Obviously VBD is extremely useful. Everyone has to admit that individual rankings are a crap shoot at best ---- Does anyone forgo the usual rankings and just go with groupings by tier ---- ie HOLT, CHAD, SMITH are the WR1 tier and it does not matter which one you get --- just wondering as year to year tiers are pretty consistent... Thoughts please ---
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!
Drafting Strategy -- tiering ---- (1 Viewer)
- Thread starter jziccfbg
- Start date
Captain Quinoa
Footballguy
I do that to some extent. Usually not with the top guys, but moreso in the middle rounds where I'm thinking "as long as I can get one of X,Y,Z as my #2 WR I'll be okay with that".
lloyd christmas
Footballguy
I always have tiers blocked out on my draft sheets. When you are drafting near the ends of rounds its very helpful to see how many players are available of similar expectations. I have Value numbers on the sheet as well, to compare between positions.For example, youre drafting 9th, its the end of the third round. There could be four WR's remaining in a given tier, but only one RB in the best tier. If you can reasonably expect any one of the four WR's to be there as the draft comes back to you in the fourth, then the RB selection is a no brainer. Tracking owners selections between you and the turn picks is essential here, and of course ADP plays a part as well.
bostonfred
Footballguy
A three player tier like that doesn't really help much unless you pick at the turn or can easily trade draft picks mid-draft. There aren't a lot of decisions you get to make where you know you could get Smith if you passed on Holt, or vice versa. In most practical situations, it only matters if a tier spans this pick and your next pick.
In those cases, from a VBD perspective, what you're really doing is saying that the dropoff from the player at the top of your tier to the player at the bottom, if there is one at all, is not as great as the dropoff would be at another position if you passed up on it.
I think most of us are willing to concede that your player projections aren't that accurate. That's not really the issue. The question, I think, is what your team looks like, and what you're trying to accomplish in that specific league.
The bigger the league is, the more a slow and steady approach will work. The smaller it is, the more you need the absolute top players. But if you're in a big tournament format, you not only need to maximize your chances of winning your league, you need to maximize your chances of winning it all. Slow and steady doesn't work as well there.
But you also want to consider your team. Say it's the fourth round, and you're looking at two RBs in the same tier. One, you think has a chance to be a top 5 RB if things work out, but he's risky. A guy like DeAngelo Williams. The other is going to be slow and steady, but probably won't be top 10. Maybe Curtis Martin. Note that I'll use those names as an example only; I don't really care if those assumptions are true. And let's assume that you're dead set on taking one of these two guys, because they're the best available players on your list, and the last two guys in their "tier".
Let's say you have a start 2 RB league, and you already have your stud RB1, WR1, and Manning. There's a good case for taking Martin at this point, since you already have studs at the top of your draft. And I'd almost definitely do this if I were in a normal league, because I'd think I had enough firepower at this point that I just want a guy who will produce points.
Change that team around, though, and say you already have two RBs. Now, for whatever reason, you think that Martin and DeAngelo are the two best available players here. I'd lean towards DeAngelo if my two RBs seemed pretty safe, or had good handcuffs. I'd lean towards Martin if I'd taken a risky RB2, like a Dayne.
Or say I was in a big tournament style league, where the emphasis was on total points for the season, and there were hundreds or thousands of players. I might be more apt to take a DeAngelo, in the hopes that he would carry my team late in the season, and I'd concede some points early at RB2 to make up for it. But if that same league emphasized making an early playoff after the first 11 weeks, I might be more apt to take the guy I thought would get surer points.
In other words, I think there's a lot more to drafting than just tiering. I think tiering helps, but building a winning team takes more.
In those cases, from a VBD perspective, what you're really doing is saying that the dropoff from the player at the top of your tier to the player at the bottom, if there is one at all, is not as great as the dropoff would be at another position if you passed up on it.
I think most of us are willing to concede that your player projections aren't that accurate. That's not really the issue. The question, I think, is what your team looks like, and what you're trying to accomplish in that specific league.
The bigger the league is, the more a slow and steady approach will work. The smaller it is, the more you need the absolute top players. But if you're in a big tournament format, you not only need to maximize your chances of winning your league, you need to maximize your chances of winning it all. Slow and steady doesn't work as well there.
But you also want to consider your team. Say it's the fourth round, and you're looking at two RBs in the same tier. One, you think has a chance to be a top 5 RB if things work out, but he's risky. A guy like DeAngelo Williams. The other is going to be slow and steady, but probably won't be top 10. Maybe Curtis Martin. Note that I'll use those names as an example only; I don't really care if those assumptions are true. And let's assume that you're dead set on taking one of these two guys, because they're the best available players on your list, and the last two guys in their "tier".
Let's say you have a start 2 RB league, and you already have your stud RB1, WR1, and Manning. There's a good case for taking Martin at this point, since you already have studs at the top of your draft. And I'd almost definitely do this if I were in a normal league, because I'd think I had enough firepower at this point that I just want a guy who will produce points.
Change that team around, though, and say you already have two RBs. Now, for whatever reason, you think that Martin and DeAngelo are the two best available players here. I'd lean towards DeAngelo if my two RBs seemed pretty safe, or had good handcuffs. I'd lean towards Martin if I'd taken a risky RB2, like a Dayne.
Or say I was in a big tournament style league, where the emphasis was on total points for the season, and there were hundreds or thousands of players. I might be more apt to take a DeAngelo, in the hopes that he would carry my team late in the season, and I'd concede some points early at RB2 to make up for it. But if that same league emphasized making an early playoff after the first 11 weeks, I might be more apt to take the guy I thought would get surer points.
In other words, I think there's a lot more to drafting than just tiering. I think tiering helps, but building a winning team takes more.
Biabreakable
Footballguy
Very good point about comparing upside of a player against another player that has shown more consistency and is more proven, but also not likely to have the same upside.
Looking for tiers and cutoffs at positions can be useful before the draft begins. As it can tell you if it is important to get 2 Rbs right away or if you should consider wr qb or te. You can count and predict what kind of cutoffs to expect if you do not get a player before the tier drops off.
As the flow of the draft goes you can have a moving tier that includes all positions. This tier is based off of how many players will be taken before your next pick. You can then look at worst case scenarios of what players you can expect to get with your next picks assuming that all players in your moving tier WILL be taken. If the other owners deviate from this then great you have another player available from that tier each time they do.
Tiers can be especialy useful if you are able to trade mid-draft. If you see your pick falling a couple slots after a tier dropoff then you might want to trade up to get at least to the bottom of that tier. Similarly if you see close to equal value in a large bunch of players for example 16 players or enough to take you on to your next pick then why not trade down? And see what you can get for it while possibly stil being able to draft a player in the same tier.
Looking for tiers and cutoffs at positions can be useful before the draft begins. As it can tell you if it is important to get 2 Rbs right away or if you should consider wr qb or te. You can count and predict what kind of cutoffs to expect if you do not get a player before the tier drops off.
As the flow of the draft goes you can have a moving tier that includes all positions. This tier is based off of how many players will be taken before your next pick. You can then look at worst case scenarios of what players you can expect to get with your next picks assuming that all players in your moving tier WILL be taken. If the other owners deviate from this then great you have another player available from that tier each time they do.
Tiers can be especialy useful if you are able to trade mid-draft. If you see your pick falling a couple slots after a tier dropoff then you might want to trade up to get at least to the bottom of that tier. Similarly if you see close to equal value in a large bunch of players for example 16 players or enough to take you on to your next pick then why not trade down? And see what you can get for it while possibly stil being able to draft a player in the same tier.
Wreckincrew
Footballguy
Tiering should just be a tool to help you make your decision easier when you are on the clock. I use it to help me assign value when I pick. I base mine off of ADP and my draft position and adjust players I like up or down.
For example, I am drafting 8th in a ten team league. I have the the top eight players in one color, the next 5 in another color, the next 15 in another, next 5, next 15 and so on. If a player falls two or three tiers I know I will be getting value if I choose him. It also helps me get in on the front end of position runs (i.e. a 3rd round run on WR's).
Like I said earlier, it isn't the be-all or end-all of lists, you use it to help you adjust to the way that each particular draft lays out.
For example, I am drafting 8th in a ten team league. I have the the top eight players in one color, the next 5 in another color, the next 15 in another, next 5, next 15 and so on. If a player falls two or three tiers I know I will be getting value if I choose him. It also helps me get in on the front end of position runs (i.e. a 3rd round run on WR's).
Like I said earlier, it isn't the be-all or end-all of lists, you use it to help you adjust to the way that each particular draft lays out.
Here is an example of where tiering presents a problem IMO....let's say that you have projected the following points:
1 Randy Moss 319 ADP 1.10 2 Torry Holt 307 ADP 2.01 3 Terrell Owens 304 ADP 2.04 4 Chad Johnson 300 ADP 3.01 5 Chris Chambers 298 ADP 3.09Now Moss, Holt and Owens are in Tier 1. Let's say your on the clock (2.08) and and Moss and Holt are off the board. Owens, CJ and Chambers are within 6 points of each other, yet are in different tiers.So if you were stricly following tiers, you would select TO b/c he was in the first tier. However, you can likely get Chambers at your next pick with only a 6 point differential.
This has led me away from tiering, b/c at the end of the day, IMO it should the difference in projected fantasy points of the remaining players in the draft pool that drive your decision to draft a player, not what tier he's in.
1 Randy Moss 319 ADP 1.10 2 Torry Holt 307 ADP 2.01 3 Terrell Owens 304 ADP 2.04 4 Chad Johnson 300 ADP 3.01 5 Chris Chambers 298 ADP 3.09Now Moss, Holt and Owens are in Tier 1. Let's say your on the clock (2.08) and and Moss and Holt are off the board. Owens, CJ and Chambers are within 6 points of each other, yet are in different tiers.So if you were stricly following tiers, you would select TO b/c he was in the first tier. However, you can likely get Chambers at your next pick with only a 6 point differential.
This has led me away from tiering, b/c at the end of the day, IMO it should the difference in projected fantasy points of the remaining players in the draft pool that drive your decision to draft a player, not what tier he's in.
Biabreakable
Footballguy
a 16 point tier is one point/game and although it makes a difference if your projections are spot on (unlikely) I don't think it is significant enough for me to consider it a drop to the next tier.
My seperation of tiers is more like 25-30 pts EOY
of course other considerations go into where I place a tier drop than just projections.
My seperation of tiers is more like 25-30 pts EOY
of course other considerations go into where I place a tier drop than just projections.
Ray Pollett
Footballguy
Since I've never seen anyone that consistently projects even close - I do use tiers. But it is not set up that stiffly. There may be 3 in tier one and 9 in tier 2; 15 in tier 3; and 5 in tier 4. It is more groupings of players at the same position that I think can give me about equall production. You look at your needs basd on starting line up, roster size and the flow of the draft. You decide which type of player that is left in the tiers gives you the best value and take one; Yes they are ranked in the tier, but that is really a feeling among players in the same tier.
NYG Endzone
Footballguy
Here's my RB tier as an example (I like having at least 5 players in a tier so I do more player analysis in my brain, instead of just going off the cheatsheet #'s) ...
1 Johnson
2 Tomlinson
3 Alexander
4 Barber
5 Jackson
6 Portis
7 Jordan
8 Brown
9 Johnson
10 James
11 Davis
12 Williams
13 Westbrook
14 McGahee
15 Droughns
16 Bell
17 Jones
18 Parker
19 Bush
20 Taylor
21 Lewis
22 Jones
23 Dunn
24 Foster
25 Taylor
26 Addai
27 Jones
28 Martin
29 Dayne
30 Dillon
31 McAlliste
1 Johnson
2 Tomlinson
3 Alexander
4 Barber
5 Jackson
6 Portis
7 Jordan
8 Brown
9 Johnson
10 James
11 Davis
12 Williams
13 Westbrook
14 McGahee
15 Droughns
16 Bell
17 Jones
18 Parker
19 Bush
20 Taylor
21 Lewis
22 Jones
23 Dunn
24 Foster
25 Taylor
26 Addai
27 Jones
28 Martin
29 Dayne
30 Dillon
31 McAlliste
Ramblin Wreck
Footballguy
You make an excellent point. I think the answer to your issue is dynamic tiers.
Biabreakable
Footballguy
I am curious about what your definition of dynamic tiers may be.Personaly I do not make static projections but projections that have a low (worst case) and a high (upside) projection range.
As BostonFred was talking about some players are more proven and solid in the numbers you can expect from them like Torry Holt who has had over 1300 yards for 6 seasons in a row now. Moss has more upside because of the high number of TDs he has scored but has a much lower worst case number to consider.
At different points during a draft you may want to take a player with greater upside than predictability and vise versa based off of the other players you allready have in the draft.
this is too big to be effective IMOfor example your first tier should be broken up into at least 2 if not 3. From there you'd know to trade up for player X if you wanted him.
If you don't know where to break it, then (for me) that tells me I need to do more research(stats and articles) to determine where the break should be and form a definitive opinion/prediction for the upcoming season.
Last edited by a moderator:
M
MLBrandow
Guest
Tier off the ADP rankings, then tier off your personal rankings.
Then assign +/- to guys higher/lower.... be a draft ninja.
Then assign +/- to guys higher/lower.... be a draft ninja.
BigRed
Footballguy
Or an auction draft, which is the only way to go
I have tiers with anywhere from 1-2 (pretty rare) to many players - 3-5ish probably the most common. It's all about perceiving the gaps. There is no "too small" or "too big" grouping IMO. eg my top QB tier is just Manning, but my 2d tier RBs have a lot of players.so once again I ask...how do we determine that Manning deserves to be in a tier all by his lonesome, while several RB's are in tier 2?
I do not disagree, mind you, that this is the case...
but for us
, can we have the logic behind such placement of players w/in positional tiers?
Ramblin Wreck
Footballguy
At each pick, re-tier your draft board. If you know you want a RB, WR, and QB with your next three picks (Let's say you have 3 of the next 25 picks), re-tiering allows you to see how many of each position you think will make it 25 picks later.
BigRed
Footballguy
There is no simple answer to that question; it all depends on the specifics of each player/players. Many factors come into consideration, eg past performance, reliability, upside, injury risk, schedule, rest of the team, etc etc etc.so once again I ask...
how do we determine that Manning deserves to be in a tier all by his lonesome, while several RB's are in tier 2?
I do not disagree, mind you, that this is the case...
but for us
bostonfred
Footballguy
Good call about the auction. I don't write them out necessarily, but I definitely use tiers at an auction.Or an auction draft, which is the only way to go
bostonfred
Footballguy
What use is there to tiering if your tiers only include players you could take at this specific pick? Wouldn't you usually just take the best guy then (with the exception of picking the best "fit" I was talking about above)?
bostonfred
Footballguy
Yeah, that's kind of what I mean, although I'm not just talking about having a couple guys I have projected to do well that have a low ADP. In my ideal world, I'd break that out into two tiers - one that include Holt and Smith and whoever else, and another that has Bennett way ahead of the others in the next tier down. I tend to go with more of an ADP based tiering, where I see a bunch of guys that will likely come off the board in the same range, and if I perceive them as being different, I try to take the best one early; if I perceive them as being virtually identical, I try to take the last one (or two if I'm lucky) in the tier.
One multi-round tier I see most years is the second tier of receivers, which usually spans the third and fourth rounds, but this year, seems to span the second and third. Usually there's 8 or so top guys that you'd take as your WR1, and I'm usually willing to let others take their first choice of those guys if it means I can improve at another position and still get a "WR1".
This year that top tier is again, about eight guys: CJ, Holt, Smith, Harrison, Owens, Fitzgerald, Boldin and Moss. Ask anyone who their favorites are, and you'll be surprised at the responses (mine are Owens and Harrison, with CJ, Smith next, Boldin and Fitzgerald next, and some people's #1 Holt is 7th on my list). If I'm in a draft where I think one of those guys would slip to my next pick, I'd pass on the first one, and take whoever's left at the bottom of that tier at my next pick.
To me, that's the right way to look at tiering. Whether I get a guy in that tier or not, I'll still take Bennett later if I think he's a sure #1, and I'd take him at or near the top of the next tier to make sure I get him because, like I said, he's going to get drafted somewhere in the next group of WRs, and I want my first choice of that next group.
The idea of tiering using 16 point VBD stuff isn't really meaningful to what I just described, though, so maybe what I'm doing isn't really what the rest of you guys are talking about when you say tiering. I don't really know.
Ray Pollett
Footballguy
"It is more groupings of players at the same position that I think can give me about equall production. " Everyone knows this pretty much. You see thread after thread on who do i TAKE, S Jackson, L Jordan, C Williams, or Ronnie Brown - Some choose Brown some choose Caddy, some choose Jackson, some choose Jordan. You look at the ADP sheets, You look at the articles about players. They are about equall so they go in the same tier. Most people have a feeling on this with out doing projections. I believe my instincts are as good if not better than most peoples projections.OK...I'm curious
you assume everyone's projections suck
how do we then determine the tiers...like:
How many?
Points between them?
Rank players w/in a tier?
I'll hang up and listen...
bostonfred
Footballguy
Good post. I agree that the consensus is that they're about equal. The thing is, if they were about equal, you'd always want to pick as late as possible in that tier, right? And if they weren't about equal, you wouldn't want to get stuck with last one, or maybe you'd even have a favorite that you wanted. Whether you call it instinct, or separate players into 16 point tiers, it's still the same kind of decision making tool. FWIW, tiering isn't a big deal when you're deciding between Jackson, Jordan, Brown and Caddilac with your first pick - they'll all be gone by your next pick, so you'd always take your favorite. Tiers really start to become an issue in the second (maybe late first) and become more important in subsequent rounds, IMO.
Biabreakable
Footballguy
I understand what your talking about Fred and I do the same thing. However that is applying ADP to your tier/projections which is another step I think.
Ramblin Wreck thanks for elaborating on what dynamic means. Fluid. Always good to continue adjusting along the way.
The usefulness I find in tiers is determining how to make the best use of each pick by staying ahead of what the other owners will do by expecting them to take all the best players on my board. If a player or more slips then gravy my tier just grew and I adjust to that. If I see a big drop off happening before my next pick then I start burning up the phone lines to trade up to where I see I want to be. If that fails then Im in the next tier and can start looking at trading down if the bucket just got deep.
Ramblin Wreck thanks for elaborating on what dynamic means. Fluid. Always good to continue adjusting along the way.
The usefulness I find in tiers is determining how to make the best use of each pick by staying ahead of what the other owners will do by expecting them to take all the best players on my board. If a player or more slips then gravy my tier just grew and I adjust to that. If I see a big drop off happening before my next pick then I start burning up the phone lines to trade up to where I see I want to be. If that fails then Im in the next tier and can start looking at trading down if the bucket just got deep.
I feel similarly and don't bother with projections. I feel like they're a waste of time for me because this works. If it didn't and I needed a way to define how well a player might perform then I'd roll with projections. I do see it's usefulness for others, I just don't care for em'.
Tick
Footballguy
I have done this in the past and been very happy with the results. The key to me for doing this is to have meaningful ADP. For example, the No Mercy leagues have eight leagues with identical rules. If you're in one of those leagues, you can get good ADP from the other drafts and rely on it. If you're just looking at general Antsports ADP, I don't trust it and fall back on my own projection tiers.
Marc Levin
Hangs out with Oscar Zeta Acosta
I look for cutoffs, but not tiers.
It is almost never true that I am "just as happy" with every guy in a certain grouping.
However, it is almost always true that I need to remind myself when the cutoff before a big positional dropoff is coming so I can figure out if I need to jump on a player (like, to maximize value or to avoid being too far behind at a position)
It is almost never true that I am "just as happy" with every guy in a certain grouping.
However, it is almost always true that I need to remind myself when the cutoff before a big positional dropoff is coming so I can figure out if I need to jump on a player (like, to maximize value or to avoid being too far behind at a position)
Captain Quinoa
Footballguy
That doesn't logically follow.XThen why would I beat you something like 7-3 if I were to draft against you head to head in a 10 team league?
Because my rankings are better than your's.
Captain Quinoa
Footballguy
Captain Quinoa
Footballguy
Larry Johnson and Ronnie Brown in the same tier? Those aren't tiers those are continents.
NYG Endzone
Footballguy
Hey Superstar. Your going to able to take Ronnie AND LJ in the draft? Otherwise, I dont follow your logic. Most of the time during the draft at least half of the tier will be crossed out as they have been taken. I account for this and make decisions on players, no one or two man tiers for me.
Val Rannous
Footballguy
Spot on. There is always that "Gee, the rest of these guys suck" point to watch out for. You will usually find 3-4 of these points in each position ranking lists, and that is your "tier". This is excellent for determining value picks in the middle of the draft (earlier you are drafting for studs, later for projects). You can tell which positions have the most room left to wait and still get a good player, and which you need to jump on right now.
Val
Captain Quinoa
Footballguy
Eeeasy there, Turbo. Do it however you like. They just strike me as big tiers. I don't tier at the top because draft position is going to dictate who you get there, but after that I'm usually looking at a group of 2-4 guys whom I will accept as "RB2", "WR2", "QB1", etc... and those become my "tiers".
Wow -- nothing like starting a thread -not being around for a couple of days and then seeing 40 postings ----
In terms of tiering --I take historical tiers and then slightly modify them for example this years tiers for RBS (# of players per tier) 2, 5, 5, 5.... WRs (3, 5, 8, 7) TE (2,2,2,3) etc. You might say why have tiers with only 2 or 3 in them --- It's because historically this is how they have fallen. Obviously, there is some subjectivity in assigning the point breakdowns of the tiers but when making the tiers it really forces you to assign your player with the grouping that inherently makes sense ---- ie 2nd group of WRs =5 (harrison, fitz, boldin, moss and TO) all are relatively equal (in my mind) and "belong" in the same tier --- If I threw Chambers or Wayne in that group (in my mind) they would not "fit" ---- My point is tiers help you to modify your projections and group relatively the same value players...
In terms of tiering --I take historical tiers and then slightly modify them for example this years tiers for RBS (# of players per tier) 2, 5, 5, 5.... WRs (3, 5, 8, 7) TE (2,2,2,3) etc. You might say why have tiers with only 2 or 3 in them --- It's because historically this is how they have fallen. Obviously, there is some subjectivity in assigning the point breakdowns of the tiers but when making the tiers it really forces you to assign your player with the grouping that inherently makes sense ---- ie 2nd group of WRs =5 (harrison, fitz, boldin, moss and TO) all are relatively equal (in my mind) and "belong" in the same tier --- If I threw Chambers or Wayne in that group (in my mind) they would not "fit" ---- My point is tiers help you to modify your projections and group relatively the same value players...
My other main purpose is tiering is to help me (loosely) to remain in some discipline in terms of relative worth ---- tier 6 RB (ranked 24-31) are not "worth" as much as tier 3 WRs (ranked 9-16) --- All of this is (normally) inherently obvious (especially the more experienced ff'rs) but when you are doing multiple drafts at the same time it does help...
bostonfred
Footballguy
I'm not arguing against taking two players from a tier, although there's nothing ideal about getting two players in the same tier in and of itself. When you take that first player, if you consider them virtually identical, you're "paying more for them" than the second one. The second player is by definition a better value, unless you think there's a significant difference between the top and bottom players in that tier.
Similar threads
