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!

Introducing the Market-Driven Baseline for VBD (1 Viewer)

freshly_shorn

Footballguy
Another freebie for all you sharks out there. FBG isn't soliciting for freelance articles any more, so I may as well offer it up to footballguy nation and see what you think.

I've heard many arguments and misunderstandings on how to 'correctly' determine baseline players for value-based drafting. Well, it struck me just yesterday that there is indeed an easy, sensible way to do it, a way that scales very well to any league size (teams and number of starters), and I offer it as a way to standardize baseline determination.

I call it market-driven baselines, and you can check it out at my ad-free blog here. Let me know what you think, guys! I liked how it valuated my players for this year's draft, it accurately reflects how my particular league ranks players.

:popcorn:

Follow me on Twitter @draftologist. Good luck!

 
Last edited by a moderator:
Thanks for the effort and your thoughts. I need to ponder it for a bit, but my very initial reactions (forgive me if I misunderstood you):

-You didn't seem to propose a solution for flex options

-Just because last year others in our league over drafted X - why should that alter OUR value of player?

-Why should I care about market value? Why should I care that market value suggests player X over player Y, if I think player Y helps me win?

-What value does this method offer over the standard(QB12, RB24, WR24, TE12, depending on league, of course)? The numbers provided should be relative - no? Whether Player X's VBD number is 120 or 50, what is important is how that number compares to the rest of the pool.

 
Thanks for the effort and your thoughts. I need to ponder it for a bit, but my very initial reactions (forgive me if I misunderstood you):-You didn't seem to propose a solution for flex options-Just because last year others in our league over drafted X - why should that alter OUR value of player?-Why should I care about market value? Why should I care that market value suggests player X over player Y, if I think player Y helps me win? -What value does this method offer over the standard(QB12, RB24, WR24, TE12, depending on league, of course)? The numbers provided should be relative - no? Whether Player X's VBD number is 120 or 50, what is important is how that number compares to the rest of the pool.
Flex options: Flex options are just the same as other positions. Your league tells you how much it values a given position. You don't start a 'Flex'. You start another RB, WR or TE. So, there really is nothing to address. It is irrelevant, IMO.- Just because & Market value... the whole point of determining a baseline is to determine value. It isn't how YOU value players- its understanding how the folks you draft against value players. Determining a baseline is simply the mechanism on how to compare players for VBD- it has nothing to do with how you personally feel about a player's value.-What value.. because this version tailors itself to whatever league size you have and accurately reflects how that league values players. But, agreed, it is a relative, arbitrary point of measurement. This method endeavors to accurately portray how a given league valuates a given position, no matter the size. Other methods can't claim that.
 
Last edited by a moderator:
Thanks for the effort and your thoughts. I need to ponder it for a bit, but my very initial reactions (forgive me if I misunderstood you):-You didn't seem to propose a solution for flex options-Just because last year others in our league over drafted X - why should that alter OUR value of player?-Why should I care about market value? Why should I care that market value suggests player X over player Y, if I think player Y helps me win? -What value does this method offer over the standard(QB12, RB24, WR24, TE12, depending on league, of course)? The numbers provided should be relative - no? Whether Player X's VBD number is 120 or 50, what is important is how that number compares to the rest of the pool.
Flex options: Flex options are just the same as other positions. Your league tells you how much it values a given position. You don't start a 'Flex'. You start another RB, WR or TE. So, there really is nothing to address. It is irrelevant, IMO.- Just because & Market value... the whole point of determining a baseline is to determine value. It isn't how YOU value players- its understanding how the folks you draft against value players. Determining a baseline is simply the mechanism on how to compare players for VBD- it has nothing to do with how you personally feel about a player's value.-What value.. because this version tailors itself to whatever league size you have and accurately reflects how that league values players. But, agreed, it is a relative, arbitrary point of measurement. This method endeavors to accurately portray how a given league valuates a given position, no matter the size. Other methods can't claim that.
-It is absolutely relevant. How does your method tell me if RB27 is worth more than TE8 as a flex?-The point of VBD is not to deterime market value, it is to determine actual value. If my projections suggest that Victor Cruz will help me win more than Marshawn Lynch, why does it matter what market value they both carry? -Let me ask this another way: My league overdrafts QBs - why should I adjust my VBD calculations to adjust for that? And what is to say my league won't smarten up and value QBs differently this year? Does this have any value in a new league, or when league member's are shuffled?
 
-It is absolutely relevant. How does your method tell me if RB27 is worth more than TE8 as a flex?

Let's step back.. this is the BASELINE, nothing else. Baselines are only relevant within the same position. So, once you determine the baselines and player values, you'll be able to make your comparisons as usual. Sorry if I wasn't clear. VBD is comprised of four steps: Projections, Baselines, X Number, Comparisons. I'm only suggesting a standard method for determining that baseline. That part has nothing to do with how you personally feel about players- it is only on deciding which player to use as a point of comparison against like players.

-The point of VBD is not to deterime market value, it is to determine actual value. If my projections suggest that Victor Cruz will help me win more than Marshawn Lynch, why does it matter what market value they both carry?: Again, we are only talking about determining baselines. Baselines only apply within the same position. The market can tell you where to draw that line. It is useful to know where the 'market' draws that line, IMO.

-Let me ask this another way: My league overdrafts QBs - why should I adjust my VBD calculations to adjust for that? And what is to say my league won't smarten up and value QBs differently this year? Does this have any value in a new league, or when league member's are shuffled?- Again, we're talking baselines here. Not your overall VBD strategy. Yes, you could apply this model to an ADP list to get an idea on how the fantasy world as a whole ranks players.

 
Last edited by a moderator:
-It is absolutely relevant. How does your method tell me if RB27 is worth more than TE8 as a flex?

Let's step back.. this is the BASELINE, nothing else. Baselines are only relevant within the same position. So, once you determine the baselines and player values, you'll be able to make your comparisons as usual. Sorry if I wasn't clear. VBD is comprised of four steps: Projections, Baselines, X Number, Comparisons. I'm only suggesting a standard method for determining that baseline. That part has nothing to do with how you personally feel about players- it is only on deciding which player to use as a point of comparison against like players.

-The point of VBD is not to deterime market value, it is to determine actual value. If my projections suggest that Victor Cruz will help me win more than Marshawn Lynch, why does it matter what market value they both carry?: Again, we are only talking about determining baselines. Baselines only apply within the same position. The market can tell you where to draw that line. It is useful to know where the 'market' draws that line, IMO.

-Let me ask this another way: My league overdrafts QBs - why should I adjust my VBD calculations to adjust for that? And what is to say my league won't smarten up and value QBs differently this year? Does this have any value in a new league, or when league member's are shuffled?- Again, we're talking baselines here. Not your overall VBD strategy. Yes, you could apply this model to an ADP list to get an idea on how the fantasy world as a whole ranks players.
-I am very familiar with VBD as a process and a philosophy. Baselines are not only relevant within the same positions when there are flex options. Why are QBs MUCH more valuable in super-flex leagues? Their unadjusted VBD is the same. The answer is because you need to start comparing QBs to the other positions. -Again - why should the market draw that line? I know it can - but why? Why should I adjust my VBD calculation because my league overdrafts backup QBs?

-Baseline's are useless without the context of an overall VBD strategy. An ADP list does what you propose your baseline does. So why do I need it?

 
Last edited by a moderator:
I'm having trouble seeing what's new or innovative about this...

Most people have been establishing the baseline as a replacement-level player (i.e. the hypothetical best available palyer from waivers at that position) for years. Hence saying "Player X is worth __ points per weeek above replacement"

How is this any different?

 
Last edited by a moderator:
I'm having trouble seeing what's new or innovative about this...Most people have been establishing the baseline as a replacement-level player (i.e. the hypothetical best available palyer from waivers at that position) for years. Hence saying "Player X is worth __ points per weeek above replacement"How is this any different?
I was thinking that as I wrote it. It isn't anything particularly new, but it does simplify things immensely. Hey, just offering it out there for folks puzzled by this concept. Whatever works!
 
I'm having trouble seeing what's new or innovative about this...Most people have been establishing the baseline as a replacement-level player (i.e. the hypothetical best available palyer from waivers at that position) for years. Hence saying "Player X is worth __ points per weeek above replacement"How is this any different?
I was thinking that as I wrote it. It isn't anything particularly new, but it does simplify things immensely. Hey, just offering it out there for folks puzzled by this concept. Whatever works!
Allrighty, well thanks for the effort. Good for people who are just starting to understand VBD. Cheers! :thumbup:
 
-I am very familiar with VBD as a process and a philosophy. Baselines are not only relevant within the same positions when there are flex options. Why are QBs MUCH more valuable in super-flex leagues? Their unadjusted VBD is the same. The answer is because you need to start comparing QBs to the other positions. You may be right, but the baselines I'm discussing are within a given position. Again, you can't baseline a flex option. That isn't what this is about.

-Again - why should the market draw that line? I know it can - but why? Why should I adjust my VBD calculation because my league overdrafts backup QBs? A key point to VBD is understanding what the other guy thinks. MDB accomplishes that. You're not adjusting your VBD calculation. You are just setting the point of comparison within a given position.

-Baseline's are useless without the context of an overall VBD strategy. An ADP list does what you propose your baseline does. So why do I need it? ADP and setting baselines have nothing to do with each other. ADP tells you where the market drafts a given player. MDB suggests one way to set baselines for sake of like-player comparison and is tied to how your league thinks rather than just drawing an arbitrary line across your projection list. How you come up with the baseline is clearly a matter of choice; it is all relative, after all. This is just one quick and dirty method that scales well and actually tries to capture how the guys in your league will draft, so you can exploit that info accordingly.

Whatever works for you is the best method for you.

 
This is just one quick and dirty method that scales well and actually tries to capture how the guys in your league will draft, so you can exploit that info accordingly.Whatever works for you is the best method for you.
Good point and something to ponder. Again, thanks for sharing.
 
-It is absolutely relevant. How does your method tell me if RB27 is worth more than TE8 as a flex?

Let's step back.. this is the BASELINE, nothing else. Baselines are only relevant within the same position. So, once you determine the baselines and player values, you'll be able to make your comparisons as usual. Sorry if I wasn't clear. VBD is comprised of four steps: Projections, Baselines, X Number, Comparisons. I'm only suggesting a standard method for determining that baseline. That part has nothing to do with how you personally feel about players- it is only on deciding which player to use as a point of comparison against like players.

-The point of VBD is not to deterime market value, it is to determine actual value. If my projections suggest that Victor Cruz will help me win more than Marshawn Lynch, why does it matter what market value they both carry?: Again, we are only talking about determining baselines. Baselines only apply within the same position. The market can tell you where to draw that line. It is useful to know where the 'market' draws that line, IMO.

-Let me ask this another way: My league overdrafts QBs - why should I adjust my VBD calculations to adjust for that? And what is to say my league won't smarten up and value QBs differently this year? Does this have any value in a new league, or when league member's are shuffled?- Again, we're talking baselines here. Not your overall VBD strategy. Yes, you could apply this model to an ADP list to get an idea on how the fantasy world as a whole ranks players.
-I am very familiar with VBD as a process and a philosophy. Baselines are not only relevant within the same positions when there are flex options. Why are QBs MUCH more valuable in super-flex leagues? Their unadjusted VBD is the same. The answer is because you need to start comparing QBs to the other positions. -Again - why should the market draw that line? I know it can - but why? Why should I adjust my VBD calculation because my league overdrafts backup QBs?

-Baseline's are useless without the context of an overall VBD strategy. An ADP list does what you propose your baseline does. So why do I need it?
What his baseline does is make the ADP fit more for your league tendencies.I am trying out his spreadsheet this year in a league we have had going for 20+ years. I take the info from prior drafts to try and see who will be available at certain points, ie at 2.10, historically we have 4QBs, 12RBs and 5WRs gone. I combine that with ADP to see who is likely to be available and whether I want to trade up or down. I have to have trades pretty much done before the draft as I am the only one drafting remotely. One thing I am looking at is trying to let Graham drop and pick him up at 3.03 (No TE has ever gone higher than 3.12 in our league and the two sharks are behind me at 4 and 5)We will see how it works. IMO it is a tool that can ADP less general for established leagues.

 
What his baseline does is make the ADP fit more for your league tendencies.I am trying out his spreadsheet this year in a league we have had going for 20+ years. I take the info from prior drafts to try and see who will be available at certain points, ie at 2.10, historically we have 4QBs, 12RBs and 5WRs gone. I combine that with ADP to see who is likely to be available and whether I want to trade up or down. I have to have trades pretty much done before the draft as I am the only one drafting remotely. One thing I am looking at is trying to let Graham drop and pick him up at 3.03 (No TE has ever gone higher than 3.12 in our league and the two sharks are behind me at 4 and 5)We will see how it works. IMO it is a tool that can ADP less general for established leagues.
MDB links the baseline to how your league values positions, not really trying to project ADP. But- yes, the Draft Planalyzer (which does help you compute a customized ADP for your league, in a way) can easily tally up the number of players at the MDB in the Predicted Position Depth table, and those become your baseline players. I used Draft Dominator when modeling this concept, and I have to say that the MDB method does a much better job reflecting how the league as a whole values positions. Knowing how the league values positions in turn helps me find value when selecting actual players. Joe's Secret Formula is decent, too, but the worst starter method looks nothing like how my league values positions.
 
This baseline has merit. What you are saying is that the last player drafted for each position has the same value.

That helps you build the value of your roster overall, but you don't play your entire roster.

You play starters across all the position.

If you use standard VBD, your roster will be set up better for starters, but will lack depth.

My answer to this is to average the two values. I posted my method under "Dual Baseline Draft Method" that I just bumped.

 
First I want to point out that this seems to just be a new name for a very old concept. Whether or not it's better or worse than other baseline methods is certainly debatable, but it's not new in any way.

Also, in reading your article it seems like you may be conflating the concepts of value and price. While they are frequently related, it's important to remember how they differ.

Who determines the value of a player? We do. We are the market.
No we don't. We determine the price of a player. That price may turn out to be way too high, or just about right, or the steal of the draft. The player's value is determined by how many more points he provides your team than other players you could have taken in his place. This is something we try to calculate based on projected fantasy stats, not "market" conditions.
We determine the relative value of players by the order in which we draft them.
Again, no we don't. We determine the price of players by the order in which we draft them. Their actual value can be quite different.Since you're using market analogies, think of players as stocks. Every stock has a price, which is determined by what the market is willing to pay. The underlying company or asset also has some intrinsic value, which is determined by their business model, profitability, etc. This value can sometimes differ greatly from the market price of the stock, which is why there are undervalued and overvalued stocks. Just because someone's willing to pay $40 for a share of Facebook doesn't mean a share of Facebook is worth $40.

You may dispute that last part by pointing out that if someone is willing to buy Facebook stock for $40, then it IS worth $40. You're right in that sense, and this is where the market analogy starts to break down when you try to translate it into fantasy football terms. Unlike stocks, you don't draft players with the intention of selling them later for a profit. You draft them because you think they will produce more points for your team. That's the value of fantasy players. If everyone else in your leage (i.e. the "market") decided that Sebastian Janikowski was worth a 1st round pick, and all tried to trade up to the #1 spot to get him there, that doesn't mean he's actually worth the #1 overall pick. In other words, even though his price (determined by the market) was 1.01, his value (not determined by the market) wasn't anywhere near that.

Let's take this to an absurd level just to really expose the flaw. Imagine you found yourself in a league with standard roster and scoring settings, but for some reason everyone drafted nothing but kickers for the first 12 rounds. Starting kickers, backup kickers, retired kickers, high school kickers... they just can't get enough kickers. Your "market-driven" baseline would determine that kickers are ubervaluable (since the baseline kicker would be like the 144th kicker down the list) and that players at all other positions are essentially worthless (since the baseline players would be the #1 guy at each position). This is the exact opposite of a winning strategy, of course - in reality, the optimal strategy would be to scoop up studs at every other position and destroy your competition every other week.

That highlights a problem with the "market-driven" value model - the way to really succeed is not to let the market determine player values for you, it's to determine player values independently of their market price, and then select the ones that are the most underpriced by the market. Really the best way to use "market" tendencies to your advantage is to use dynamic baselines during the draft.

 
First I want to point out that this seems to just be a new name for a very old concept. Whether or not it's better or worse than other baseline methods is certainly debatable, but it's not new in any way.

Also, in reading your article it seems like you may be conflating the concepts of value and price. While they are frequently related, it's important to remember how they differ.

Who determines the value of a player? We do. We are the market.
No we don't. We determine the price of a player. That price may turn out to be way too high, or just about right, or the steal of the draft. The player's value is determined by how many more points he provides your team than other players you could have taken in his place. This is something we try to calculate based on projected fantasy stats, not "market" conditions.
We determine the relative value of players by the order in which we draft them.
Again, no we don't. We determine the price of players by the order in which we draft them. Their actual value can be quite different.Since you're using market analogies, think of players as stocks. Every stock has a price, which is determined by what the market is willing to pay. The underlying company or asset also has some intrinsic value, which is determined by their business model, profitability, etc. This value can sometimes differ greatly from the market price of the stock, which is why there are undervalued and overvalued stocks. Just because someone's willing to pay $40 for a share of Facebook doesn't mean a share of Facebook is worth $40.

You may dispute that last part by pointing out that if someone is willing to buy Facebook stock for $40, then it IS worth $40. You're right in that sense, and this is where the market analogy starts to break down when you try to translate it into fantasy football terms. Unlike stocks, you don't draft players with the intention of selling them later for a profit. You draft them because you think they will produce more points for your team. That's the value of fantasy players. If everyone else in your leage (i.e. the "market") decided that Sebastian Janikowski was worth a 1st round pick, and all tried to trade up to the #1 spot to get him there, that doesn't mean he's actually worth the #1 overall pick. In other words, even though his price (determined by the market) was 1.01, his value (not determined by the market) wasn't anywhere near that.

Let's take this to an absurd level just to really expose the flaw. Imagine you found yourself in a league with standard roster and scoring settings, but for some reason everyone drafted nothing but kickers for the first 12 rounds. Starting kickers, backup kickers, retired kickers, high school kickers... they just can't get enough kickers. Your "market-driven" baseline would determine that kickers are ubervaluable (since the baseline kicker would be like the 144th kicker down the list) and that players at all other positions are essentially worthless (since the baseline players would be the #1 guy at each position). This is the exact opposite of a winning strategy, of course - in reality, the optimal strategy would be to scoop up studs at every other position and destroy your competition every other week.

That highlights a problem with the "market-driven" value model - the way to really succeed is not to let the market determine player values for you, it's to determine player values independently of their market price, and then select the ones that are the most underpriced by the market. Really the best way to use "market" tendencies to your advantage is to use dynamic baselines during the draft.
Great post, but I naturally have to disagree with some things (and agree with some, too). That's why I posted it here. First, I agree most of the concepts aren't new, but I've never seen anyone baseline this way- it's new to me in that sense. When I applied the model to this years draft in DD, I was struck at how well it reflects the reality of how my league values some positions over others. Combined with the custom ADP I've generated based on historical drafts, this model delivers a very accurate way to mock draft and zero in on where the value will be in my league. It certainly looks a lot more sensible than the 'worst starter' method, which ranks a DL as the 49th most valuable player- ridiculous if you ask me.Anyway, I'm certainly aware of the difference between value and price. But we absolutely do set the value of the players. There is no way to know a player's true value until the season is underway and they actually start performing. We are setting the value of anticipated future performance (projections). Who sets those projections? We do- the market. So, yes, the market *does* determine the value of a player. Now, as the season wears on, our valuation no longer is relevant, true. But we absolutely do set the value of the players through projections. I guess the closest proxy in the stock market is commodity futures, right? We can guess what they're worth, and buy/sell accordingly, but we're only estimating their value in the future. We're speculators.

I get what you are saying, believe me. I'd argue the 'flaw' you so aptly demonstrate in MDB is one in many baseline models. But in the real world, the model works perfectly well and I believe it better captures how the league values positions. That is quite different from how I value positions and players. But I find it very useful to puzzle out how my league drafts overall, where it places value, using things like custom ADP and this MDB.

I enjoyed your post immensely, certainly thoughtful and a valid argument! Ironically, my post has done nothing to alleviate the debate or confusion. Oh well! Thanks!

 
Last edited by a moderator:
Anyway, I'm certainly aware of the difference between value and price. But we absolutely do set the value of the players. There is no way to know a player's true value until the season is underway and they actually start performing. We are setting the value of anticipated future performance (projections). Who sets those projections? We do- the market.
Again, no we don't. The market sets the price of the players. The market doesn't create the projected values. Each individual sets their own projected value for each player (or gets them from somewhere like Footballguys), and by the end of the season we can determine (in hindsight) each player's actual value. Price, projected value, and actual value are three different things, and the only one that is set by the market is the price.
I'd argue the 'flaw' you so aptly demonstrate in MDB is one in many baseline models.
No, the flaw was specifically shown to be the result of letting the "market" determine player values. Any other baseline that isn't determined by the market won't be similarly influenced. If you used MDB in the all-kicker draft, you'd end up with a bunch of kickers on your team, because your market would be telling you that kickers are the most valuable players by far. If you used pretty much any other baseline, you'd end up with a stud-filled team almost guaranteed to go 16-0.
But in the real world, the model works perfectly well and I believe it better captures how the league values positions. That is quite different from how I value positions and players. But I find it very useful to puzzle out how my league drafts overall, where it places value, using things like custom ADP and this MDB.
Right, even here you admit that how you value players is quite different from how the rest of your league values those same players. That's how it's supposed to work - you come up with your own projections and rankings, which are hopefully better than the ones your opponents come up with, and you draft based on the values you've come up with. Again, it seems like you're combining two things that have different meanings or are meant to be separate. I agree that it's very useful to know your league's draft tendencies - this allows you to do more accurate mock drafts ahead of time, for example, and if you have the #2 pick it's useful to have an educated guess as to what the guy behind you is going to do at the 12/1 turn. That's one thing.

You also need to determine player values. That's another thing, and it's not dependent on your league's draft tendencies (as my exaggerated all-kicker draft example illustrates). Based on your projections, for instance, Aaron Rodgers has some value above some other "baseline" QBs you could draft. That value, whatever it is and however you determine it, is independent of your league's draft tendencies (i.e. the market). Aaron Rodgers is X points better than Jake Locker, whether they are drafted ten rounds apart or with back to back picks. His price WILL be determined the market, and it's your job to decide during the draft whether his price is higher or lower than his value.

I do think I see what you're trying to accomplish - for example, if your league devalues QBs, then presumably you want to also devalue them so you don't draft one too early. You don't want to waste a first round pick on Aaron Rodgers if he'd still be sitting there in the third round. In that sense it's important to know your leaguemates' draft tendencies. But your MDB doesn't capture that kind of info. You know that 21 QBs will be off the board by the end of the 12th round, but setting your baseline at the 21st best QB doesn't tell you how early the studs will be drafted. Every owner might grab a starter by the end of the 4th round, or Rodgers might sit there until the 5th round, but as long as 21 QBs end up being taken by the 12th round, MDB doesn't differentiate between these two scenarios. Only a more dynamic approach during the draft can inform those kinds of decisions.

I'm sorry if I've been overly critical - there's no agreed-upon "right" way to select a baseline, and ignoring absurd cases like the all-kicker draft, I don't think your method is especially worse than any of the others people commonly use. But I don't think it's especially better, either.

 
Actually, it does capture perfectly how my league acts. We typically take 3-4the abs in first two rounds, and when I set baselines that's what DD shows. Combined with custom asp, mdb accurately depicts how the league values a position.

We'll have to agree to disagree on who determines value.

 
Actually, it does capture perfectly how my league acts. We typically take 3-4the abs in first two rounds, and when I set baselines that's what DD shows. Combined with custom asp, mdb accurately depicts how the league values a position.

We'll have to agree to disagree on who determines value.
It isn't really an agree to disagree thing. Elenchi does a great job of explaining the difference between value and price - everyone paying more for something doesn't necessarily mean it is more valuable.Please explain the bolded for me, because I am assuming there is a typo there but I can't quite figure out what the typo is. Maybe I'm just missing something.

 
re: price vs value - it's semantics. It really doesn't matter to the crux of what freshly_shorn is trying to demonstrate.

 
Instinctive- QBs. Darn autocorrect. Moleculo- exactly. Getting stuck on terminology.
Ok, I kept thinking RBs but it didn't quite make sense - no worries.To try and put what Elenchi is saying into more layman's terms: What you value a guy at and what his value is are two different things. As such, what your league values a guy at and what his value is are also different. You shouldn't let your league determine his value - they could be wrong.

You should definitely take into account league tendencies, which is clearly what this is trying to do, but I am not sure it does so as effectively as it could/as is necessary.

I do think I see what you're trying to accomplish - for example, if your league devalues QBs, then presumably you want to also devalue them so you don't draft one too early. You don't want to waste a first round pick on Aaron Rodgers if he'd still be sitting there in the third round. In that sense it's important to know your leaguemates' draft tendencies. But your MDB doesn't capture that kind of info. You know that 21 QBs will be off the board by the end of the 12th round, but setting your baseline at the 21st best QB doesn't tell you how early the studs will be drafted. Every owner might grab a starter by the end of the 4th round, or Rodgers might sit there until the 5th round, but as long as 21 QBs end up being taken by the 12th round, MDB doesn't differentiate between these two scenarios. Only a more dynamic approach during the draft can inform those kinds of decisions.
The bolded is the part that the other poster says which I am trying to get at here. How does your system get you to knowing that 3-4 QBs are taken in the first round? I don't think it does - I think you're combining your system with other knowledge you have (a good thing, what you should do) but then attributing all of it to the system (incorrectly).Please correct me if I am wrong in those assertions and explain a little more, because I'm always looking to improve the VBD model and I'd like to determine if this actually does.

 
Actually, it does capture perfectly how my league acts.
I never said it didn't, but that's not by design. It's more a coincidence generated by the fact that most of your leaguemates are probably using projections that are roughly similar to the ones you use, so for example their model is telling them that Aaron Rodgers should go in the mid-first round, and so is yours. It's not much of a thought experiment, but imagine if (for whatever reason) all your leaguemates decide this year to wait on QBs. Maybe they're using projections from a different site that has QBs scoring fewer points this year, maybe the guy who won the league last year had a bunch of stud WRs on their team so everyone wants to go WR early this year, whatever. In reality, QBs would be drafted later than usual, but your model would still spit out the same exact results that it is now, so it wouldn't accurately capture the way your league would act.

Again, your model just says "21 QBs will be gone by the end of the 12th round," it doesn't say "2 QBs will be taken in the first round, 3 QBs will be taken in the second round, etc." To the extent that the results of your model "capture perfectly" what your league does isn't a feature MDB, it's just a consequence of the fact that most people use similar projections and most drafts follow a fairly predictable pattern.

Anyway, my only point is that "how your league acts," while valuable ancillary information for optimizing your draft strategy, is not necessarily how you should determine how valuable a player will be on your fantasy team. Aaron Rodgers is a great QB. He's probably going to put up some gaudy stats again this year - more than most other QBs in the league. That gives him some intrinsic value as a fantasy QB. Whether or not your leaguemates think he's worth an early first rounder or a late third rounder doesn't impact how much better he'll be than the other QBs in the NFL this year. If you think no one in your league is going to take him until the third round, that's useful information, and in the sense that you're using the word, that means his "value" is lower because it would be a mistake to take him in the first round if he'll still be there in the second. But MDB won't tell you whether or not he'll still be there in the second round. It'll tell you (based on your projections) where he should be taken, but the fact that 21 QBs will be off the board by round 12 doesn't give you any indication where Rodgers will go among those 21.

Another little thought experiment. Say you think your league will draft 21 QBs in the first 12 rounds. So your baseline is the 21st best QB, and let's say you have that guy projected to score 280 points this year. Let's say you also have Aaron Rodgers projected to score 480 points this year. So Aaron Rodgers value in the MDB system is 200, and that might place him squarely in the middle of the first round. Now imagine that instead, you had determined that your leaguemates wait forever for backup QBs and therefore there will only be 14 QBs drafted by the end of the 12th round. So your baseline is the 14th best QB. Assume he's projected to score 320 points this year. So now Aaron Rodgers value in the MDB system is reduced to 160. Does it make sense that Aaron Rodger's draft stock would fall simply because your leaguemates decided to wait longer on drafting backup QBs? I'm not necessarily saying it doesn't make sense, but that's a consequence of MDB so consider whether that operates the way you think it should. Obviously that's a consequence of changing the baseline in any VBD system that uses a static baseline, so I'm not saying it's a flaw of MDB in particular, or even a flaw at all. I'm just curious if you think Aaron Rodgers would lose value in a scenario like that.

 
re: price vs value - it's semantics. It really doesn't matter to the crux of what freshly_shorn is trying to demonstrate.
Right, the only reason I'm trying to distinguish between the terms is to clarify exactly what OP is trying to do, because I think he's mixing them in ways they shouldn't be mixed.There is some... thing - call it price, or value, or whatever - but it's "what I have to pay to get this player."There is some other thing - call it value or X-value or whatever - but it's "how much will this player help me win fantasy football games."Those two things, whatever you decide to call them, are often closely related but they're not the same. Ultimately you want to maximize the second thing. That may mean you need to also maximize the first thing, or you may be able to outwit your opponents and maximize the second thing while keeping the first thing minimal. But the second thing is what matters. We talk about drafting for value, but strictly speaking, you win by scoring the most points, not by getting the best deal during the draft. For me, the first thing is "price" and it is determined by the market. The second thing is "value" and it is not determined by the market.
 
Actually, it does capture perfectly how my league acts. We typically take 3-4the abs in first two rounds, and when I set baselines that's what DD shows. Combined with custom asp, mdb accurately depicts how the league values a position.We'll have to agree to disagree on who determines value.
Value is determined by the number of points players will score in your starting lineup. The purpose of the Draft Dominator is to determine where your league-mates are inaccurately assessing value, thus giving you opportunity to get higher-value players than they are. If you set up the DD to map to your league's tendencies, it might help you predict what someone else is going to do, but it won't tell you what you want to know, which is whether that was the right thing to do or not.
 
'Ignoratio Elenchi said:
'freshly_shorn said:
Actually, it does capture perfectly how my league acts.
I never said it didn't, but that's not by design. It's more a coincidence generated by the fact that most of your leaguemates are probably using projections that are roughly similar to the ones you use, so for example their model is telling them that Aaron Rodgers should go in the mid-first round, and so is yours. It's not much of a thought experiment, but imagine if (for whatever reason) all your leaguemates decide this year to wait on QBs. Maybe they're using projections from a different site that has QBs scoring fewer points this year, maybe the guy who won the league last year had a bunch of stud WRs on their team so everyone wants to go WR early this year, whatever. In reality, QBs would be drafted later than usual, but your model would still spit out the same exact results that it is now, so it wouldn't accurately capture the way your league would act.

Again, your model just says "21 QBs will be gone by the end of the 12th round," it doesn't say "2 QBs will be taken in the first round, 3 QBs will be taken in the second round, etc." To the extent that the results of your model "capture perfectly" what your league does isn't a feature MDB, it's just a consequence of the fact that most people use similar projections and most drafts follow a fairly predictable pattern.

Anyway, my only point is that "how your league acts," while valuable ancillary information for optimizing your draft strategy, is not necessarily how you should determine how valuable a player will be on your fantasy team. Aaron Rodgers is a great QB. He's probably going to put up some gaudy stats again this year - more than most other QBs in the league. That gives him some intrinsic value as a fantasy QB. Whether or not your leaguemates think he's worth an early first rounder or a late third rounder doesn't impact how much better he'll be than the other QBs in the NFL this year. If you think no one in your league is going to take him until the third round, that's useful information, and in the sense that you're using the word, that means his "value" is lower because it would be a mistake to take him in the first round if he'll still be there in the second. But MDB won't tell you whether or not he'll still be there in the second round. It'll tell you (based on your projections) where he should be taken, but the fact that 21 QBs will be off the board by round 12 doesn't give you any indication where Rodgers will go among those 21.

Another little thought experiment. Say you think your league will draft 21 QBs in the first 12 rounds. So your baseline is the 21st best QB, and let's say you have that guy projected to score 280 points this year. Let's say you also have Aaron Rodgers projected to score 480 points this year. So Aaron Rodgers value in the MDB system is 200, and that might place him squarely in the middle of the first round. Now imagine that instead, you had determined that your leaguemates wait forever for backup QBs and therefore there will only be 14 QBs drafted by the end of the 12th round. So your baseline is the 14th best QB. Assume he's projected to score 320 points this year. So now Aaron Rodgers value in the MDB system is reduced to 160. Does it make sense that Aaron Rodger's draft stock would fall simply because your leaguemates decided to wait longer on drafting backup QBs? I'm not necessarily saying it doesn't make sense, but that's a consequence of MDB so consider whether that operates the way you think it should. Obviously that's a consequence of changing the baseline in any VBD system that uses a static baseline, so I'm not saying it's a flaw of MDB in particular, or even a flaw at all. I'm just curious if you think Aaron Rodgers would lose value in a scenario like that.
I agree with this, but this is really a characteristic of static vs dynamic VBD. I think that this MDB is probably an improvement over static VBD but not as good as dynamic VBD.The problem with dynamic VBD is it's difficult to do w/o a computer - if you want to draft with a piece of paper and that's it, you are kind of stuck with static, which is where this may help.

 
Wow, what a thread.

OK, trying to clarify my position, as requested.

MDB is simply a way to determine which players should be your baseline. It is dynamic in the sense that it sizes itself to the league and number of starters in question. MDB tells you where to measure a player's X-Number from, just like any other baseline model.

Value, as Elenchi is stressing, only comes into play when you begin comparing players across positions, and while I use the term value throughout the article, what I mean, specifically, is the emphasis/premium/value that the league, as a whole, places on a particular position, as measured by the quantity taken at the MDB.

MDB is only meant to determine where to begin measuring, and I think this method does a much better job, in a simple way, then most models since it does incoporate your league's tendencies. I never said MDB tells you where to take players- that's up to you. Custom ADP can tell you where your league will take players. MDB only tells you what player should be your zero player, that's all, just like any baseline model.

I can see where my statement 'the market determines the value' should be amended, and thanks to Elenchi & CalBear for helping me realize that. I think the market can tell you where to begin measuring value, as perceived by the league as a whole. Your projections will then reveal- within the context of how much emphasis/premium/value the league places on a particular position.

Great thread, guys, this is what I hoped for.

 
'CalBear said:
'freshly_shorn said:
Actually, it does capture perfectly how my league acts. We typically take 3-4the abs in first two rounds, and when I set baselines that's what DD shows. Combined with custom asp, mdb accurately depicts how the league values a position.

We'll have to agree to disagree on who determines value.
Value is determined by the number of points players will score in your starting lineup. The purpose of the Draft Dominator is to determine where your league-mates are inaccurately assessing value, thus giving you opportunity to get higher-value players than they are. If you set up the DD to map to your league's tendencies, it might help you predict what someone else is going to do, but it won't tell you what you want to know, which is whether that was the right thing to do or not.
Bingo. This article does a great job of explaining the flaws in Joe's Top 100 method, which I think apply to a baseline set by your league's tendencies. Using economic theory, VBD baselines should be based on value over replacement player. The trick in fantasy football is to use a value measure that accounts for greater value for those players that produce as starters for your fantasy team as opposed to those who are on your bench but are above replacement player value. I'm working on a model for this and hope to release it after a season of testing. A lot of this is based on the Retrospective Draft that was held in the Shark Pool a few years back. I'm sure I'll post about it in the Shark Pool, but you can follow @QSPAR on Twitter if you want to be aware of updates. It's not going to happen this season.
 
Last edited by a moderator:
'Ignoratio Elenchi said:
'freshly_shorn said:
Actually, it does capture perfectly how my league acts.
I never said it didn't, but that's not by design. It's more a coincidence generated by the fact that most of your leaguemates are probably using projections that are roughly similar to the ones you use, so for example their model is telling them that Aaron Rodgers should go in the mid-first round, and so is yours. <snip> Another little thought experiment. Say you think your league will draft 21 QBs in the first 12 rounds. So your baseline is the 21st best QB, and let's say you have that guy projected to score 280 points this year. Let's say you also have Aaron Rodgers projected to score 480 points this year. So Aaron Rodgers value in the MDB system is 200, and that might place him squarely in the middle of the first round. Now imagine that instead, you had determined that your leaguemates wait forever for backup QBs and therefore there will only be 14 QBs drafted by the end of the 12th round. So your baseline is the 14th best QB. Assume he's projected to score 320 points this year. So now Aaron Rodgers value in the MDB system is reduced to 160. Does it make sense that Aaron Rodger's draft stock would fall simply because your leaguemates decided to wait longer on drafting backup QBs? I'm not necessarily saying it doesn't make sense, but that's a consequence of MDB so consider whether that operates the way you think it should. Obviously that's a consequence of changing the baseline in any VBD system that uses a static baseline, so I'm not saying it's a flaw of MDB in particular, or even a flaw at all. I'm just curious if you think Aaron Rodgers would lose value in a scenario like that.
Point 1-Quite to the contrary, it can't be anything but by design. If I'm counting how many of each position my league takes by a given point in the draft, how can it look any other way? If past results say my league takes around 21 QBs by pick 144 every year, and I want to tailor my baseline to that fact, that is quite by design. Now, if as you say everyone shifts their philosophy and we're bitten by the 'past results don't guarantee future performance', then of course all this flies out the window- just like with any other system. Just because the X-number says the player should go in the first round has nothing to do whether he will or won't.. that's what ADP tells you.Point 2-MDB does not change the value of Rodgers as he compares to other QBs- the relative difference in X-numbers doesn't change. If the league decides to start snapping up RBs and leaving QBs to later, it doesn't change his value at all. The point is- MDB predicts how many of any position will be gone by a certain point. I'm not determining anything other than how a value a player within that context. And as you say- you can ask that question of any baselining model. I think this model does a better job of placing your personal valuations in the league (market) context then most, and in a simple fashion.
 
Last edited by a moderator:
'CalBear said:
'freshly_shorn said:
Actually, it does capture perfectly how my league acts. We typically take 3-4the abs in first two rounds, and when I set baselines that's what DD shows. Combined with custom asp, mdb accurately depicts how the league values a position.

We'll have to agree to disagree on who determines value.
Value is determined by the number of points players will score in your starting lineup. The purpose of the Draft Dominator is to determine where your league-mates are inaccurately assessing value, thus giving you opportunity to get higher-value players than they are. If you set up the DD to map to your league's tendencies, it might help you predict what someone else is going to do, but it won't tell you what you want to know, which is whether that was the right thing to do or not.
Bingo. This article does a great job of explaining the flaws in Joe's Top 100 method, which I think apply to a baseline set by your league's tendencies. Using economic theory, VBD baselines should be based on value over replacement player. The trick in fantasy football is to use a value measure that accounts for greater value for those players that produce as starters for your fantasy team as opposed to those who are on your bench but are above replacement player value. I'm working on a model for this and hope to release it after a season of testing. A lot of this is based on the Retrospective Draft that was held in the Shark Pool a few years back. I'm sure I'll post about it in the Shark Pool, but you can follow @QSPAR on Twitter if you want to be aware of updates. It's not going to happen this season.
Thanks for sharing the link, I'll check it out. I may have seen it before but bears rereading. Eager to see your model when complete.
 
'Ignoratio Elenchi said:
'freshly_shorn said:
Actually, it does capture perfectly how my league acts.
I never said it didn't, but that's not by design. It's more a coincidence generated by the fact that most of your leaguemates are probably using projections that are roughly similar to the ones you use, so for example their model is telling them that Aaron Rodgers should go in the mid-first round, and so is yours. <snip>

Another little thought experiment. Say you think your league will draft 21 QBs in the first 12 rounds. So your baseline is the 21st best QB, and let's say you have that guy projected to score 280 points this year. Let's say you also have Aaron Rodgers projected to score 480 points this year. So Aaron Rodgers value in the MDB system is 200, and that might place him squarely in the middle of the first round. Now imagine that instead, you had determined that your leaguemates wait forever for backup QBs and therefore there will only be 14 QBs drafted by the end of the 12th round. So your baseline is the 14th best QB. Assume he's projected to score 320 points this year. So now Aaron Rodgers value in the MDB system is reduced to 160. Does it make sense that Aaron Rodger's draft stock would fall simply because your leaguemates decided to wait longer on drafting backup QBs? I'm not necessarily saying it doesn't make sense, but that's a consequence of MDB so consider whether that operates the way you think it should. Obviously that's a consequence of changing the baseline in any VBD system that uses a static baseline, so I'm not saying it's a flaw of MDB in particular, or even a flaw at all. I'm just curious if you think Aaron Rodgers would lose value in a scenario like that.
Point 1-Quite to the contrary, it can't be anything but by design. If I'm counting how many of each position my league takes by a given point in the draft, how can it look any other way? If past results say my league takes around 21 QBs by pick 144 every year, and I want to tailor my baseline to that fact, that is quite by design. Now, if as you say everyone shifts their philosophy and we're bitten by the 'past results don't guarantee future performance', then of course all this flies out the window- just like with any other system. Just because the X-number says the player should go in the first round has nothing to do whether he will or won't.. that's what ADP tells you.

Point 2-

MDB does not change the value of Rodgers as he compares to other QBs- the relative difference in X-numbers doesn't change. If the league decides to start snapping up RBs and leaving QBs to later, it doesn't change his value at all. The point is- MDB predicts how many of any position will be gone by a certain point. I'm not determining anything other than how a value a player within that context. And as you say- you can ask that question of any baselining model. I think this model does a better job of placing your personal valuations in the league (market) context then most, and in a simple fashion.
So perhaps a better way to use this would be to have two scores - one on this model and one on the last starter/first nonstarter model - and compare the two to find and better capitalize on market inefficiencies?Your model helps to show a variation of the league's valuation of a player, the other model shows their value to your lineup, and the differences can show you where the league is in line with the value you have for guys.

 
'Ignoratio Elenchi said:
'freshly_shorn said:
Actually, it does capture perfectly how my league acts.
I never said it didn't, but that's not by design. It's more a coincidence generated by the fact that most of your leaguemates are probably using projections that are roughly similar to the ones you use, so for example their model is telling them that Aaron Rodgers should go in the mid-first round, and so is yours. <snip>

Another little thought experiment. Say you think your league will draft 21 QBs in the first 12 rounds. So your baseline is the 21st best QB, and let's say you have that guy projected to score 280 points this year. Let's say you also have Aaron Rodgers projected to score 480 points this year. So Aaron Rodgers value in the MDB system is 200, and that might place him squarely in the middle of the first round. Now imagine that instead, you had determined that your leaguemates wait forever for backup QBs and therefore there will only be 14 QBs drafted by the end of the 12th round. So your baseline is the 14th best QB. Assume he's projected to score 320 points this year. So now Aaron Rodgers value in the MDB system is reduced to 160. Does it make sense that Aaron Rodger's draft stock would fall simply because your leaguemates decided to wait longer on drafting backup QBs? I'm not necessarily saying it doesn't make sense, but that's a consequence of MDB so consider whether that operates the way you think it should. Obviously that's a consequence of changing the baseline in any VBD system that uses a static baseline, so I'm not saying it's a flaw of MDB in particular, or even a flaw at all. I'm just curious if you think Aaron Rodgers would lose value in a scenario like that.
Point 1-Quite to the contrary, it can't be anything but by design. If I'm counting how many of each position my league takes by a given point in the draft, how can it look any other way? If past results say my league takes around 21 QBs by pick 144 every year, and I want to tailor my baseline to that fact, that is quite by design. Now, if as you say everyone shifts their philosophy and we're bitten by the 'past results don't guarantee future performance', then of course all this flies out the window- just like with any other system. Just because the X-number says the player should go in the first round has nothing to do whether he will or won't.. that's what ADP tells you.

Point 2-

MDB does not change the value of Rodgers as he compares to other QBs- the relative difference in X-numbers doesn't change. If the league decides to start snapping up RBs and leaving QBs to later, it doesn't change his value at all. The point is- MDB predicts how many of any position will be gone by a certain point. I'm not determining anything other than how a value a player within that context. And as you say- you can ask that question of any baselining model. I think this model does a better job of placing your personal valuations in the league (market) context then most, and in a simple fashion.
So perhaps a better way to use this would be to have two scores - one on this model and one on the last starter/first nonstarter model - and compare the two to find and better capitalize on market inefficiencies?Your model helps to show a variation of the league's valuation of a player, the other model shows their value to your lineup, and the differences can show you where the league is in line with the value you have for guys.
I like the idea of using MDB to gauge league valuation, but use another model to determine value to lineup. Shows you opportunity for arbitrage, which is helpful in deciding which positions to take early in the draft.</snip>

 
Last edited by a moderator:
So perhaps a better way to use this would be to have two scores - one on this model and one on the last starter/first nonstarter model - and compare the two to find and better capitalize on market inefficiencies?Your model helps to show a variation of the league's valuation of a player, the other model shows their value to your lineup, and the differences can show you where the league is in line with the value you have for guys.
I like that quite a bit. Great concept!
 
MDB is simply a way to determine which players should be your baseline. It is dynamic in the sense that it sizes itself to the league and number of starters in question. MDB tells you where to measure a player's X-Number from, just like any other baseline model.
So far, so good. If we're going to use some kind of VBD system, we need to pick baselines. You're proposing a way to determine what those baselines are.
Value, as Elenchi is stressing, only comes into play when you begin comparing players across positions, and while I use the term value throughout the article, what I mean, specifically, is the emphasis/premium/value that the league, as a whole, places on a particular position, as measured by the quantity taken at the MDB.
Here you're clarifying what you mean when you use the term "value." No major objections here though I don't think you're really capturing much useful information about the emphasis your league places on particular positions.
MDB is only meant to determine where to begin measuring, and I think this method does a much better job, in a simple way, then most models since it does incoporate your league's tendencies.
OK here's where you start to lose me. Why do you think this method does a much better job than most models? Why do you assume that incorporating your league's tendencies makes it better? Upthread Concept Coop asked a lot of really great questions which were basically getting to this point, and while you responded to all of them I don't think you ever really gave an answer.
I never said MDB tells you where to take players- that's up to you.
But that's the whole point of selecting an appropriate baseline. You pick the baseline, which produces a bunch of "x-values", which tell you where to take players. If you pick different baselines, then you're going to get different x-values, which would change where you would take players.
I think the market can tell you where to begin measuring value, as perceived by the league as a whole.
Why? Why would I begin there as opposed to somewhere else more suited to calculating actual player values (as opposed to the perceived values of your leaguemates)?
Your projections will then reveal- within the context of how much emphasis/premium/value the league places on a particular position.
Again, why? Why do I want to adjust my projections to account for how much emphasis the league places on a particular position? That's what ADP is for (either general ADP or, if you have a long-running league, then maybe you have more league-specific ADP). If I know that guys in my league place a premium on QBs, and therefore take QBs earlier, then I can keep that in mind while I'm drafting but (1) it doesn't make QBs more valuable (in terms of contributing to my fantasy score every week) and (2) that kind of info isn't captured by MDB anyway. Calbear said in one line something I've been trying to express in a thousand words: You're trying to incorporate information about where players are going to get picked without learning anything about whether or not those are the right decisions. The whole point is to exploit the mistakes your co-owners make, not rely on them to determine value.

Ultimately you can't say you're "just" choosing a baseline. Along with making projections, choosing a baseline is what VBD is all about - in fact, everything after that is elementary. The baseline you choose does directly determine where you draft players. If you use MDB, then MDB is going to tell you where to take players. As I illustrated earlier, if your league loves to take kickers early and often, then MDB is going to tell you to take kickers, too, which is the opposite of what you should do, and is a flaw unique to MDB because you're relying on what your leaguemates think players are worth instead of what the players are actually worth (or at least what you think they're actually worth).

What to choose as an appropriate baseline is an open question, but I still just don't see where you've made the case that your method is any better than any other.

 
I do like this system a lot and I've essentially been doing the same thing by dictating my baseline based on the top 100 players taken in the ADP. What always confused me in draft strategies is how to handle leagues that are drafting too QB heavy. Yes, part of me wants to say, "Fine you guys take your QB's and I'll stock up on my RB's" but in the end it seems like they all walk out of the draft with a good QB AND still get good RB's because the RB pool end up being less depleted.

 
Bingo. This article does a great job of explaining the flaws in Joe's Top 100 method, which I think apply to a baseline set by your league's tendencies. Using economic theory, VBD baselines should be based on value over replacement player. The trick in fantasy football is to use a value measure that accounts for greater value for those players that produce as starters for your fantasy team as opposed to those who are on your bench but are above replacement player value. I'm working on a model for this and hope to release it after a season of testing. A lot of this is based on the Retrospective Draft that was held in the Shark Pool a few years back. I'm sure I'll post about it in the Shark Pool, but you can follow @QSPAR on Twitter if you want to be aware of updates. It's not going to happen this season.
:blackdot: this is interesting
 
Point 1-

Quite to the contrary, it can't be anything but by design. If I'm counting how many of each position my league takes by a given point in the draft, how can it look any other way? If past results say my league takes around 21 QBs by pick 144 every year, and I want to tailor my baseline to that fact, that is quite by design. Now, if as you say everyone shifts their philosophy and we're bitten by the 'past results don't guarantee future performance', then of course all this flies out the window- just like with any other system. Just because the X-number says the player should go in the first round has nothing to do whether he will or won't.. that's what ADP tells you.
By design, you'll have 21 QBs off the board by pick 144. But that's it. If you happen to have 2 QBs off the board by round 1, and your league actually does draft 2 QBs in the first round, that's not a success of the system by design, that's simply a coincidental consequence of everyone on earth having access to the same kinds of information. The whole point of any "value" system is to determine where players SHOULD be drafted. Where they WILL be drafted is a totally different issue, and that's how you build a better team than your opponents. If your leaguemates take players too early, then they're losing value. If your leaguemates wait too long on other players, you can get that value by scooping them up. If you determine their value based on where they're being picked, you're doing it backwards and giving up the whole advantage of creating a value system in the first place. I hate to keep beating the same dead horse but I don't think you've addressed this: If your whole league drafted kickers too early, would you start drafting kickers, too? Because that's what MDB would have you do.

Point 2-

MDB does not change the value of Rodgers as he compares to other QBs- the relative difference in X-numbers doesn't change. If the league decides to start snapping up RBs and leaving QBs to later, it doesn't change his value at all. The point is- MDB predicts how many of any position will be gone by a certain point. I'm not determining anything other than how a value a player within that context. And as you say- you can ask that question of any baselining model. I think this model does a better job of placing your personal valuations in the league (market) context then most, and in a simple fashion.
Value is used across positions, not within positions. Of course his QB rank won't change relative to other QBs, the only thing that would do that is if you changed his projected stats. Value (which is directly determined by the baselines you choose) tells you when to pick a QB instead of a player at another position.For example, say you've made the following projections:

QB1 Aaron Rodgers: 480 points

...

QB14 Josh Freeman: 320 points

...

QB21 Andrew Luck: 280 points

RB1 Arian Foster: 330 points

...

RB49 Tim Hightower: 150 points

...

RB56 Jahvid Best: 130 points

If you use QB21 and RB49 as baselines, then Aaron Rodgers has a value of 200 and Arian Foster has a value of 180. So if you have the #1 pick, you would take Aaron Rodgers, because he's more valuable. However, if you use QB14 and RB56 as baselines (essentially saying your league would take a few more RBs and few fewer QBs by the end of the 12th round), then Aaron Rodgers has a value of 160 and Arian Foster has a value of 200, so you would take Arian Foster with the #1 pick.

Now as I asked before, do you think that the rate at which half of your leaguemates draft backup QBs and RBs in the later rounds affects whether or not Aaron Rodgers is more valuable than Arian Foster? Because that's the basis upon which you'd be making such a decision. Is this system working as intended?

 
Now as I asked before, do you think that the rate at which half of your leaguemates draft backup QBs and RBs in the later rounds affects whether or not Aaron Rodgers is more valuable than Arian Foster? Because that's the basis upon which you'd be making such a decision. Is this system working as intended?
I think this is the question that needs to be answered. Well said.
 
Nice job Fresh, good effort, keep at it and don't let folks try and stop you. Feedback is always good, there is room for everyone.
Nobody's trying to stop him - just helping understand the implications. This year I've really started to understand why you rub people the wrong way...
 
Nice job Fresh, good effort, keep at it and don't let folks try and stop you. Feedback is always good, there is room for everyone.
Nobody's trying to stop him - just helping understand the implications. This year I've really started to understand why you rub people the wrong way...
Show me in the rules where I am not allowed to offer support for other writers as I deem fit? Hope you have a better afternoon Instinct, sorry you feel that way.
 
"Do you think that the rate at which half of your leaguemates draft backup QBs and RBs in the later rounds affects whether or not Aaron Rodgers is more valuable than Arian Foster?

For example, say you've made the following projections:

QB1 Aaron Rodgers: 480 points

...

QB14 Josh Freeman: 320 points

...

QB21 Andrew Luck: 280 points

RB1 Arian Foster: 330 points

...

RB49 Tim Hightower: 150 points

...

RB56 Jahvid Best: 130 points

If you use QB21 and RB49 as baselines, then Aaron Rodgers has a value of 200 and Arian Foster has a value of 180. So if you have the #1 pick, you would take Aaron Rodgers, because he's more valuable. However, if you use QB14 and RB56 as baselines (essentially saying your league would take a few more RBs and few fewer QBs by the end of the 12th round), then Aaron Rodgers has a value of 160 and Arian Foster has a value of 200, so you would take Arian Foster with the #1 pick."

Yes, I believe that if my league drafts more RBs than QBs at a certain point of the draft, then RBs are perceived to be the more valuable position. Yes, changing the baseline changes the value. So what? The system is working just as any other baseline system does- you move the line, you change the value of every player on the board. MDB is no different from any other system in that regard.

I understand the argument that where you set the baseline is important- wouldn't have written the article otherwise. So if you couldn't care less how your league drafts overall, then MDB serves no purpose for you. But, the point of this whole thing was to come up with a standard way to simply set a line that is appropriate for any size league with whatever number of starters. If you have 6 teams with 10 starters, what the heck good is a baseline of pick 100? And if you go position by position, picking a distinct baseline for each, how exactly do you determine what a baseline player is?

Obviously, this whole exercise is about identifying value- finding players that are undervalued by the 'market' and picking them at the appropriate spot. I totally disagree that a players value tells you when to take them. ADP tells you more about that. VBD tells you how a player ranks against other players and ranks them accordingly. You have to decide when to take them, using all available information.

" I hate to keep beating the same dead horse but I don't think you've addressed this: If your whole league drafted kickers too early, would you start drafting kickers, too? Because that's what MDB would have you do."

OK, lets go to fantasy land. Let's say the 32 picks in the draft are kickers, before the MDB. So, out of the 144 picks at the MDB, 32 are kickers. Looking at DD, Gostowski in my league is projected to have 136 FP. Kicker #32, Josh Scobee, is worth 95. So, with Mr. Scobee being the baseline, Gostowski's X-number is 41. OK, so we have the 112 other picks now. Let's say 21 QBs still went. Rodgers is worth 441, QB #21- Jay Cutler- is worth 259. That's an X-number of 182. So, that proves that MDB holds up just fine in this ridiculous example. It's the projections that keep everything honest. Less valuable positions will remain so, and this method accurately conveys exactly where your league perceives value to be.

 
Last edited by a moderator:
Nice job Fresh, good effort, keep at it and don't let folks try and stop you. Feedback is always good, there is room for everyone.
Nobody's trying to stop him - just helping understand the implications. This year I've really started to understand why you rub people the wrong way...
Show me in the rules where I am not allowed to offer support for other writers as I deem fit? Hope you have a better afternoon Instinct, sorry you feel that way.
Thanks bud. Just academic conversation going on in here. I love it. I'll be using MDB to my heart's content, so I'm good! Just seeing what the masses think, and very happy to have an engaged audience. Even if they think I'm a complete dolt. :thumbup:
 

Users who are viewing this thread

Back
Top