moleculo
Footballguy
Over the years, I've done lots of different things for lots of different drafts. This will by my 13th year doing the fantasy football; I've tried all sorts of things, to varying success (including subscriptions to FBG, draft dominator and the like).
Without getting too much into the nuts and bolts, here is what I'm currently doing:
projections
projections are the foundation of any draft list. Projections are much better than rankings, because projections are used to compare across positions, and to determine tiers. With simple rankings, you really have no way to guess how close two people may or may not be. Projections also let you quickly customize per your leagues scoring - I play in multiple leagues with multiple scoring systems, and a simple ranking system won't show any differences across different leagues.
Projections don't have to be completely balanced (i.e. QB passing yards do not have to equal team receiving yards) and they don't have to be all that accurate. I say this because IMO, one should always project as if everyone will play 16 games, but for an injury prone RB, you might want to throw some extra opportunity towards his back-up.
In the past, I've done full-scale projections from scratch, which is a ton of work. In the end, it wasn't very good because, IMO, I don't have enough time to be an expert on 32 teams - not with a full time job, wife, and kids. Last year, I relied on some free projections as a base-line, and then did modifications as I saw fit. This hurt me because I was relying on assumptions made by the original projector, which may or may not have been great. As an example, the projections I had last year had Dominik Hixon as the NYG WR to have...I ended up reaching for him, which didn't work out. I would have been better off with pretty much any other NYG WR. I applaud the projector for making a call, but it was the wrong one and it cost me.
This year, I'm averaging 5 sets of projections that I've found over the net (including phenoms). I also look at differences across projections and tweaking players for whom there is lots of disagreement. If 5 independent projectors think Gore will score between 200 and 210 points, for example, my time is better spent thinking about and adjusting projections for other players. Using 5 projections will also hopefully average out projectors making calls that may or may not be any good.
Value
The next step is to translate full projections into FF points - pretty straight forward. Next, we calculate the VBD score, which is used to compare across positions and make a master list. Similar to the FBG method, I subtract a "baseline" from the projected score to develop a "value" score. The way I think of it, "value" is my appraisal of what a player is worth.
Baseline:
I calculate baseline by taking the # of league wide projected starters at a particular position, multiply by 1.5, and then average the projected FF score of that many players....for example, in a 12 man league, I average the top 18 QB FF scores. I do this because the baseline is not subject to one players score, it also means that no particular player is the "baseline". I came up with the 1.5 by trial and error, and I like the way it fits together.
ADP
Next, I download the latest ADP I can find. The ADP is my road map - I use this to estimate what my other 11 league mates will do, and to estimate who will be drafted soon. I think of ADP as the markets appraisal of what a players value is. Next, I compare my value with ADP, and that pretty clearly lets me know who is being drafted too high and who is being drafted too low, based on my projections. If I think someone is being drafted to soon, I won't take them at that position....maybe next round if he's there. If I think someone is being drafted too late, I'll go ahead and reach for him if that's what I feel I need to do. I also examine where my value and ADP differ, and make sure that I am ok with this.
At the draft
I don't like to nerd it up by using a laptop. There's enough nerd in the process to get here that I'm tired of computers at this point, and want to enter the draft with paper, a pen, and that's it. Here is my draft kit:
sheet 1: full blown unmodified ADP, 20+ rounds, with indications of who I think is over and under priced.
sheet 2. positional ranking list, with "value" reported. I use "value" here because it let's me compare across positions, and within one position, it would be exactly the same as projected FF points.
sheet 3: table to write down who my league-mates have taken, so I can estimate what positions are likely to be drafted in the next round - i.e. player X has two RB's and no WR's...good chance he won't draft a RB in the next round.
sheet 4: latest injury list
sheet 5: latest depth charts (useful in later rounds)
During the draft, I simply go down the ADP, look for someone who I think presents good value for where I'm picking, and try to fill out my team. I try to keep it as open as I possibly can: I hate thinking about "I'm gonna draft RB-RB-QB-WR", because if you do that, you are blind to value that drops to you. I want the draft to come to me. I also look at tiers and drop-offs, think about who will be available next round, and all of that. I will even skip order of my lists - this is where I use my gut. Sometimes I just have a feeling, and it's ok to deviate from your lists. In the end, all of the above is simply a guide - it's not a law.
In a nutshell, I find I'm relying on ADP to a pretty heavy degree.I acknowledge that it's ok to reach for a player, but I only want to reach if I think it's worth it. I think that drafting with ADP as a guide might be unique - I have no idea of anyone else doing it this way. The FBG way is to use VBD for the first 55 picks or so, and then to fill out your team. Like Timsochet found, simply looking at your VBD can cause you to draft players earlier than you should; it's ok to do that but we should be conscious of when we overpay.
I'm open to comments, criticisms, and feedback.
Without getting too much into the nuts and bolts, here is what I'm currently doing:
projections
projections are the foundation of any draft list. Projections are much better than rankings, because projections are used to compare across positions, and to determine tiers. With simple rankings, you really have no way to guess how close two people may or may not be. Projections also let you quickly customize per your leagues scoring - I play in multiple leagues with multiple scoring systems, and a simple ranking system won't show any differences across different leagues.
Projections don't have to be completely balanced (i.e. QB passing yards do not have to equal team receiving yards) and they don't have to be all that accurate. I say this because IMO, one should always project as if everyone will play 16 games, but for an injury prone RB, you might want to throw some extra opportunity towards his back-up.
In the past, I've done full-scale projections from scratch, which is a ton of work. In the end, it wasn't very good because, IMO, I don't have enough time to be an expert on 32 teams - not with a full time job, wife, and kids. Last year, I relied on some free projections as a base-line, and then did modifications as I saw fit. This hurt me because I was relying on assumptions made by the original projector, which may or may not have been great. As an example, the projections I had last year had Dominik Hixon as the NYG WR to have...I ended up reaching for him, which didn't work out. I would have been better off with pretty much any other NYG WR. I applaud the projector for making a call, but it was the wrong one and it cost me.
This year, I'm averaging 5 sets of projections that I've found over the net (including phenoms). I also look at differences across projections and tweaking players for whom there is lots of disagreement. If 5 independent projectors think Gore will score between 200 and 210 points, for example, my time is better spent thinking about and adjusting projections for other players. Using 5 projections will also hopefully average out projectors making calls that may or may not be any good.
Value
The next step is to translate full projections into FF points - pretty straight forward. Next, we calculate the VBD score, which is used to compare across positions and make a master list. Similar to the FBG method, I subtract a "baseline" from the projected score to develop a "value" score. The way I think of it, "value" is my appraisal of what a player is worth.
Baseline:
I calculate baseline by taking the # of league wide projected starters at a particular position, multiply by 1.5, and then average the projected FF score of that many players....for example, in a 12 man league, I average the top 18 QB FF scores. I do this because the baseline is not subject to one players score, it also means that no particular player is the "baseline". I came up with the 1.5 by trial and error, and I like the way it fits together.
ADP
Next, I download the latest ADP I can find. The ADP is my road map - I use this to estimate what my other 11 league mates will do, and to estimate who will be drafted soon. I think of ADP as the markets appraisal of what a players value is. Next, I compare my value with ADP, and that pretty clearly lets me know who is being drafted too high and who is being drafted too low, based on my projections. If I think someone is being drafted to soon, I won't take them at that position....maybe next round if he's there. If I think someone is being drafted too late, I'll go ahead and reach for him if that's what I feel I need to do. I also examine where my value and ADP differ, and make sure that I am ok with this.
At the draft
I don't like to nerd it up by using a laptop. There's enough nerd in the process to get here that I'm tired of computers at this point, and want to enter the draft with paper, a pen, and that's it. Here is my draft kit:
sheet 1: full blown unmodified ADP, 20+ rounds, with indications of who I think is over and under priced.
sheet 2. positional ranking list, with "value" reported. I use "value" here because it let's me compare across positions, and within one position, it would be exactly the same as projected FF points.
sheet 3: table to write down who my league-mates have taken, so I can estimate what positions are likely to be drafted in the next round - i.e. player X has two RB's and no WR's...good chance he won't draft a RB in the next round.
sheet 4: latest injury list
sheet 5: latest depth charts (useful in later rounds)
During the draft, I simply go down the ADP, look for someone who I think presents good value for where I'm picking, and try to fill out my team. I try to keep it as open as I possibly can: I hate thinking about "I'm gonna draft RB-RB-QB-WR", because if you do that, you are blind to value that drops to you. I want the draft to come to me. I also look at tiers and drop-offs, think about who will be available next round, and all of that. I will even skip order of my lists - this is where I use my gut. Sometimes I just have a feeling, and it's ok to deviate from your lists. In the end, all of the above is simply a guide - it's not a law.
In a nutshell, I find I'm relying on ADP to a pretty heavy degree.I acknowledge that it's ok to reach for a player, but I only want to reach if I think it's worth it. I think that drafting with ADP as a guide might be unique - I have no idea of anyone else doing it this way. The FBG way is to use VBD for the first 55 picks or so, and then to fill out your team. Like Timsochet found, simply looking at your VBD can cause you to draft players earlier than you should; it's ok to do that but we should be conscious of when we overpay.
I'm open to comments, criticisms, and feedback.
Last edited by a moderator:
