. Followed by a second tier that runs through 22.

 

Interestingly, you have 23 Wr's listed.

 

I see your point and cut offs, espcially for Wr's and Qb's, can be arbitrary.

 

I still find tiering to be very useful however, especially when it comes to the draft runs at certain positions that you eluded to.

 

For example, if I see that there is a run on Wr's in the mid rounds, I look at my tiers and see if I take a 3rd Rb or a Qb, and still have a Wr that I have tiered with the ones that are going, fall to me with my next pick. If that makes sense.

 

Tiers, like other ranking factors, are extremely subjective and just another tool that I find makes my drafting a little more structured.

[Let Da Big Dog Eat 42]

In my initial Wr rankings I have an elite tier of 8. Followed by a second tier that runs through 22.

 

Interestingly, you have 23 Wr's listed.

 

I could add more. I thought it was a reasonable sample. I went to 23 based on top WR to the first WR below the 90 pt mark. That is a difference of about 50 pts (more than a 1/3 dropoff). Seems like a huge difference to say only 2 tiers????????? How many rankings would you like to see?

[Parja 0]

Steve Smith CAR 139.3

Torry Holt STL 135.4

Larry Fitzgerald ARI 130.3

Chad Johnson CIN 124

------------------------------

Marvin Harrison IND 117

Hines Ward PIT 116.2

Donald Driver GB 114

Terrell Owens DAL 111

Chris Chambers MIA 107.6

Darrell Jackson SEA 105.3

Randy Moss OAK 103.7

Lee Evans BUF 103.5

Anquan Boldin ARI 102.8

-------------------------------

Plaxico Burress NYG 99.8

Reggie Wayne IND 99.7

Deion Branch NE 97.5

Eddie Kennison KC 97.3

Matt Jones JAC 96.8

Rod Smith DEN 96.1

Joey Galloway TB 94.1

Terry Glenn DAL 90.5

Santana Moss WAS 90.3

T.J. Houshmandzadeh CIN 86.4

 

I typically use 16 point tiers (1 point per week) and make note of the point differential between tiers (7 from 1-2, 3 from 2-3) which sometimes shows some huge dropoffs between tiers. Like if you go by the FFToday RB projections, you have Larry Johnson alone in Tier 1, then SA alone in Tier 2...but after a 51 point (3.2 tier) dropoff!

[t.j 35]

I use tiers but I don't split them up based on gaps in the sorted projections. I do it based on how much I trust a guy to produce, compared to the players projected above and below him. So for example, based on the projections posted earlier in this thread, I might put my best guess (my projection) for Harrison and Ward as roughly equal. But still put Harrison in a higher tier because I think Ward's numbers rely too much on the assumption that he'll be the primary receiving target in the red zone again. Or something like that.

[VaTerp 0]

I could add more. I thought it was a reasonable sample. I went to 23 based on top WR to the first WR below the 90 pt mark. That is a difference of about 50 pts (more than a 1/3 dropoff). Seems like a huge difference to say only 2 tiers????????? How many rankings would you like to see?

 

Its not about how many rankings I want to see. I was just making a point that the cut-offs can be arbitrary and that for Wr's I usually have fairly large groups in my tiers. The overall discrepancy in pts is rather large but sicne Wr's are so inconistent from year to year, its based on projections.

 

I will prob. break my second tier down further but I agree w/ TJ here. Im not as much of a projection guy as some other people here. A lot of my rankings are based on feel and my confidence that a player will produce conistently.

 

Using Parja's tiers as an example. I have no more confidence in Hines Ward, D-Jax, or L. Evans than I do in Wayne and Burress. Yet he has them in a seperate tier.

 

As others have pointed out, with the Wr posisiton being so inconsistent from year to year, I use my tiering more for the things I stated in my first post rather than as surefire ranking factor.

[t.j 35]

I agree with all of VaTerp's points about how the year-to-year inconsistency of WRs affects tiering.

[Let Da Big Dog Eat 42]

I apologize all. It just occured to me that I broke my own 1st Commandment of Posting. If the scoring system is of any relavance, post it. So here it is: 1/20/6, no ppr.

 

Good points all and I am sure that all of us who know anything about what they are doing, use tiering to a certain degree whether cognizent of it or not.

 

I was simply trying to assess how divergent people were in where/how they tier. The premise was that there would not be "hard tiers" but merely a kind of "red flag" approach that would help people to draft a balanced roster and not get caught saying to themselves "how did I end up with such crappy WR's?" It seems the premise is correct.

 

If anyone is interested in a deeper listing, it could be provided along with other positions. And if you want to post a scoring system, I will be happy to provide results. Won't happen until Sunday tho, golf outing then dinner, sex, drunk and pass out is the plan for tomorrow. Just another Saturday.

 

I would be interested in persuing this topic to see if there is divergence of opinion and why. Last, I think if I had provided a deeper pool it would have made a better topic. The next 25 WR's likely would have been harder to tier due to decreasing differences from one to the next.

[Gepetto 1,465]

All I know is it's a big mistake to have Lee Evans in the same tier as Randy Moss, Terrell Owens, and Marvin Harrison.

[Kopy 595]

I use tiers but I don't split them up based on gaps in the sorted projections. I do it based on how much I trust a guy to produce, compared to the players projected above and below him. So for example, based on the projections posted earlier in this thread, I might put my best guess (my projection) for Harrison and Ward as roughly equal. But still put Harrison in a higher tier because I think Ward's numbers rely too much on the assumption that he'll be the primary receiving target in the red zone again. Or something like that.

 

 

Same with me. I will base my tiers on the players themselves. Then rank the tiers accordingly useing my projections. I always have tiers with projections that do look alittle screwey, But It's worked for me.

An example for me (useing your numbers) would be putting Owens in tier 1, but still keeping guys like Ward and Driver in tier 2.

We also use ppr. scoreing also. So that alone affects alot of players.

[Parja 0]

For those of you who disagree with the tiers...

It doesn't mean you disagree with how the tiers are done. You're really just disagreeing with the FFToday projections. If you think a guy should be in a tier with another group of players, change his projections to get him there. Or, you might find that changing his stats to get him there is kind of a reach and he's really not worthy of being in that tier. It's that simple. Tiering is a valuable way to put together your own projections and vice versa.

[Franknbeans 46]

All I know is it's a big mistake to have Lee Evans in the same tier as Randy Moss, Terrell Owens, and Marvin Harrison.

 

[ratbastard1 0]

Tiering avoids the need for projections.

 

Is RB "x" worthy of a #1 in your league? If so then he should be in a tier above a guy who is only worthy of a #2 slot.

[Gepetto 1,465]

Tiering avoids the need for projections.

 

Is RB "x" worthy of a #1 in your league? If so then he should be in a tier above a guy who is only worthy of a #2 slot.

 

Actually, I believe projections are needed before you can tier players. How else do you tier players, on a hunch?

[Let Da Big Dog Eat 42]

Tiering avoids the need for projections.

 

Is RB "x" worthy of a #1 in your league? If so then he should be in a tier above a guy who is only worthy of a #2 slot.

 

I am confused about both parts of this post. If you don't project ANY kind of stats it would appear that tiers are purely arbitrary.

 

Second part of the post seems to say that you tier at 12, 24, 36. That also seems arbitrary as one may like the #25 WR way more than the #22 RB. According to your scenario, one would be obligated to take the #22 RB.

[ratbastard1 0]

You dont really need to project any true stats to know the first tier of RBs is LJ, LT2 and Alexander do you? Obviously i project domination out of those 3 RBs. I then break down the rest by who i would want as my #1 RB, #2 RB etc. (or even further 2a, 2b) I guess i project their tiers...sure, whatever.

 

Why do you need to say a RB will have 1400 yds and 10 tds? You dont, you just need to figure what tier that player represents to your league's roster slots. When i look at Ronnie Brown i consider his keeper value and his potential as a #1 RB. I can more easily predict what tier he falls in than how many exact TDs he's going to get.

 

As your tiers dwindle you obviously switch positions back and fourth. This is tiering which can be used in conjunction with a ranking sheet but doesnt need to. Lets say I have CBrown, TJones and CMart in the same tier. Im looking for my #3/4 RB. Why should I order them. When I get to them my squad may already be injury prone, old or RBBC and my team will influence the pick though i may have a favorite or take a different position.

 

Obviously tiering is somewhat projecting but id rather stay on top of the info than waste my time projecting stats that have NO chance of being correct. My tiers certainly can for many players.

 

Making the lists in tiers can be redone all the time with little effort. Add the projections in and you have to think too much to move players up and down. Gut feeling certainly comes into play, but dosnt it always?

[Slim Pickins 0]

Hm. I've always used the tiering system for a more relevant purpose: to ensure I draft the best available person at a position. As the tiering philosophy goes, if you are looking at a full tier of #3 WR and there is a tier 2 RB or a tier 1 TE, then you should draft from those positions. Tiering is relevant to you as you are saying everyone in that tier has about the same chance of scoring TOTAL points during the season. Whether projections, a hunch, homerism, or crystal ball, it doesn't matter. What matters is you have the best available person from each position. As for tier delineation, VaTerp explained why and Parja's use of 16 is a simple benchmark if you don't want to go through all the mental gymnastics (like me).

[TD Ryan2 316]

some good discussion in here.

here's a couple of my thoughts:

 

- I believe "tiering" for WRs this year is especially difficult as there really isn't an "Elite 3" as there has been in the past (Moss, TO, Harrison). The WR talent is more of a cluster than ever.

 

- I agree that you can tier players without stats projections.

 

- I use tiering to make sure that I get value during a draft.

 

 

If I were to "tier" your list from above, I guess it would go something like this:

 

Terrell Owens

Steve Smith

Torry Holt

Marvin Harrison

Randy Moss

___________________________

Chad Johnson

Larry Fitzgerald

Chris Chambers

Anquan Boldin

Reggie Wayne

___________________________

 

Hines Ward

Darrell Jackson

Donald Driver

Santana Moss

___________________________

 

T.J. Houshmandzadeh

Matt Jones

Joey Galloway

Plaxico Burres

Terry Glenn

___________________________

Lee Evans

Deion Branch

Eddie Kennison

Rod Smith

[ratbastard1 0]

I definately use them when drafting but i look at it like this and using your tiers which are not mine:

 

1A--I want one of these guys bad, i might overpay or draft early:

Terrell Owens

Steve Smith

Torry Holt

Marvin Harrison

Randy Moss

___________________________

 

1B--I want one of these guys as my #1 if I dont go WR in 1A:

Chad Johnson

Larry Fitzgerald

Chris Chambers

Anquan Boldin

Reggie Wayne

Lee Evans

Deion Branch

___________________________

 

1C--I will be fine with one of these guys as my #1 but will go WR very soon again if so:

Donald Driver

Santana Moss

Plaxico Burres

___________________________

 

2A--If I dont have 2 guys from the 1s already I will focus on making one of these my #2s:

T.J. Houshmandzadeh

Matt Jones

Joey Galloway

Terry Glenn

Lee Evans

Deion Branch

___________________________

 

2B--If Im still looking for my #2, i must get one of these guys or I can have a great #3:

Eddie Kennison

Rod Smith

 

etc

 

 

In the draft I usually go RBs early and often and this year is by far the worst to predict for WRs, I hate everyone of them except probably Harrison, and the Ariz guys. I break down the tiers but dont combine positions except for maybe the first 30-35 players which i often rank in order of preference.

 

Its really very much like projections just much less work which opens up more time to mock and read, and also tweak my tiers. Gut feeling comes into play after all other factors like sched and injuries etc are considered. I often go injury safe when i already have an injury prone team which i only know right then and there at that pick.

 

Roster slots are often different and knowing your league really helps.

[Bottles 0]

Here are the season point projections using my league scoring and FF Today projections. Where would you draw lines for your tiers? It seems like a pretty smooth curve with no real obviously breakpoints.

 

It's not that I don't understand the concept. It seems kind of arbitrary when you look at the raw stats. And, I don't agree with all the projections. So, I don't use tiers. I prefer to make picks based on what runs are going on in the draft and what I need.

 

Steve Smith CAR 139.3

Torry Holt STL 135.4

Larry Fitzgerald ARI 130.3

Chad Johnson CIN 124

Marvin Harrison IND 117

Hines Ward PIT 116.2

Donald Driver GB 114

Terrell Owens DAL 111

Chris Chambers MIA 107.6

Darrell Jackson SEA 105.3

Randy Moss OAK 103.7

Lee Evans BUF 103.5

Anquan Boldin ARI 102.8

Plaxico Burress NYG 99.8

Reggie Wayne IND 99.7

Deion Branch NE 97.5

Eddie Kennison KC 97.3

Matt Jones JAC 96.8

Rod Smith DEN 96.1

Joey Galloway TB 94.1

Terry Glenn DAL 90.5

Santana Moss WAS 90.3

T.J. Houshmandzadeh CIN 86.4

 

 

If you are supremely confident in these projections, and these numbers factor in varience in productivity due to injuries or other unknowns, then you are correct that there is no point in creating tiers. Others, might argue that the top 4 or 5 or however many are more "locks" to hit their projected numbers, and once you get below that, it's more of a crapshoot, hence a reason for at least two tiers.