How do you do rankings? (1 Viewer)

[lebowski]

Just curious how fellow FBG's do their rankings. I personally tweek projections in the projection dominator.

 

[The Kansas Comet]

All of the above. I make my own projections, using statistics and gut feelings to come up with my numbers. Then I plug them into the Projections Dominator (I weight my numbers 20-25%), and then break the resulting lists into tiers.

 

[jurb26]

I typically start from scratch (unless I'm feeling lazy) making note of personal changes I feel are noteworthy and then adjust for talent. Rarely I adjust for injury risk. Then, I take a break (about a week) and come back to digest the information I've now put on paper (excel sheet typically) and tweek based on some gut feelings.

So I voted "other."

 

[The Man with the Plan]

What I do is i'll take a look at any standard cheatsheet. And then I go down the list and see if there's a player that I like better than the player him. I'll make an adjustment. For example:

Let's say the first five running backs are listed like this:

LaDainian Tomlinson

Larry Johnson

Steven Jackson

Frank Gore

Joseph Addai

I see that LT is ranked above Larry Johnson. Then I ask myself do I like him than better Larry Johnson(the next ranked player on the cheatsheet) this season. Since I do then I rank LT the #1 running back. I type his name out onto a spreadsheet/database and cross his name off the cheatsheet.

Now I look at Larry Johnson and ask myself do I like him better than the next ranked player on the cheatsheet which is Steven Jackson. Since I do not I'll put Steven Jackson ahead of him for the time being. And because there is no other running back for me to put Steven Jackson behind I'll rank him the #2 running back. He's crossed off the list. You're wondering what happens with Larry Johnson.

I'll compare Larry Johnson to the remaining running backs on the cheatsheet. Until I find one that he's better than. For argument's sake (not necessarily my actual belief) let's say I like him less than Frank Gore but more than Joseph Addai. That means he would be the #4 running back.

This process is the same for every running back on the cheatsheet. Once it's completed I now have my ranking of the top 60 rbs. I then do this for every position and player on the cheatsheet and have made my own ranking list. For now all the positions are ranked separately (top 30 QB, 60 RB, 60 WR, 30 TE) Then as things happen during training camp and preseason I tweak the list using the same process of "Do I like this guy better than this guy"?

The projections part is actually the easiest part of the process for me. I use a 3-year AVT list and match the point total to where I have the players ranked. And it's from the combination of the ranking list and the AVT list that I use to create a final VBD list that ranks the player by their value number and I draft off that list (for the most part).

I've probably made it sound really complicated but that's mainly because I suck with words. Trust me it's ridiculously simple and a lot less stressful than how I've done it in the past. I had the projections done in January and the ranking list done in May. Now it's just a matter of getting as much good information as possible between now and the draft and tweaking the list accordingly.

P.S. By AVT I mean Average Value Theory. Here a link that tells you all you need to know. Average Value Theory

 

[dirtywaters20]

What I do is i'll take a look at any standard cheatsheet. And then I go down the list and see if there's a player that I like better than the player him. I'll make an adjustment. For example:

Let's say the first five running backs are listed like this:

LaDainian Tomlinson

Larry Johnson

Steven Jackson

Frank Gore

Joseph Addai

I see that LT is ranked above Larry Johnson. Then I ask myself do I like him than better Larry Johnson(the next ranked player on the cheatsheet) this season. Since I do then I rank LT the #1 running back. I type his name out onto a spreadsheet/database and cross his name off the cheatsheet.

Now I look at Larry Johnson and ask myself do I like him better than the next ranked player on the cheatsheet which is Steven Jackson. Since I do not I'll put Steven Jackson ahead of him for the time being. And because there is no other running back for me to put Steven Jackson behind I'll rank him the #2 running back. He's crossed off the list. You're wondering what happens with Larry Johnson.

I'll compare Larry Johnson to the remaining running backs on the cheatsheet. Until I find one that he's better than. For argument's sake (not necessarily my actual belief) let's say I like him less than Frank Gore but more than Joseph Addai. That means he would be the #4 running back.

This process is the same for every running back on the cheatsheet. Once it's completed I now have my ranking of the top 60 rbs. I then do this for every position and player on the cheatsheet and have made my own ranking list. For now all the positions are ranked separately (top 30 QB, 60 RB, 60 WR, 30 TE) Then as things happen during training camp and preseason I tweak the list using the same process of "Do I like this guy better than this guy"?

The projections part is actually the easiest part of the process for me. I use a 3-year AVT list and match the point total to where I have the players ranked. And it's from the combination of the ranking list and the AVT list that I use to create a final VBD list that ranks the player by their value number and I draft off that list (for the most part).

I've probably made it sound really complicated but that's mainly because I suck with words. Trust me it's ridiculously simple and a lot less stressful than how I've done it in the past. I had the projections done in January and the ranking list done in May. Now it's just a matter of getting as much good information as possible between now and the draft and tweaking the list accordingly.

P.S. By AVT I mean Average Value Theory. Here a link that tells you all you need to know. Average Value Theory

I just started doing something similar. How do you account for each player's value relative to players at a different position (for example Larry Johnson to Peyton Manning). I just started in a 2 QB, 2 RB, 4WR + 1 flex (RB, WR, or TE) & am unsure of what baseline to use to compare the value across the positions.

 

[tomarken]

I put them in tiers, after doing my own projections from scratch, which involves a lot of stats and a little gut feeling. I only use FBG projections to compare to mine to see if I'm way off base somewhere - and even then I don't necessarily change my projections, I just note that I'm going against the grain and maybe see if there is another perspective I'm not considering.

Doing your own projections is definitely the way to go; I started doing it a couple years ago and it makes the game so much more fulfilling. I know on draft day I'm drafting who I want, not what some expert/FBG/magazine/whatever says I should draft. That way even if I totally #### the bed, I know I went with my gut and I won't be pissed off that I relied on someone else's ideas for my team.

 

[LionsFan78]

I start off with a list of players, usually cut/pasted from the site, and start from there. I'll put everyone into an excel spreadsheet, color-coded by tiers and roles, and add-in last year's stats, newslinks of interest, comments etc. I'll conduct multiple mock drafts (Done about 37 so far ), and make notes of which players I drafted ahead of other players, then I tweak my list according to how I drafted each player. That's pretty much it. Then at least once a week I'll run down my list and see if my feelings have changed on a specific player.

 

[Mystery Achiever]

The projections part is actually the easiest part of the process for me. I use a 3-year AVT list and match the point total to where I have the players ranked. And it's from the combination of the ranking list and the AVT list that I use to create a final VBD list that ranks the player by their value number and I draft off that list (for the most part).

I've probably made it sound really complicated but that's mainly because I suck with words. Trust me it's ridiculously simple and a lot less stressful than how I've done it in the past. I had the projections done in January and the ranking list done in May. Now it's just a matter of getting as much good information as possible between now and the draft and tweaking the list accordingly.

P.S. By AVT I mean Average Value Theory. Here a link that tells you all you need to know. Average Value Theory

fwiw, see post 12 in this thread for fbg thoughts on avt; yes, this thread!http://forums.footballguys.com/forum/index...howtopic=328615

 

[The Man with the Plan]

What I do is i'll take a look at any standard cheatsheet. And then I go down the list and see if there's a player that I like better than the player him. I'll make an adjustment. For example:

Let's say the first five running backs are listed like this:

LaDainian Tomlinson

Larry Johnson

Steven Jackson

Frank Gore

Joseph Addai

I see that LT is ranked above Larry Johnson. Then I ask myself do I like him than better Larry Johnson(the next ranked player on the cheatsheet) this season. Since I do then I rank LT the #1 running back. I type his name out onto a spreadsheet/database and cross his name off the cheatsheet.

Now I look at Larry Johnson and ask myself do I like him better than the next ranked player on the cheatsheet which is Steven Jackson. Since I do not I'll put Steven Jackson ahead of him for the time being. And because there is no other running back for me to put Steven Jackson behind I'll rank him the #2 running back. He's crossed off the list. You're wondering what happens with Larry Johnson.

I'll compare Larry Johnson to the remaining running backs on the cheatsheet. Until I find one that he's better than. For argument's sake (not necessarily my actual belief) let's say I like him less than Frank Gore but more than Joseph Addai. That means he would be the #4 running back.

This process is the same for every running back on the cheatsheet. Once it's completed I now have my ranking of the top 60 rbs. I then do this for every position and player on the cheatsheet and have made my own ranking list. For now all the positions are ranked separately (top 30 QB, 60 RB, 60 WR, 30 TE) Then as things happen during training camp and preseason I tweak the list using the same process of "Do I like this guy better than this guy"?

The projections part is actually the easiest part of the process for me. I use a 3-year AVT list and match the point total to where I have the players ranked. And it's from the combination of the ranking list and the AVT list that I use to create a final VBD list that ranks the player by their value number and I draft off that list (for the most part).

I've probably made it sound really complicated but that's mainly because I suck with words. Trust me it's ridiculously simple and a lot less stressful than how I've done it in the past. I had the projections done in January and the ranking list done in May. Now it's just a matter of getting as much good information as possible between now and the draft and tweaking the list accordingly.

P.S. By AVT I mean Average Value Theory. Here a link that tells you all you need to know. Average Value Theory

I just started doing something similar. How do you account for each player's value relative to players at a different position (for example Larry Johnson to Peyton Manning). I just started in a 2 QB, 2 RB, 4WR + 1 flex (RB, WR, or TE) & am unsure of what baseline to use to compare the value across the positions.

I like to keep things simple. I go by the worst starter for my baseline. So for example in your league (12 team) the baseline would be the 24th ranked QB, 24th ranked RB, and 48th ranked WR. Then I subtract every team's point total from the worst starter's number and then get an X number. So for example I have the 24th ranked QB with 175 fantasy points I then subtract that number from every other QB's point total. I have the #1 ranked QB at 371 so 371-175=196 the #1 QB's X number would be 196. You do this for every player that you think is going to be drafted. Once you have all the X numbers you put all of them from every position in order from highest to lowest and boom you have a VBD cheatsheet.

You can't totally go off the cheatsheet though. You have to use common sense. I have the #1 tight end, and #1 quarterback in my overall top ten but there's no way i'd draft either one in the first round. But if they're still around in later rounds I have to seriously consider it you know.

 

[tomarken]

I like to keep things simple. I go by the worst starter for my baseline. So for example in your league (12 team) the baseline would be the 24th ranked QB, 24th ranked RB, and 48th ranked WR. Then I subtract every team's point total from the worst starter's number and then get an X number.

So for example I have the 24th ranked QB with 175 fantasy points I then subtract that number from every other QB's point total. I have the #1 ranked QB at 371 so 371-175=196 the #1 QB's X number would be 196. You do this for every player that you think is going to be drafted. Once you have all the X numbers you put all of them from every position in order from highest to lowest and boom you have a VBD cheatsheet.

You can't totally go off the cheatsheet though. You have to use common sense. I have the #1 tight end, and #1 quarterback in my overall top ten but there's no way i'd draft either one in the first round. But if they're still around in later rounds I have to seriously consider it you know.

Why not? What's the point of doing this if you ignore the results?

 

[The Man with the Plan]

The projections part is actually the easiest part of the process for me. I use a 3-year AVT list and match the point total to where I have the players ranked. And it's from the combination of the ranking list and the AVT list that I use to create a final VBD list that ranks the player by their value number and I draft off that list (for the most part).

I've probably made it sound really complicated but that's mainly because I suck with words. Trust me it's ridiculously simple and a lot less stressful than how I've done it in the past. I had the projections done in January and the ranking list done in May. Now it's just a matter of getting as much good information as possible between now and the draft and tweaking the list accordingly.

P.S. By AVT I mean Average Value Theory. Here a link that tells you all you need to know. Average Value Theory

fwiw, see post 12 in this thread for fbg thoughts on avt; yes, this thread!http://forums.footballguys.com/forum/index...howtopic=328615

No offense but when the creator of a fantasy football site (whose foundation is built on making projections for players) has a problem with a system that does his job quicker, simpler, and ultimately just as effective it's kind of hard to put too much stock in it. I can see why the "experts' would have a problem with AVT seeing as how it makes that part of their job to be frank, useless.Like I said before. I went down that road trying to figure out how many completions, attempts, catches, yards, and touchdowns a player is going to have or how many a player is "supposed" to have or any number of yards. I just got tired of trying to plan out every player's entire season knowing that I'm going to be wrong anyways. Once I did the AVT thing there was just no going back to the old ways.

AVT and doing your own projections (or going off somebody else's) is like 2 people trying to get from point A to point B except one person is walking and the other is driving a car. I'll let you guess which one is which.

 

[The Man with the Plan]

I like to keep things simple. I go by the worst starter for my baseline. So for example in your league (12 team) the baseline would be the 24th ranked QB, 24th ranked RB, and 48th ranked WR. Then I subtract every team's point total from the worst starter's number and then get an X number.

So for example I have the 24th ranked QB with 175 fantasy points I then subtract that number from every other QB's point total. I have the #1 ranked QB at 371 so 371-175=196 the #1 QB's X number would be 196. You do this for every player that you think is going to be drafted. Once you have all the X numbers you put all of them from every position in order from highest to lowest and boom you have a VBD cheatsheet.

You can't totally go off the cheatsheet though. You have to use common sense. I have the #1 tight end, and #1 quarterback in my overall top ten but there's no way i'd draft either one in the first round. But if they're still around in later rounds I have to seriously consider it you know.

Why not? What's the point of doing this if you ignore the results?

As I said before you have to use common sense. Why would I draft Antonio Gates in the 1st round when I can get him in the 3rd round? Plus you have to look at the team as a whole. Which is better? Having the #1 tight end and getting the #7 running back or getting the #1 tight end too soon and ending up with the #15 running back on your list? I don't know about you but for me it's a simple choice.The list is just a guideline so that you take the best player available at the position you need at the time. I'm not ignoring it at all. And I need my running backs early dammit.

 

[tomarken]

No offense but when the creator of a fantasy football site (whose foundation is built on making projections for players) has a problem with a system that does his job quicker, simpler, and ultimately just as effective it's kind of hard to put too much stock in it. I can see why the "experts' would have a problem with AVT seeing as how it makes that part of their job to be frank, useless.Like I said before. I went down that road trying to figure out how many completions, attempts, catches, yards, and touchdowns a player is going to have or how many a player is "supposed" to have or any number of yards. I just got tired of trying to plan out every player's entire season knowing that I'm going to be wrong anyways. Once I did the AVT thing there was just no going back to the old ways.AVT and doing your own projections (or going off somebody else's) is like 2 people trying to get from point A to point B except one person is walking and the other is driving a car. I'll let you guess which one is which.

I completely disagree with this. AVT is simplistic and completely arbitrary - all you have are a list of averages, and then you still have to plug in the players you think will achieve those averages. You might as well not use AVT at all and just list players you like in order from 1-100; it's the same exact thing, except by spending ten minutes using AVT numbers it tricks you into thinking there is some statistical basis for your list.

 

[RC94]

I base mine on my gut feelings, using a tiering system while taking into account my projections and other people's rankings. I mostly rely on my own projections/gut feelings, but will use other's rankings as sort of a tiebreaker or if I get lazy on the lower level players. I play only in dynasty leagues, so I tend to go very deep in the rankings.

 

[BigRed]

Tier players by position. Do very general/casual projections here and there in my head which help w/the tiering, but it's far from a be all/end all. Gut feelings are of course part of it too.

 

[The Man with the Plan]

No offense but when the creator of a fantasy football site (whose foundation is built on making projections for players) has a problem with a system that does his job quicker, simpler, and ultimately just as effective it's kind of hard to put too much stock in it. I can see why the "experts' would have a problem with AVT seeing as how it makes that part of their job to be frank, useless.

Like I said before. I went down that road trying to figure out how many completions, attempts, catches, yards, and touchdowns a player is going to have or how many a player is "supposed" to have or any number of yards. I just got tired of trying to plan out every player's entire season knowing that I'm going to be wrong anyways. Once I did the AVT thing there was just no going back to the old ways.

AVT and doing your own projections (or going off somebody else's) is like 2 people trying to get from point A to point B except one person is walking and the other is driving a car. I'll let you guess which one is which.

I completely disagree with this. AVT is simplistic and completely arbitrary - all you have are a list of averages, and then you still have to plug in the players you think will achieve those averages. You might as well not use AVT at all and just list players you like in order from 1-100; it's the same exact thing, except by spending ten minutes using AVT numbers it tricks you into thinking there is some statistical basis for your list.

Yes it's simplistic. That's the whole point dude. It's about working smarter not harder. As for arbitrary well aren't making projections the same thing? You're arbitrarily deciding who's going to have more yards and touchdowns than so and so this year. After doing the research you still have to make the final choice of who to rank. AVT works backwards from regular projections in that instead of predicting how many points a player is going to score and then ranking them it's the other way around with first the rankings and the stats. The point of the rankings is to make a VBD list which I feel helps me greatly in my draft.I resent the fact that you think that because my list is based off of AVT that it somehow lacks statistical basis. That's totally condescending of you (though on this site you wouldn't be the first FF snob I've come across). It's the fact that instead of focusing on exactly how many yards and receptions and carries and touchdowns a person is supposed to have (which in the end will always be wrong) I'm paying attention to the stuff that's going on in the NFL and getting as much information as I can between now and my draft.

People are going to believe what they want and no one's changing anybody's mind here so I'm done arguing about this. You have your way of doing things and I have mine. No one's wrong or right but I've done it both ways and I know after doing the AVT process I'm never going back to the old ways. It just doesn't make any sense to me.

 

[tomarken]

No offense but when the creator of a fantasy football site (whose foundation is built on making projections for players) has a problem with a system that does his job quicker, simpler, and ultimately just as effective it's kind of hard to put too much stock in it. I can see why the "experts' would have a problem with AVT seeing as how it makes that part of their job to be frank, useless.

Like I said before. I went down that road trying to figure out how many completions, attempts, catches, yards, and touchdowns a player is going to have or how many a player is "supposed" to have or any number of yards. I just got tired of trying to plan out every player's entire season knowing that I'm going to be wrong anyways. Once I did the AVT thing there was just no going back to the old ways.

AVT and doing your own projections (or going off somebody else's) is like 2 people trying to get from point A to point B except one person is walking and the other is driving a car. I'll let you guess which one is which.

I completely disagree with this. AVT is simplistic and completely arbitrary - all you have are a list of averages, and then you still have to plug in the players you think will achieve those averages. You might as well not use AVT at all and just list players you like in order from 1-100; it's the same exact thing, except by spending ten minutes using AVT numbers it tricks you into thinking there is some statistical basis for your list.

Yes it's simplistic. That's the whole point dude. It's about working smarter not harder. As for arbitrary well aren't making projections the same thing? You're arbitrarily deciding who's going to have more yards and touchdowns than so and so this year. After doing the research you still have to make the final choice of who to rank. AVT works backwards from regular projections in that instead of predicting how many points a player is going to score and then ranking them it's the other way around with first the rankings and the stats. The point of the rankings is to make a VBD list which I feel helps me greatly in my draft.I resent the fact that you think that because my list is based off of AVT that it somehow lacks statistical basis. That's totally condescending of you (though on this site you wouldn't be the first FF snob I've come across). It's the fact that instead of focusing on exactly how many yards and receptions and carries and touchdowns a person is supposed to have (which in the end will always be wrong) I'm paying attention to the stuff that's going on in the NFL and getting as much information as I can between now and my draft.

People are going to believe what they want and no one's changing anybody's mind here so I'm done arguing about this. You have your way of doing things and I have mine. No one's wrong or right but I've done it both ways and I know after doing the AVT process I'm never going back to the old ways. It just doesn't make any sense to me.

What's with people getting all defensive lately? I'm not being condescending, I'm pointing out a fundamental flaw in AVT. If you want to use it then feel free, we'll just agree to disagree. I just want to point out the important distinction between simple and simplistic. You like AVT because it is the former. I don't like AVT because it is the latter.

 

[dirtywaters20]

What I do is i'll take a look at any standard cheatsheet. And then I go down the list and see if there's a player that I like better than the player him. I'll make an adjustment. For example:

Let's say the first five running backs are listed like this:

LaDainian Tomlinson

Larry Johnson

Steven Jackson

Frank Gore

Joseph Addai

I see that LT is ranked above Larry Johnson. Then I ask myself do I like him than better Larry Johnson(the next ranked player on the cheatsheet) this season. Since I do then I rank LT the #1 running back. I type his name out onto a spreadsheet/database and cross his name off the cheatsheet.

Now I look at Larry Johnson and ask myself do I like him better than the next ranked player on the cheatsheet which is Steven Jackson. Since I do not I'll put Steven Jackson ahead of him for the time being. And because there is no other running back for me to put Steven Jackson behind I'll rank him the #2 running back. He's crossed off the list. You're wondering what happens with Larry Johnson.

I'll compare Larry Johnson to the remaining running backs on the cheatsheet. Until I find one that he's better than. For argument's sake (not necessarily my actual belief) let's say I like him less than Frank Gore but more than Joseph Addai. That means he would be the #4 running back.

This process is the same for every running back on the cheatsheet. Once it's completed I now have my ranking of the top 60 rbs. I then do this for every position and player on the cheatsheet and have made my own ranking list. For now all the positions are ranked separately (top 30 QB, 60 RB, 60 WR, 30 TE) Then as things happen during training camp and preseason I tweak the list using the same process of "Do I like this guy better than this guy"?

The projections part is actually the easiest part of the process for me. I use a 3-year AVT list and match the point total to where I have the players ranked. And it's from the combination of the ranking list and the AVT list that I use to create a final VBD list that ranks the player by their value number and I draft off that list (for the most part).

I've probably made it sound really complicated but that's mainly because I suck with words. Trust me it's ridiculously simple and a lot less stressful than how I've done it in the past. I had the projections done in January and the ranking list done in May. Now it's just a matter of getting as much good information as possible between now and the draft and tweaking the list accordingly.

P.S. By AVT I mean Average Value Theory. Here a link that tells you all you need to know. Average Value Theory

I just started doing something similar. How do you account for each player's value relative to players at a different position (for example Larry Johnson to Peyton Manning). I just started in a 2 QB, 2 RB, 4WR + 1 flex (RB, WR, or TE) & am unsure of what baseline to use to compare the value across the positions.

I like to keep things simple. I go by the worst starter for my baseline. So for example in your league (12 team) the baseline would be the 24th ranked QB, 24th ranked RB, and 48th ranked WR. Then I subtract every team's point total from the worst starter's number and then get an X number. So for example I have the 24th ranked QB with 175 fantasy points I then subtract that number from every other QB's point total. I have the #1 ranked QB at 371 so 371-175=196 the #1 QB's X number would be 196. You do this for every player that you think is going to be drafted. Once you have all the X numbers you put all of them from every position in order from highest to lowest and boom you have a VBD cheatsheet.

You can't totally go off the cheatsheet though. You have to use common sense. I have the #1 tight end, and #1 quarterback in my overall top ten but there's no way i'd draft either one in the first round. But if they're still around in later rounds I have to seriously consider it you know.

Thanks for taking the time. I hear you on the common sense, I have copies of the last 5 years drafts from the league and have a pretty good grasp of what the ADP will look like based on that. I plan make a cheetsheet based on the league's historical data & see how it compares with the DD list.