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!
Anyone have a FF trade value chart? (1 Viewer)
- Thread starter zed2283
- Start date
David Yudkin
Footballguy
CommuterMan
Footballguy
For dynasty leagues try the one that Jeff Pasquino created: The Dynasty Calculator Sorry I can't seem to get the link to post but if you go to any of Jeff's postings in these threads the Dynasty Calculator is in his sig at the bottom...For redrafts try the FBG's under tools.... David provide the link...
Last edited by a moderator:
robb
Footballguy
I believe this is the Dynasty Calculator article:
http://subscribers.footballguys.com/2006/0...dynastycalc.php
Robb
http://subscribers.footballguys.com/2006/0...dynastycalc.php
Robb
Toads
Footballguy
It's called a Draft Pick Trade Chart and it's proven to be illusive.Reason it's illusive? Because no one has one for an 8 team, 10 team, 12 team, 14 team and 16 team league. The "draft pick calculators" seem to work up to about a 10 team league....it's Yahoo time!
If they do have a version at all, they think that the NFL 16 team version works, which it doesn't.
It makes sense that the #16 pick in a 16 team league has a different value than the #16 pick in an 8 team league.
So, the question: where is there a 16 team chart for a six round draft?
Last edited by a moderator:
Jeff Pasquino
Footballguy
I made the charts and it'll be posted shortly to the site.
Jeff Pasquino
Footballguy
Jeff Pasquino
Footballguy
Hi Magic,Can you elaborate on the bolded part?
While I can see how Team 16's first pick is more intrinsically valuable to Team 16, as it is his (or her) first pick, but I fail to see why Pick 16's value varies from league size to league size.
The pick values are relative to the rough value of each player. So while in an 8 team league, Team 1 gets picks 1 and 16 and lots of value, so do the other 7 teams (getting 2/15, 3/14, etc.).
In a 16 team league, the hope would be that the value in getting pick 1 is offset by having to wait until Pick 32 to select the second player.
The values are all relative to one another, and are a guide. Some years Players 1-3 are >>> the next tier, but in other years the difference isn't as great. Each year is different.
GregR_2
Footballguy
Picks values change when the league size changes because the value of the player's being selected with the picks change. Because positions do not all decline in value at the same rate as you go down the positional list, those changes in value are not relative.Take last year's results in one of my leagues and go from an 8 to a 12 team league with 1 QB, 2 RB, 2 WR, 1 TE. QBs gain 28 VBD value points, RBs gain 20, WRs gain 21, and TEs gain 18. Going from 8 to 12 teams in a 2 QB league would change QB value by 68 using last year's results. The value of picks will change going from an 8 to a 12 team league both because the order of players picked will change, and because even if the players at 2 given picks stay the same, they will change in value by different amounts if they are at different positions, so the change is definitely not a relative one.
If you need to see more concrete examples just grab the VBD app or Draft Dominator and change the league size from 8 to 12 and recalculate and look at two picks and see how the value of the players taken there changed despite it being the same starting lineup and same set of projections used to find the value of the players.
Last edited by a moderator:
Toads
Footballguy
Hi, Jeff.I keep on trying to resurrect this thread, or should I say these threads, because there are more than one of them. "They" all have one thing in common: there's never a published result for the 8-16 team leagues.
If you want to construct a trade in any league, it's a function of commonly held values....evidence the NFL and it's OBJECTIVE 32 team Trade Value chart. It's a commonly held point of reference. The subjective evaluation comes in when a team is motivated to give, or in the opposite case....to extract, differential value (evidence the CLE and DAL trade for Brady Quinn, CLE paid a price there.).
So, back to the 16 teamer. It makes sense that once published, these charts would exibit some relationship to statistical formula, that the variance in the curves that are demonstrated in the backup article that is at the bottom of the page in the linked article and used in the linked article, apply. The proportional decline in value in an 8 teamer is different than the proportional decline in value in a 32 teamer.....and therein lies the challenge.
Another way of looking at the proportional decline statement is to take an 8 team draft over six rounds....it declines over 48 picks. A 16 team draft over six rounds declines over 96 picks. Hence, the value for any isolated pick in these draft, choose a pick, say the #18 pick, has to be different in each isolated draft.
The 16 teamer....yet again: if you "interpolate" the 32 teamer into the 16 teamer by taking the average for each of the two consecutively listed picks (hence, the interpolated #1 pick is actual #1 pick of 3,000 points + the actual #2 pick of 2,600 points / 2 or 2,800 points for the "adjusted" #1 pick), you can modify the 32 teamer into the 16 teamer by doing this exercise for the entire 32 team chart. When you are finished with this exercise, you've "interpolated" a 16 Team Draft Value Trade Chart. It declines over Seven Rounds, not the six that most leagues employ....but it's "close enough for the girls we go with."
Only one problem with the above example: it declines at a 32 team (choice/round) rate....not at 16 team (choices/round) rate. Because of this, it's inaccurate for a 16 team (choices/round) scenario.
So, Jeff, when we find a stastical trend analyser that can comprehend this mess, it'll be interesting to see if the 32 teamer has any relationship to the 16 teamer and the 16 teamer has any relationship to the 8 teamer? If our hypothesis is true, they have no relationship to each other because they all decline as a function of their own mathmatic curves.
Last edited by a moderator:
Toads
Footballguy
Jeff Pasquino
Footballguy
Let's understand what I did (when it gets published, and Dodds and Clayoton are working on this) - All I did was to lay out the Pick Values for each individual pick for each slot in a chart. That is, Pick 33 is worth Pick 33 - whether it is in the 5th round of an 8-team draft, the 4th in 10-teams, or the third in 12-teams (or the first third-round pick in a 16 teamer, etc.).Magic Desert Toads said:
Now debating whether or not that makes sense is a different issue altogether, and it may or may not have merit, but I agree with you that these charts aren't published anywhere as a reference. That's what I worked to correct.
Jeff Pasquino
Footballguy
Hi GregR,You a solid analyzer here in the Shark Pool, but I think we're going to disagree on this one.GregR said:
I believe you're interpreting Pick Value as VBD value, which is a mistake.
Do VBDs change based on league size? Absolutely. (It also matters what flavor of VBD you take - worst starter, average starter, "Joe's Secret Formula", etc.) However, relative to one another, picks are picks. The order in which you would select certain positions or players will vary based on league size, but overall I don't think this would affect the value of Pick 37 as Pick 37.
To look at your example, I did change the Draft Dominator from 8, 12, and 16 teams and see where the baseline was. In all 3 cases, 1 QB (roughly), 2.4 RBs, 3.7 WRs and about 0.5 TEs had VBDs > 0.
It's been a long day and I may have missed something, but I don't see the discrepancy in interpreting the VBD this way.
Toads
Footballguy
Jeff, let's go back to the orginal link.When I utilize the link, there's alot of information there. There are several attempts to identify progressions, including a "normalized NFL chart". So, let's assume that you are on to something and that it takes a bit more pouring over to make sense of what's there.
The one part that I can't get to work is the "download the chart here" reference as there's a sample screen and some color coded input directions but there is no download that I can see or get to operate.....nothing lights up when you attempt to "download here".
Can you assist with that download?
If you can help and get that to work, it's simply a matter of using it and seeing what happens when data is entered.
Last edited by a moderator:
Jeff Pasquino
Footballguy
What original link? My Calculator?The link for it and the article are in my signature.Jeff, let's go back to the orginal link.When I utilize the link, there's alot of information there. There are several attempts to identify progressions, including a "normalized NFL chart". So, let's assume that you are on to something and that it takes a bit more pouring over to make sense of what's there.
The one part that I can't get to work is the "download the chart here" reference as there's a sample screen and some color coded input directions but there is no download that I can see or get to operate.....nothing lights up when you attempt to "download here".
Can you assist with that download?If you can help and get that to work, it's simply a matter of using it and seeing what happens when data is entered.
Toads
Footballguy
Thanks, that signature link get's me there....the reference in the article didn't.The Figure 1-Rookie Draft Pick Values (in the article) and it's reference to one scenario "decreasing in value far faster" than another example and all this discussion as it relates to the Dynasty Factor....the sound of the rubber hitting the road.What original link? My Calculator?The link for it and the article are in my signature.Jeff, let's go back to the orginal link.When I utilize the link, there's alot of information there. There are several attempts to identify progressions, including a "normalized NFL chart". So, let's assume that you are on to something and that it takes a bit more pouring over to make sense of what's there.
The one part that I can't get to work is the "download the chart here" reference as there's a sample screen and some color coded input directions but there is no download that I can see or get to operate.....nothing lights up when you attempt to "download here".
Can you assist with that download?If you can help and get that to work, it's simply a matter of using it and seeing what happens when data is entered.
It looks like the Dynasty Factor is the key to the relative decline in the Draft Pick Value in the Trade Chart.
I varied the input to see how that effected the result. As a for instance, since I'm searching for a 16 teamer, I varied the RB starters between 1 and 2 and varied the WR starters between 2 and 3 (this mirrors a flex condition....start either 2-RB's and 2-WR's or start 1-RB and 3-WR's) while maintaining a constant roster size, and the 2-2 combo had a Dynasty Factor of 4.152 vs. the 3.907 for the 1-3 combo.
So, you've done it: you've created the perfect mouse trap.
Added By Edit: The Calculator doesn't take into account the number of drafting rounds but it looks like the size of the sample is static at 100 draft picks....?
Last edited by a moderator:
humphrp
Footballguy
I am trying to put together a trade package for the 1.02 in my draft (CJ). In the process of using this calculator I have two questions.
One, I could not see where flex positions can be input in the starting lineup. Am I missing something or is it just not there. If it is not there how hard would it be to add that.
Two, What are ya'll thoughts into valuing future draft picks. What kind of present value would you give these. I have often valued them at about 75% of current value.
One, I could not see where flex positions can be input in the starting lineup. Am I missing something or is it just not there. If it is not there how hard would it be to add that.
Two, What are ya'll thoughts into valuing future draft picks. What kind of present value would you give these. I have often valued them at about 75% of current value.
Jeff Pasquino
Footballguy
Thanks, I'm glad you like the tool.I believe it goes up to 190 picks, but after a certain point they go to zero value so it will look like there's no pick there.Magic Desert Toads said:Thanks, that signature link get's me there....the reference in the article didn't.The Figure 1-Rookie Draft Pick Values (in the article) and it's reference to one scenario "decreasing in value far faster" than another example and all this discussion as it relates to the Dynasty Factor....the sound of the rubber hitting the road.What original link? My Calculator?The link for it and the article are in my signature.Jeff, let's go back to the orginal link.When I utilize the link, there's alot of information there. There are several attempts to identify progressions, including a "normalized NFL chart". So, let's assume that you are on to something and that it takes a bit more pouring over to make sense of what's there.
The one part that I can't get to work is the "download the chart here" reference as there's a sample screen and some color coded input directions but there is no download that I can see or get to operate.....nothing lights up when you attempt to "download here".
Can you assist with that download?
It looks like the Dynasty Factor is the key to the relative decline in the Draft Pick Value in the Trade Chart.
I varied the input to see how that effected the result. As a for instance, since I'm searching for a 16 teamer, I varied the RB starters between 1 and 2 and varied the WR starters between 2 and 3 (this mirrors a flex condition....start either 2-RB's and 2-WR's or start 1-RB and 3-WR's) while maintaining a constant roster size, and the 2-2 combo had a Dynasty Factor of 4.152 vs. the 3.907 for the 1-3 combo.
So, you've done it: you've created the perfect mouse trap.
Added By Edit: The Calculator doesn't take into account the number of drafting rounds but it looks like the size of the sample is static at 100 draft picks....?
That is to say, with the Dynasty Factor (DF) >1, picks will be <the Dodds calculator value. as the DF increases, the values decrease.
As the DF approaches 1, it goes back to Dodds' normal values.
There is no limitation to the number of rounds, but as I stated before after 100 rookies it looks like no value (perhaps even sooner if DF is high).
Also most rookie drafts don't even get to 100, even with IDP. Perhaps a 16 teamer with 7 rounds would, but that's really pushing it.
Let me know if there's additional questions.
Jeff Pasquino
Footballguy
Flex isn't there, but I will likely add it in the next turn of this tool. If you find other features to add, let me know - right now flex is about all I've heard missing.As for PV of future picks, I discount 1 round per year in the future (but some would disagree on this board).humphrp said:
GregR_2
Footballguy
I don't understand how it can't affect the value.The value of a pick is based on the value of the player that can be taken with that pick. There is no absolute value associated with Pick 37 that is non-changing. The value of pick 37 is relative to what player can be taken at it and what his value is vs the players who will be taken at other picks. Right?Go from one league to another and the value of those players change so the value of the picks change with them. Go from one year to the next and the value of the picks change too.The way to evaluate any trade is to evaluate the team before and after and see how much the team improved in points scored, depth, etc. When you can replace a pick with the player that will likely be taken with the pick, you shouldn't ever use a trade value chart, you should just deal with the players and your beliefs about them.But when we're dealing with a year or two years away it's harder to put player's names to the picks. So then what can you do to determine the value of the pick? You can grab some historical numbers on the value of players taken at those picks and use that as a guideline, which is what a pick value chart in FF should be. I'd ask, if the value of players taken with a pick are not the method used to determine value in the current charts, then exactly what it is based on?
Jeff Pasquino
Footballguy
GregR,You're throwing a ton of stuff into one number - which really cannot be done, yet there are charts that try to do just that.Charts are a guide - a tool - nothing more. Year-to-year variability cannot be charted. Your particular beliefs on a certain player in a trade cannot be charted. So, that being said, I'm going to agree with you on a few points you made and try to go from there.
Agreed, provided you can quantify it (no subjectivity here - a player you like or dislike - it is just player #37, for example. In a perfect 20/20 hindsight world, we'd all take this player at Pick #37.If you can't agree with this (I will call this Principle #1), we're stuck.
Disagree and agree at once. I agree that it does change year to year, but you have to make the assumption / leap of faith that such a number can exist - otherwise there would never be a chart to use as a guide in the first place.This is the basis of every chart / calculator in existence - there's no variable for "strength of field" or "year to year variation" - the value of Pick #37 in 2007 is the same as Pick #37's value in 2006. Again, if you can't agree with this point either (I will call this Principle #2), we're stuck again.
Agreed - this is applying VBD.
This is where we start to diverge. The list of the players and who is in what order varies from league to league, but Player #37 in 20/20 hindsight is/was/always will be Player #37. Now, if you try and make the case that this changes year-to-year, you're violating Principle #2, and we're stuck.
Whoa - hang on here. While I agree that this is likely the best way for an owner to view his own team and trade offers, you've now gone away from quantified analysis (a pick value chart) and to your own gut or beliefs. That's perfectly fine and a good way to go (I'm not 100% analytical at all), but if you're making a statement about the value of a pick vs. the number on a chart, this isn't a good argument. We've violated Principle #1, where Player #37 is Player #37 (in a perfect world).
See Principles #1 and #2. The basis of all charts is to boil it all down to one number, and have those numbers reflect relative value to each other. Dodds pegged #1 at 1889 points. I've seen it so much I don't even have to look it up, I just know that number now. Does that help me? What does 1889 mean? Absolutely nothing without context. The NFL Draft Chart starts at 3000. Is that better or worse? Does it matter?The answer is no. What matters is that Pick #2 in the NFL is 2600 points, or 13.3% less value than Pick #1. That's all that matters. The absolute value of that number means nothing without the context of what #2 or #3 or the rest of them are worth.The same is true for 1889. Pick #2 is 1823. That's a difference of 66 points, or about 3-4%. Now if you believe that LT2 > SJax or LJ by more than 4%, then you wouldn't accept 4% of value plus Pick #2 for Pick 1 - but you'd be violating the chart - WHICH IS PERFECTLY FINE TO DO.The chart / values are a GUIDE, not mandatory rules of trading.
Last edited by a moderator:
GregR_2
Footballguy
I wrote a reply and then deleted it because it was just going to continue going down some side path that has little to do with the point we've been trying to make. I don't have long so let me try to put it in a nutshell.
The point myself and others have been making is:
Whoever comes up with a draft pick value chart, has to have some method for determining the values.
We would assume they aren't pulling the values out of their behind, so they must have some heavy basis on fantasy drafts and players from past seasons.
Some system for quantifying player value has got to be involved, such as VBD but not necessarily in the normal fashion (i.e. not last starter-ish baselines).
If player value is being determined in such a fashion, we need to account for the fact that values change from one league set up to the next. Because those values were what were fed into the draft pick value number, and they aren't the same for all leagues.
That value distribution changes over a span of years in the NFL. That a fantasy draft pick value chart made in 1999 probably isn't better for use in 2007 then one made in 2007 that incorporated trends from 2000-2006. Or even better, incorporates beliefs about 2007/2008.Now, if you are telling us #2 is wrong, that Dodds just wrote down some numbers from his gut, we'll all do a
.
Barring that, what we've been trying to discuss with you is #4 and/or #5. If you don't agree with this, which you've stated you don't, then please discuss the actual method used to come up with Dodd's 1889 or your number or whoever's it is, and why the variation over time or over league setup washes out and doesn't change things enough that their inclusion is warranted. Show us some actual results that show the variation is magnitudes smaller than the changes between picks which leads you to believe they aren't necessary.
The point myself and others have been making is:
Whoever comes up with a draft pick value chart, has to have some method for determining the values.
We would assume they aren't pulling the values out of their behind, so they must have some heavy basis on fantasy drafts and players from past seasons.
Some system for quantifying player value has got to be involved, such as VBD but not necessarily in the normal fashion (i.e. not last starter-ish baselines).
If player value is being determined in such a fashion, we need to account for the fact that values change from one league set up to the next. Because those values were what were fed into the draft pick value number, and they aren't the same for all leagues.
That value distribution changes over a span of years in the NFL. That a fantasy draft pick value chart made in 1999 probably isn't better for use in 2007 then one made in 2007 that incorporated trends from 2000-2006. Or even better, incorporates beliefs about 2007/2008.Now, if you are telling us #2 is wrong, that Dodds just wrote down some numbers from his gut, we'll all do a
.Barring that, what we've been trying to discuss with you is #4 and/or #5. If you don't agree with this, which you've stated you don't, then please discuss the actual method used to come up with Dodd's 1889 or your number or whoever's it is, and why the variation over time or over league setup washes out and doesn't change things enough that their inclusion is warranted. Show us some actual results that show the variation is magnitudes smaller than the changes between picks which leads you to believe they aren't necessary.
Biabreakable
Footballguy
This is one nut that I am hoping never is cracked. Personaly I believe one of my edges on my compitition in dynansty leagues stems from my evaluation of rookie draft picks and thier mutable value relative to veteran players. This is an edge that I want to keep.
As unknowledgable as I am of the college game never watching it I still seem to be able to evaluate college talent well enough in how it will translate into pro level performance that I am ushualy able to make good use of rookie draft picks that I aquire. Also in trading I have a tendency to be able to aquire picks at a value that ends up being an advantage to me.
The closest thing I have seen to defining rookie pick value relative to veteran players is beto's study which generaly is very sound. That study uses historical performance of players taken in rookie drafts and how they have performed comparativly to veteran players. This study is objective although it follows the patterns of beto's specific league I do think that there are enough similarities in scoring systems and league dynamics that his study can be used as a general template of relative value of rookie picks compared to veteran players. This comparison is important because by having veteran players as a basis of comparison one can use VBD and ADP principles to find at least approximate player value of the picks. The positions or specific players will change from year to year based off of league rules and dynamics as well as each years rookie crops relative strengths and weaknesses. But it is at least a good guideline for using value principles such as VBD and ADP even as the specific players change from year to year.
As far as comparing rookie picks to rookie picks based off of what seems to me a arbitrary number value system that I do not see a correlation for in actual performance... well I think the NFLs value chart is really off and so any chart that would mirror that would also be very off. Thinking from a FF perspective of course is different than the goals of a NFL team. But when looking at NFL drafts I often see trade scenarios that seem like good moves for teams but that are never done.
Why?
Because of the value chart? Or more likely other considerations...
But for example the Eli Manning trade in 2004. Has to be one of the more one sided deals I have ever seen even before we saw how the players that were picked with those picks panned out (we are still evaluating that). The Vick trade before that also seems very skewed and one sided.
Not somthing I want to use as a tool in making my decisions. I will stick with what I have been doing and not really looking to completly lay out all the different things that I do when evaluating relative value of picks to player performance in the dynamic of dynasty leagues. Beto has done enough damage in clearing up that picture in my opinion as it is.
As an asside if more people start using value pick calculators such as Jeff has developed then I will pay more attention to them so I can find the loopholes in it to be exploited in specific trades that involve picks where I am targeting specific players with those picks. As it is right now not enough people use it for me to consider exploiting these loopholes yet.
As unknowledgable as I am of the college game never watching it I still seem to be able to evaluate college talent well enough in how it will translate into pro level performance that I am ushualy able to make good use of rookie draft picks that I aquire. Also in trading I have a tendency to be able to aquire picks at a value that ends up being an advantage to me.
The closest thing I have seen to defining rookie pick value relative to veteran players is beto's study which generaly is very sound. That study uses historical performance of players taken in rookie drafts and how they have performed comparativly to veteran players. This study is objective although it follows the patterns of beto's specific league I do think that there are enough similarities in scoring systems and league dynamics that his study can be used as a general template of relative value of rookie picks compared to veteran players. This comparison is important because by having veteran players as a basis of comparison one can use VBD and ADP principles to find at least approximate player value of the picks. The positions or specific players will change from year to year based off of league rules and dynamics as well as each years rookie crops relative strengths and weaknesses. But it is at least a good guideline for using value principles such as VBD and ADP even as the specific players change from year to year.
As far as comparing rookie picks to rookie picks based off of what seems to me a arbitrary number value system that I do not see a correlation for in actual performance... well I think the NFLs value chart is really off and so any chart that would mirror that would also be very off. Thinking from a FF perspective of course is different than the goals of a NFL team. But when looking at NFL drafts I often see trade scenarios that seem like good moves for teams but that are never done.
Why?
Because of the value chart? Or more likely other considerations...
But for example the Eli Manning trade in 2004. Has to be one of the more one sided deals I have ever seen even before we saw how the players that were picked with those picks panned out (we are still evaluating that). The Vick trade before that also seems very skewed and one sided.
Not somthing I want to use as a tool in making my decisions. I will stick with what I have been doing and not really looking to completly lay out all the different things that I do when evaluating relative value of picks to player performance in the dynamic of dynasty leagues. Beto has done enough damage in clearing up that picture in my opinion as it is.
As an asside if more people start using value pick calculators such as Jeff has developed then I will pay more attention to them so I can find the loopholes in it to be exploited in specific trades that involve picks where I am targeting specific players with those picks. As it is right now not enough people use it for me to consider exploiting these loopholes yet.
Jeff Pasquino
Footballguy
Bia,Have you read my article and/or tried the calculator?It is solely for rookie picks.As for beto's study, I agree and it validates my rough rule of thumb of 1 slot == 1 round.Also, what's your take on larger (and/or smaller) leagues - redraft and dynasty?
Biabreakable
Footballguy
I did read it some months back Jeff somewhat lazily I admit as I did not see it being somthing that works for me. I do understand that it is for rookie picks vs. rookie picks but what myself and I think Greg are saying is that those picks will = players and those players careers once the picks are made. And the only tangible way to compare those values is based off of those players performance which straighten me out if I am wrong.. but I do not think your calculator takes this into account?That is somthing that beto's study does do even though it is geared for his specific league. I am able to make adjustments for my leagues.I know I have said this many times before but I do think rookie picks have a value in themselves as extra roster slots. So in my eyes this adds to thier value. Also the wildcard nature of picks that can end up being higher or lower in a round and can be used to fill any position adds to a picks value in my eyes until that pick is used. I find picks to be great tools in trades also going both ways. The wildcard nature of picks helps them fit any teams needs more easily than a specific player that every owner has a different opinion of.As far as league size on the extra roster slot dynamic of rookie pick value it definitly does matter and IIRC your logic on this makes sense as the larger the roster size the more useful later round picks become compared to smaller rosters. Also obviously the number of teams has an impact on each rounds value. A 1st round pick in a 8 team league has less variance in overall slotting than a 1st round pick in a 16 team league. This gets offset however by how likely a 1st round (or later round) pick will be able to help a team in a 8 team league compared to a 16 team league. In a 8 team league rookie picks are a lot less likely to help a team even if those teams have deep rosters when compared to a 16 team league.*edit coz I don't know what a calburator is
Last edited by a moderator:
Jeff Pasquino
Footballguy
Based on your answers (and I believe GregR's as well) neither of you want to use a calculator for your evaluations - which I can appreciate. I would, however, suggest that many people ARE looking for one, so knowledge of what they are looking for and reading about is valuable. To make a poker analogy (it is WSOP time) - if you know that certain players will play certain cards a certain way, you can tell what they have by their playing style - even if you don't play that way. In essence, you know "the book" on them even if you don't believe in the book.I make mention of beto's study and I've gone over it several times, but there's really no point in making a calculator match beto's study - it is what it is. The "1 slot = 1 round" rule of thumb is a close enough approximation to his study and you can obviously tailor it to your league and/or preferences.
I'm guessing you use a taxi squad / development squad?
Biabreakable
Footballguy
Not at all.I look at it this way if you have a roster of 36 players for example and a rookie draft of 6 rounds each team then has 42 slots of players when you consider the rookie picks as roster slots.If you trade for additional rookie picks this further increases your total roster value. Lets say I traded 3 players for 3 picks in the 2008 draft. I can now pick up 3 players off of waivers for developmental purposes still within my 36 player limit. But I now also have 9 rookie picks that can be used in trades or on new players to upgrade my roster in the next rookie draft.Now not all picks are equal which is the main point of a rookie pick calculator or beto's study showing pick value relative to players rostered.The rookie picks are the taxi/developmental squad.The smaller the league the less likely rookie picks are to be good enough to roster. But at the same time the smaller the roster size the greater the benifit rookie picks have on overall roster value when one looks at them as extra slots in the way I have described.
GregR_2
Footballguy
I have no problem with having a calculator of some sort to use as input for my decision. But if so I want a calculator that reflects the information that is available.If it's a trade of rookie picks from this year we have a very good idea what the pool of players looks like and the value of the picks should be based very strongly on those, and not based on what the value of pools of players in past years were.If it's a trade of future years where we can't even predict who will be the top players (see Matt Leinart 2005 vs 2006 draft predictions), then I want to know that the calculator reflects my league.I can guarantee you that any such calculator made for a standard setup won't come anywhere close to my dynasty league. I wouldn't expect a FBG calculator to necessarily have as much flexibility as would be needed to support my own league (though if it did take people's league parameters as an input similar to the VBD app, man you would have an offering that would blow away any other site). But I would expect it to be able to differentiate between the common league differences like 8/10/12/16 teams, IDPs or not, etc.
Jeff Pasquino
Footballguy
There is no definitive way to evaluate future draft picks, mostly because you don't know what pick they'll be until you are less than 6 months (or less) from the NFL Draft.
A good team may have a bad season and pick first, and a bad team may catch lightning in a bottle and eek into the playoffs then win the league.
These things happen. Play long enough, and you'll see it all.
So, as a result, you can't project a value on a pick > 6-9 months in advance. It doesn't work. You can try (like I think Cleveland, KC and Oakland will struggle this year and be picking early in 2008 (or Dallas will ), but there's no guarantee.
GregR's thoughts on a calculator based on VBD is interesting. I'll have to think on that one.
A good team may have a bad season and pick first, and a bad team may catch lightning in a bottle and eek into the playoffs then win the league.
These things happen. Play long enough, and you'll see it all.
So, as a result, you can't project a value on a pick > 6-9 months in advance. It doesn't work. You can try (like I think Cleveland, KC and Oakland will struggle this year and be picking early in 2008 (or Dallas will
), but there's no guarantee.GregR's thoughts on a calculator based on VBD is interesting. I'll have to think on that one.
TBLCommish
Footballguy
GregR_2
Footballguy
Not having any luck in this thread getting people to discuss methods for creating such a chart, but maybe we can spark the conversation by using this as an example? Please don't take these comments personally, if you know me you'll know by now I'm very practical and my desire is to discuss how to create the best chart possible, not to go criticizing others work for criticism's sake.On this chart, I can't say for sure what went into making this one, but it looks like it took straight from the NFL trade value chart for the first 16 picks. Then it used a drop off 25 points each pick for round 2. Then 20 points for 2 picks, then 10 points out to pick 60, then 8 point drops for a couple of picks, then 6, then 5 for awhile, then 4.TBLCommish said:
My first thought is that I don't think it's likely that the real value of the fantasy rookie picks is the same as in the NFL, or that it ends up being so uniform. I could be wrong, it could be that this was made from looking at actual fantasy trades to assign the values, but it seems too similar to the NFL chart to me for this to be likely, thus my comment.
The NFL really has no such thing as FF's tangible fantasy points for quantifying the value of rookie players. How do you quantify the value of a one particular left guard vs a WR vs a weak side LB vs a DE used on pass rush situations but not 3rd and short, accounting for the talent of each individual as well as positional value?
So when they ended up doing their draft value chart, they couldn't easily base it on direct quantified predictions of player value since those are a lot harder to quantify in the real NFL. They instead created it by looking at actual trades that had taken place over the years and coming up with a chart that reflected the subjective value held by the teams based on those actual trades.
I'm not saying that's a bad thing. In fact if someone did that for their own league I think it could be useful because that tells you what owner perception on the value of the picks is. Though I'd imagine you should have a lot of trades as input to work off of.
But the kind of chart I'm talking about for FF that I think is the most useful isn't based off of owner perception, but off of the actual value of the players. This isn't a big change even from the NFL, is it? How many posts and articles and TV talking head discussions are there about the outrageous cost of trading for the #1 pick? That comes from when people look at what is the actual expected value of the player taken #1, vs what do teams ask for in trade to move up there, and the two numbers are often very far off (see Chargers-Giants Eli Manning trade).
If you had both of these charts, one to reflect what owners are often seeking in trade, and one that tells you what the value of the players normally taken there actually is, wow what a great tool that would be! You could look at what your league's owners trade tendencies are, and compare it with what you might expect the actual fantasy value of the players to be, and find where the value is greater than the owner's expectations and go after those picks in trade talks.
In an ideal world here is what I think a site like FBG should offer.
1. A chart that is made at the end of the college season, or at least near the end, when there is a pretty decent idea about what players will be available for the draft. As an example, the changes in value from picks 1.1 to 1.5 this year were a lot different than people saw last year. So come December or January when we have a good idea who is going to be drafted and where we expect to slot them in a dynasty sense and with what kind of gaps between, that is what the value chart based on actual player value should work off of.
2. A chart that is made using an average of the value at that position from previous years, which is updated with new data as each new year happens to keep reflecting the changing trends in the NFL like the change in QB value with the rule emphasis change on pass interference.
Further, I think both charts should reflect league size and common league parameters. That means a separate one for an 8 team league than for a 16 team league, unless it is shown by actually creating the charts and comparing them, that the changes between such leagues is neglible. Which I've never seen anyone do, and which until I do I won't take at face value that one size chart fits all.
Does that make sense to anyone?
Toads
Footballguy
It would be interesting for you to spell out the league's starting mix, roster size, ec., in short, all the variables that go into Jeff's calculator.Let's install those variables into Jeff's calculator and compare one guy trading the 1.05 for the other guys 1.09 and 1.11 and see what comes up.TBLCommish said:
This calculator is interesting. Take, for instance, a 16 team league with 32 roster size/14 starters where one of the starter's is a Punter. You can actually get a modified trade value for the league that includes an "odd starter" (the Punter position) by considering him under the Def (Team D) designation.
Last edited by a moderator:
GregR_2
Footballguy
Wanted to make one other comment about using actual player value when we have an idea what it is, but didn't want to put too many discussion points in the other post.
Jeff, I don't think enough stress is made to FBG users when it's appropriate to use the existing draft pick value chart, and when it isn't. We see countless posts asking if there is such a chart, and they go up in frequency come July and August. I'm not talking rookie pick here I'm talking redraft, but much of the same would apply to rookies.
But does anyone really believe the majority of people asking for such a chart are wanting it to trade the NEXT year's picks? Of course not. Most are asking because they have a draft coming up in a month or less and they want to see if an offer they were made is good or not.
And I hope everyone would agree that when we're at a point when we have projections for the players involved in the draft, we shouldn't be using charts that try to approximate by averaging the value of the pick from previous years when we can use the actual value of the player we expect to be taken with the picks.
As I said previously, ideally the draft pick value chart for the coming season should be based off of FBG projections for that season once they exist, and then there should be a separate chart that would be used for picks a year out or more.
But I bet you that 75% of the use of the existing chart, is for picks for the current year when projections for those players already exist.
Jeff, I don't think enough stress is made to FBG users when it's appropriate to use the existing draft pick value chart, and when it isn't. We see countless posts asking if there is such a chart, and they go up in frequency come July and August. I'm not talking rookie pick here I'm talking redraft, but much of the same would apply to rookies.
But does anyone really believe the majority of people asking for such a chart are wanting it to trade the NEXT year's picks? Of course not. Most are asking because they have a draft coming up in a month or less and they want to see if an offer they were made is good or not.
And I hope everyone would agree that when we're at a point when we have projections for the players involved in the draft, we shouldn't be using charts that try to approximate by averaging the value of the pick from previous years when we can use the actual value of the player we expect to be taken with the picks.
As I said previously, ideally the draft pick value chart for the coming season should be based off of FBG projections for that season once they exist, and then there should be a separate chart that would be used for picks a year out or more.
But I bet you that 75% of the use of the existing chart, is for picks for the current year when projections for those players already exist.
Toads
Footballguy
Loud and clear, Kemosabe.I'm getting a lot out of this cause I've been working on it for a long time....kinda like the Quest For The Holy Grail.
I've been playing with Jeff's Calulator and I find that it has capacities that I never dreamed that it would have:
1.) You can install flex into a league by using a half player and watching the Dynasty Value change as you do it. Now, that's incredibly cool. That "half player" is the Flex player. Just try it....set up a 16 team leage, roster fixed at 32 and start playing with the variables and watch the DV start adjusting up and down; use 1.5 payers for RB. 2.5 players for WR in a start RB or WR flex.
2.) Once you start following the adjustment of the DV factor, start messing with a standard trade, say the 1.05 in exchange for the 1.09 plus the 1.11. [Note: when you enter any picks into the calulator, be sure and do it in the pick number field (say pick #32), for instance, not in the pick/round designation (the 2.16 in a 16 teamer, for instance). Pick/round designations do not reset into the pick # eqivalent).]
3.) Start playing around with this deal. It is a real significant development. If you don't think so, try comparing a 32 roster/15 starter, flex OFF and DEF, including a Punter as the 15th starter with a fxed league that's 32 roster/14 starter that's 2-RB, 2-WR.
Let's play around with this deal....there's significant variation in the results.
And, as to the changes in values as league composition changes, I'm not bying into that one at all. Let's do a clinical survey, a Scientific Methods Approach and see what we come up with.
Scientific Method:
1.) For starters in a 16 team league, use 1 (QB, TE, PK), 2 (RB, WR, DL, DB) and 3-LB spread over a 32 man roster.....that's 14 starters over a 32 man roster.
2.) Next, take the same leage/roster dynamics and go to 1 (QB, TE, PK, RB), 2 (DL, DB) 3 (WR, LB)....same league, the RB/WR is flexed with the flex being subtacted fro the RB and added to the WR position. Actually, there's a flex on the DEF side of the ball here but it's remaining constant as the #3-LB but for this trial, we'll only flex the RB/WR position.
3.) Last, take 1(QB, TE, PK), 1.5 (RB), 2(DL, DB), 2.5 (WR), and 3 (LB)....same league but the flexed RB/WR position counted as a half a starter.
Now in each of those combo's Note the settings on a 1.05 trade for the 1.09 plus the 1.11 trade.
Make note of the different results:
1.) the DV for test #1, #2 and #3....we'll call those DV-1. DV-2 and DV-3
2.) the picks need to balance the trade in all three instances....we'll call those Pick #1, Pick #2 and Pick #3.
I'll do this and report back and we'll see if there's a major difference based in Draft Pick Trade Analysis based on the league settings. I'll also report back as to how those three examples stack up to the NFL 32 Team Draft Pick Trade Chart
Last edited by a moderator:
Toads
Footballguy
Actully, when you look at it, it's suspiciously like the NFL 32 teamer, just declined for a 16 team league.Wait...it is the NFL 32 Teamer declined for a 16 team league.
Fraudlent attempt to install a 16 teamer noted. This is nothing more than the standard attempt, to use the NFL trade chart for all purposes for all leagues, no matter how many teams are in your league.
It doesn't work that way....which is (one of) the purpose(s) of this thread.
Last edited by a moderator:
Toads
Footballguy
The calculator is specifically ID'ed as a Rookie Draft Pick Trade Chart. It has no application for players, player stat's projections, next year's picks....nothing but Rookie Draft pick exchange in the current period.
Toads
Footballguy
Results are as follows:Trial #1: DV of 4.152
Trade Result: 1.05 (1056 pts.) vs 1.09 + 1.11 (1078 pts.): variance ='s 2.12 %.
Needed to balance the trade: the 1.05 owner had to give an additional #44 pick....or, a very close
to even trade, 2% being close enough for the girl's we go with but the pick equivalent is a
significant pick. The #44 pick in a 16 teamer is the 3.12 pick.
Trial #2: DV of 3.907
Trade Result: 1.05 (1093 pts.) vs. 1.09 + 1.11 (1161 pts.): variance ='s 6.03%
Needed to balance the trade: the 1.05 owner had to give up an additional #32 pick, that's
the last pick in the second round (pick 2.16)....a significant pick in a 16 teamer.
Trial #3: DV of 4.079, about what we'd expect being right in the middle of the previous two values
[the average of the two is (4.152 + 3.907)/2 ='s 4.0295].
Trade result: 1.05 (1067 pts) vs. 1.09 + 1.11 (1102 pts.); variance ='s 3.29%
Needed to balance the trade: the 1.05 owner contributes the #39 pick, what amounts to
the 3.03 pick in a 16 teamer. (Note tha the pick is still centered in the range though. If it's
stastically "in the middle of the previous two ranges", it would be the #44 + #32)/2 = #38 pick.)
Trial #4: DV for the NFL 32 Teamer is unknown because we're using the 16 team/converted from the
32 team chart, we're using the NFL Trade Chart.
Trade Result: 1.05 (1,800 points) vs. 1.09 + 1.11 (2,650 pts.): variance ='s 47.2%
Needed to balance the trade: the 1.05 owner has to contribute 850 points or the #22 pick; a
2.06 pick in a 16 teamer.
Conclusion: those are significant variances....from the #3.12 pick to the #2.06 pick needed to balance our example is a wide range in a trade for Rooke Draft Choices! It's a variance of 45% [47.2 - (2.12)]='s 45.08% variance).
Good stuff for us Holy Grailers!
Last edited by a moderator:
Biabreakable
Footballguy
Sorry but this needs to be said.
So what?
If you think I am going to trade picks 1.09 and 1.11 AND pick 3.03 (or whatever) for pick 1.05.... especialy from the 2007 crop of rookies... just because this chart tells me that is what is fair.... your a few cards short of a deck and I am playing 52 pickup. Keep the jokers.
That would be equivilent of me trading you (based off FBGs consensus rankings):
1.09 Sidney Rice
1.11 Dwayne Jarrett
3.03 Johnie Lee Higgins
for
1.05 Robert Meachum
Thanks. But no thanks.
So what?
If you think I am going to trade picks 1.09 and 1.11 AND pick 3.03 (or whatever) for pick 1.05.... especialy from the 2007 crop of rookies... just because this chart tells me that is what is fair.... your a few cards short of a deck and I am playing 52 pickup. Keep the jokers.
That would be equivilent of me trading you (based off FBGs consensus rankings):
1.09 Sidney Rice
1.11 Dwayne Jarrett
3.03 Johnie Lee Higgins
for
1.05 Robert Meachum
Thanks. But no thanks.
GregR_2
Footballguy
Right, but my point was that I believe the majority of use these charts actually get is people using them for picks when the players were known... meaning they should have mocked up a draft and then decided from the players available if they wanted to do the trade, and not use a chart that doesn't have the real player's value built in.I don't think the novice or even median level FF owner realizes that, and I think it should be spelled out more clearly.
Edit to add: Haven't had time to through the stuff you did in detail and play with the chart more but am looking forward to it when I get some time I can concentrate on it for awhile.
Last edited by a moderator:
Toads
Footballguy
Same old stick....not reading, not comprehending, what's presented.The purpose of the exercise is the relative value of Rookie Draft Picks....not the relative value of some player that you choose to associate with those picks.
The exercise was to see what the effect the different make-ups, the different variables, had on the outcome. You're positioning of the 3.03 pick was on the wrong side of the board.... in the example(s), the 1.05 owner had to supplement the deal with the 3.03 (or, as you say: "what ever"), not receive it. Please re-read and advise if thats not what's said.
Pay attention, work in concepts, not cemented-in-place semi-realities. The pliable thoughts are what's happening here....the mix of the variables is what's being worked on.
If you can't formulate a system that works in creating a trade scenario, and do so with the assets available to both sides in a dynasty trade, you can't trade. This calculator deal does that.
So, pay attention and send me a trade offer, but don't attack if you can't follow the presentation. The chart will tell you what's fair, for any league scenario. You just have to be able to use it.
Last edited by a moderator:
Biabreakable
Footballguy
Okay I missed that in my brief reading as far as what side was being.. balanced? Perhaps I could pay better attention if I were not having to wade through so many superfluous statements about how great it is that math works in the calculator although there is still no basis for each picks value that has any substance in fact of what has really happened or will happen. And what players these picks ultimatly become.Correcting my misunderstanding here we end up with:
1.09 Sidney Rice
1.11 Dwayne Jarrett
for
1.05 Robert Meachum
3.03 Johnie Lee Higgins
Now the trade may look more fair with 2 players all of the same position on each side of the transaction.
However when applying a form of ADP to the results it does not bear out equal value imo when the actual or at least percieved value of Rice Jarrett and Meachum are all very close in the same tier yet Higgins is not.
Taking this a step further by looking at points scored results from 2006 draft (using one of my leagues here) this trade would have resulted in this:
1.05 RB Lendale White 32.75pts
3.03 TE Leonard Pope 18.1 pts
for
1.09 QB Jay Cutler 83.04 pts
1.11 Santonio Holmes 122.82 pts
So trading 50.85 pts for 205.86 pts
Another way to look at it would be to use beto's study and compare point values of those picks historicly to one another as a way of checking the results.
Until the calculations start factoring in these tangible results (and I do think they need to use a large sample size in doing so) I do not see what basis the calculations have for determining the value of each picks relative worth.
Here showing you that I am paying attention by responding specificly to each point. Working is the concepts without having a basis in reality.. not semi-reality but real tangible results imo is flawed and does not meet the goal intended. A trade value chart.
What is the point of testing the variance of the chart without testing the results? Now obviously people make bad picks and even reasonable picks bust without any previous indication of that result happening. However the collective analysis of ADP will without a doubt in my mind yeild better results more often than this more abstract point value assigned to each pick does. Taking that a step further and looking a large sample of points scored by ADP over somthing like a decade would provide a much better measurment of each picks value than the abstract point values assigned to each pick in this chart as well.
That would require a lot of data mining. It would be intruiging to see those results and compare the 2 systems although I am very certain that the method I am suggesting would be better.
Not that I really want such a thing to become readily available when I think that one of my edges over my compitition stems from having over 10 years of experience working from such results built in to my analysis of each rookie crop.
I have never had a problem forming trade proposals using my assessment of the players available with basis of past experience as a guide and I never needed or wanted a calculator to do so. Trading has much more with the communication of ideas and agreement than anything else. If my trade partner is using a calculator to check thier decisions I can work with or around that by using my own judgement.
There is no way I would follow this. Depending on each year and each possible combinations of trades I can see myself being on any side of the same trade. But my assesment of that will be based off of the specific players targeted with each pick and what my value/expectations are of them. Never because of a chart telling me what is value. Which honestly based off of trades done in the NFL such as the Eli or Vick deals this shows just how unbalanced and unfair the basis of this chart actualy is.Fair is a matter of agreement if it is a consideration at all. It is very rare for a trade to end up fair regardless of the perception of either party at the time.So, pay attention and send me a trade offer, but don't attack if you can't follow the presentation. The chart will tell you what's fair, for any league scenario. You just have to be able to use it.
:(
Last edited by a moderator:
Jeff Pasquino
Footballguy
Good stuff in here - glad you liked using the tool. Always looking for feedback or ways to make it better.Right now I think a flex option (or several) are needed, but you have found the easy workaround (Fractional players).
Nice job noticing the "pick to make even" - I love that feature.
Jeff Pasquino
Footballguy
That's pretty much the purpose of the calculator.It's a tool and a guide.Once you set a standard for evaluation, people adopt it as a reference. Is it 100% correct and perfect in every instance? Of course not. But it is a guide, and if everyone acknowledges it as such, you can treat it as such and even use it as part of your negotiations in trade.Use it, enjoy it, and we can make it even better if you find something to add.Magic Desert Toads said:
Toads
Footballguy
In order for a trade to happen, each side to the bargain has to hold onto some semblance of what "a fair trade" is, otherwise there would never be a trade.I really agree with your target statement, about targeting a range and trading up or trading down to get within that range in order to draft a specific player, all the while having a backup player in mind just in case my #1 option is off the table.Okay I missed that in my brief reading as far as what side was being.. balanced? Perhaps I could pay better attention if I were not having to wade through so many superfluous statements about how great it is that math works in the calculator although there is still no basis for each picks value that has any substance in fact of what has really happened or will happen. And what players these picks ultimatly become.Correcting my misunderstanding here we end up with:Same old stick....not reading, not comprehending, what's presented.
The purpose of the exercise is the relative value of Rookie Draft Picks....not the relative value of some player that you choose to associate with those picks.
The exercise was to see what the effect the different make-ups, the different variables, had on the outcome. You're positioning of the 3.03 pick was on the wrong side of the board.... in the example(s), the 1.05 owner had to supplement the deal with the 3.03 (or, as you say: "what ever"), not receive it. Please re-read and advise if thats not what's said.
1.09 Sidney Rice
1.11 Dwayne Jarrett
for
1.05 Robert Meachum
3.03 Johnie Lee Higgins
Now the trade may look more fair with 2 players all of the same position on each side of the transaction.
However when applying a form of ADP to the results it does not bear out equal value imo when the actual or at least percieved value of Rice Jarrett and Meachum are all very close in the same tier yet Higgins is not.
Taking this a step further by looking at points scored results from 2006 draft (using one of my leagues here) this trade would have resulted in this:
1.05 RB Lendale White 32.75pts
3.03 TE Leonard Pope 18.1 pts
for
1.09 QB Jay Cutler 83.04 pts
1.11 Santonio Holmes 122.82 pts
So trading 50.85 pts for 205.86 pts
Another way to look at it would be to use beto's study and compare point values of those picks historicly to one another as a way of checking the results.
Until the calculations start factoring in these tangible results (and I do think they need to use a large sample size in doing so) I do not see what basis the calculations have for determining the value of each picks relative worth.
Here showing you that I am paying attention by responding specificly to each point. Working is the concepts without having a basis in reality.. not semi-reality but real tangible results imo is flawed and does not meet the goal intended. A trade value chart.
What is the point of testing the variance of the chart without testing the results? Now obviously people make bad picks and even reasonable picks bust without any previous indication of that result happening. However the collective analysis of ADP will without a doubt in my mind yeild better results more often than this more abstract point value assigned to each pick does. Taking that a step further and looking a large sample of points scored by ADP over somthing like a decade would provide a much better measurment of each picks value than the abstract point values assigned to each pick in this chart as well.
That would require a lot of data mining. It would be intruiging to see those results and compare the 2 systems although I am very certain that the method I am suggesting would be better.
Not that I really want such a thing to become readily available when I think that one of my edges over my compitition stems from having over 10 years of experience working from such results built in to my analysis of each rookie crop.
I have never had a problem forming trade proposals using my assessment of the players available with basis of past experience as a guide and I never needed or wanted a calculator to do so. Trading has much more with the communication of ideas and agreement than anything else. If my trade partner is using a calculator to check thier decisions I can work with or around that by using my own judgement.There is no way I would follow this. Depending on each year and each possible combinations of trades I can see myself being on any side of the same trade. But my assesment of that will be based off of the specific players targeted with each pick and what my value/expectations are of them. Never because of a chart telling me what is value. Which honestly based off of trades done in the NFL such as the Eli or Vick deals this shows just how unbalanced and unfair the basis of this chart actualy is.Fair is a matter of agreement if it is a consideration at all. It is very rare for a trade to end up fair regardless of the perception of either party at the time.So, pay attention and send me a trade offer, but don't attack if you can't follow the presentation. The chart will tell you what's fair, for any league scenario. You just have to be able to use it.
The point here is both of the considerations you've brought up, targeting and defining a what a "fair trade" is are valid statements. One thing though, we are both omitting a trade partner and that trade partner has to be 1.) motivated and 2.) preceive that the trade has value, that the trade is fair.
That's where the calculator shines. It's a tool for trading and for proviing a comfort zone for any trade. As such, and if I felt that it was advantageous to do so, I would not hesitate to share it with any potential trade partner. I'm not looking to skin the opposite party, I'm just looking to accomplish my goal....to move my trade assets to get into a position to get to the range where I can draft my targeted player(s)....and to draft them as economically as possible.
Where this tool realy shines is when you are in a short time window. It allows for a cross check 1.) into the league settings, and 2.) a cross check to verify the settings that both patrties to the trade are referencing.... so that both arrive at the same place in the DV settings, both are referencing the same criteria and agreeing that it's accurate.
Once you get to that point, it's just a matter of playing one guys's draft holding vs. the other guys draft holdings to arrive at a deal that get's done....and there's really the key, getting it done and not getting close but getting there.
The goal is making the trade and this calulator deal get's that done.
Last edited by a moderator:
Biabreakable
Footballguy
Ah. I never was looking at it in that way. In the words of the big lebowski "my thinking about this has gotten really uptight" 
I have been looking at it only from the other way around such as if you suggested that because of this draft chart that I should trade Xpicks for Xpicks I would look at what players I am seeing and assess the deal the same way I have been for years. If I see the trade working vastly not in my favor even though the trade chart is saying that the trade is fair this would cause me to either laugh in your face or else chuckle under my breath depending on my mood at the time.
I do make many proposals that involve draft picks. More often than not there are veteran players involved as well so this is a dynamic that is not covered by this chart. That being said I have figured out some things about draft pick value just from experience that makes it fairly easy for me to formulate offers that involve them. I rarely know what to expect my trade partners reaction to be when I make these offers or if it is the draft picks or somthing else that they liked or did not like about the deal. However over time I do start learning owners tendencies for how they value picks. No two owners are the same as some value picks more than others in general and some see bigger gaps in value from 1st round picks to later round picks than others.
Being a pretty honest person it would be hard for me to sell a pick value chart to someone else because I do not believe in it. But I do see your point about it possibly being a tool that could lead a horse to water. That is somthing worth considering instead of me explaining so much about my own specific reasoning to the owner (which gives them a lot of information and information about me).
Thanks for pointing this out DMT.
I have been looking at it only from the other way around such as if you suggested that because of this draft chart that I should trade Xpicks for Xpicks I would look at what players I am seeing and assess the deal the same way I have been for years. If I see the trade working vastly not in my favor even though the trade chart is saying that the trade is fair this would cause me to either laugh in your face or else chuckle under my breath depending on my mood at the time.
I do make many proposals that involve draft picks. More often than not there are veteran players involved as well so this is a dynamic that is not covered by this chart. That being said I have figured out some things about draft pick value just from experience that makes it fairly easy for me to formulate offers that involve them. I rarely know what to expect my trade partners reaction to be when I make these offers or if it is the draft picks or somthing else that they liked or did not like about the deal. However over time I do start learning owners tendencies for how they value picks. No two owners are the same as some value picks more than others in general and some see bigger gaps in value from 1st round picks to later round picks than others.
Being a pretty honest person it would be hard for me to sell a pick value chart to someone else because I do not believe in it. But I do see your point about it possibly being a tool that could lead a horse to water. That is somthing worth considering instead of me explaining so much about my own specific reasoning to the owner (which gives them a lot of information and information about me).
Thanks for pointing this out DMT.
GregR_2
Footballguy
Well, I'm not as easily convinced as Biabreakable. They broke you, Bia!
To be serious though, which output of these two do you expect a pick value calculator to give you? And which is most important to you?
1. A relative measure of the actual value your team will get when the pick is converted into a player.
2. A relative measure of what another owner may consider an equitable trade, without necessarily any semblance on the actual value your team will get when the pick is converted into a player.
I hope you answered #1 for both questions, though if you could have both that would be the ideal world. You need to know #1 because that's the only one that tells you if you should have an expectation that your team improves, which ultimately is the question you need to answer to decide to make the trade or not. #2 might help you in negotiations, but it doesn't tell you the real value of the pick.Well guess what? The NFL draft pick value chart is #2. It is taking a bunch of trades from NFL history and fitting a curve to it and saying "this is the value of the pick". In actuality it is the PERCEIVED value of the pick. It is the perception of the NFL GM's (and hands on owners like Jerry Jones and Al Davis) of the value of the pick, but that perception may not match what history tells us the picks normally turn out to be worth.
Two professors, Massey and Thaler, published a paper on the actual value of NFL picks vs the perceived value. Note this part of the abstract:
Jeff, the calculator you produced could best be summarized as showing "these are how fantasy owners tend to value picks, but not necessarily how YOU should value them". Though, I have one additional beef that if that is what you want to produce, you shouldn't base it off of NFL GM's perceptions of NFL draft picks, because those are not necessarily the same as fantasy owner's perceptions of fantasy draft picks. If you want to produce that chart, then find a bunch of dynasty leagues, group them by similar league parameters, and do some curve fits to fit the actual trades that have been made. Or use some other similar method, but at least use actual fantasy trades as the input since the NFL and fantasy aren't necessarily the same in this respect.
The second point I'd have is that if indeed that is what you're going to make, you should be clear about what it is showing. Because I think I can safely say 99% of FF owners when told something is a draft pick value chart, believe it reflects the value of the pick to their team as it relates to their team's scoring bottom line.
What myself and Bia and some others have said is that to produce the second chart, the one that tells you the value to your team, you probably need to work off of historical data for fantasy player success, similar to how Massey and Thaler worked off of historical NFL player success taken with each pick to find the value of the pick. You need to find a way to roll up all the factors you think are important into a quantified number that is the expected value of taking a player with the pick and having him on your team. You can't predict it perfectly and no one expects it to be. If they can predict the players perfectly they shouldn't be using the chart anyway, they should be using the players like Bia did when he came into the thread. But to get to the best answer you can for when you can't directly predict the players, you need to use some historical results that apply to the fantasy situation.
Let me leave this with another quote Massey and Thaler again, about why perceptions are often out of whack with actual value:
To be serious though, which output of these two do you expect a pick value calculator to give you? And which is most important to you?
1. A relative measure of the actual value your team will get when the pick is converted into a player.
2. A relative measure of what another owner may consider an equitable trade, without necessarily any semblance on the actual value your team will get when the pick is converted into a player.
I hope you answered #1 for both questions, though if you could have both that would be the ideal world. You need to know #1 because that's the only one that tells you if you should have an expectation that your team improves, which ultimately is the question you need to answer to decide to make the trade or not. #2 might help you in negotiations, but it doesn't tell you the real value of the pick.Well guess what? The NFL draft pick value chart is #2. It is taking a bunch of trades from NFL history and fitting a curve to it and saying "this is the value of the pick". In actuality it is the PERCEIVED value of the pick. It is the perception of the NFL GM's (and hands on owners like Jerry Jones and Al Davis) of the value of the pick, but that perception may not match what history tells us the picks normally turn out to be worth.
Two professors, Massey and Thaler, published a paper on the actual value of NFL picks vs the perceived value. Note this part of the abstract:
From their results, the most valuable picks in the NFL are at the end of the first round. The expectation for those picks based on actual results is that you get a slightly less talented/successful player but at significantly lower costs (i.e. salary). In fact they found that going by actual value to the team, the worst pick in the first round was the first pick. For the meat of their conclusions, go start in the middle of page 34 and read through page 36.Anyway, back to the fantasy topic. This is half of what I've been complaining about -- Biabreakable was too before he caved! -- that the chart that people want to see is the chart of the actual value. Worse, that's what they THINK they are seeing. Judging from Biabreakable's reaction when he came into the thread, I think that's what he was expecting to see. It's the result of the professor study, not of the NFL draft chart, that is the piece of data you most need, so expect that's what the chart is giving you. They want answered: "Does this trade most likely help my team, or hurt it, or is it truly even?"
Jeff, the calculator you produced could best be summarized as showing "these are how fantasy owners tend to value picks, but not necessarily how YOU should value them". Though, I have one additional beef that if that is what you want to produce, you shouldn't base it off of NFL GM's perceptions of NFL draft picks, because those are not necessarily the same as fantasy owner's perceptions of fantasy draft picks. If you want to produce that chart, then find a bunch of dynasty leagues, group them by similar league parameters, and do some curve fits to fit the actual trades that have been made. Or use some other similar method, but at least use actual fantasy trades as the input since the NFL and fantasy aren't necessarily the same in this respect.
The second point I'd have is that if indeed that is what you're going to make, you should be clear about what it is showing. Because I think I can safely say 99% of FF owners when told something is a draft pick value chart, believe it reflects the value of the pick to their team as it relates to their team's scoring bottom line.
What myself and Bia and some others have said is that to produce the second chart, the one that tells you the value to your team, you probably need to work off of historical data for fantasy player success, similar to how Massey and Thaler worked off of historical NFL player success taken with each pick to find the value of the pick. You need to find a way to roll up all the factors you think are important into a quantified number that is the expected value of taking a player with the pick and having him on your team. You can't predict it perfectly and no one expects it to be. If they can predict the players perfectly they shouldn't be using the chart anyway, they should be using the players like Bia did when he came into the thread. But to get to the best answer you can for when you can't directly predict the players, you need to use some historical results that apply to the fantasy situation.
Let me leave this with another quote Massey and Thaler again, about why perceptions are often out of whack with actual value:
Jeff Pasquino
Footballguy
Greg, How many different ways can I say this before you agree that I've said it already.Well, I'm not as easily convinced as Biabreakable. They broke you, Bia!
To be serious though, which output of these two do you expect a pick value calculator to give you? And which is most important to you?
1. A relative measure of the actual value your team will get when the pick is converted into a player.
2. A relative measure of what another owner may consider an equitable trade, without necessarily any semblance on the actual value your team will get when the pick is converted into a player.
I hope you answered #1 for both questions, though if you could have both that would be the ideal world. You need to know #1 because that's the only one that tells you if you should have an expectation that your team improves, which ultimately is the question you need to answer to decide to make the trade or not. #2 might help you in negotiations, but it doesn't tell you the real value of the pick.Well guess what? The NFL draft pick value chart is #2. It is taking a bunch of trades from NFL history and fitting a curve to it and saying "this is the value of the pick". In actuality it is the PERCEIVED value of the pick. It is the perception of the NFL GM's (and hands on owners like Jerry Jones and Al Davis) of the value of the pick, but that perception may not match what history tells us the picks normally turn out to be worth.
Two professors, Massey and Thaler, published a paper on the actual value of NFL picks vs the perceived value. Note this part of the abstract:
From their results, the most valuable picks in the NFL are at the end of the first round. The expectation for those picks based on actual results is that you get a slightly less talented/successful player but at significantly lower costs (i.e. salary). In fact they found that going by actual value to the team, the worst pick in the first round was the first pick. For the meat of their conclusions, go start in the middle of page 34 and read through page 36.Anyway, back to the fantasy topic. This is half of what I've been complaining about -- Biabreakable was too before he caved! -- that the chart that people want to see is the chart of the actual value. Worse, that's what they THINK they are seeing. Judging from Biabreakable's reaction when he came into the thread, I think that's what he was expecting to see. It's the result of the professor study, not of the NFL draft chart, that is the piece of data you most need, so expect that's what the chart is giving you. They want answered: "Does this trade most likely help my team, or hurt it, or is it truly even?"
Jeff, the calculator you produced could best be summarized as showing "these are how fantasy owners tend to value picks, but not necessarily how YOU should value them". Though, I have one additional beef that if that is what you want to produce, you shouldn't base it off of NFL GM's perceptions of NFL draft picks, because those are not necessarily the same as fantasy owner's perceptions of fantasy draft picks. If you want to produce that chart, then find a bunch of dynasty leagues, group them by similar league parameters, and do some curve fits to fit the actual trades that have been made. Or use some other similar method, but at least use actual fantasy trades as the input since the NFL and fantasy aren't necessarily the same in this respect.
The second point I'd have is that if indeed that is what you're going to make, you should be clear about what it is showing. Because I think I can safely say 99% of FF owners when told something is a draft pick value chart, believe it reflects the value of the pick to their team as it relates to their team's scoring bottom line.
What myself and Bia and some others have said is that to produce the second chart, the one that tells you the value to your team, you probably need to work off of historical data for fantasy player success, similar to how Massey and Thaler worked off of historical NFL player success taken with each pick to find the value of the pick. You need to find a way to roll up all the factors you think are important into a quantified number that is the expected value of taking a player with the pick and having him on your team. You can't predict it perfectly and no one expects it to be. If they can predict the players perfectly they shouldn't be using the chart anyway, they should be using the players like Bia did when he came into the thread. But to get to the best answer you can for when you can't directly predict the players, you need to use some historical results that apply to the fantasy situation.
Let me leave this with another quote Massey and Thaler again, about why perceptions are often out of whack with actual value:
The calculator I made for the Rookie Draft Picks is a TOOL. It is NOT an absolute value. That's not possible for everyone. You can't do that. It is a guide - a rough order of value.
The calculator was made by using several NFL parameters about players coupled with the FBG draft calculator (Dodds' values). I'm sure you won't like hearing that either and want 20 years of FF historical trade data to validate Dodds' numbers - be my guest.
I've asked repeatedly for people to exercise the calculator on their Dynasty leagues - and you know what? It's pretty darn accurate.
I do not use the NFL draft chart in the calculator at all. I use it as a frame of reference in the article about the calculator and as a reference point - nothing more.
The entire point of the calculator is to take YOUR LEAGUE SETTINGS and to adjust the parameter known as the Dynasty Factor. That is derived based on NFL data - not the NFL chart - but on NFL player data.
Now, you cannot make a tool that does every league and every possible trade. They don't make that Swiss Army Knife.
What this calculator does is take models of trades (and yes I've looked at MANY) and tests the trades against them. I looked at what 1.01 was worth vs. say 1.05 and 1.07 in a deal. I've seen deals that match these parameters. I invite you to go ahead and test the calculator against any historical Dynasty league you wish.
I've tried on several occasions here to converse with you on this, but it seems that there is no pleasing you. You state that you want historical trade data to get more realistic answers, yet you obscure the fact that there could be bad trades in the data. What if the owner had 8 picks in the first 2 rounds and couldn't roster them all, but he wanted to move up? Of course he might overpay and do a 2 for 1 that most others wouldn't, but it made sense to him and his team. Where's that accounted for in your analysis model?
Even if I had those results and those numbers, you still wouldn't be happy as you would then want to put names and faces with the picks and then decide on value. I've said, time and again, you can't have a calculator that changes every year and in every crop of new rookies. A Reggie Bush year is different than a Calvin Johnson year. Some drafts are rich at RB, some at WR. If your leagues treat them differently, trades won't align well to the calculator at least one of those two times.
I've come to the conclusion that you will argue most any point against this calculator, so the discussion is pointless. You've clearly made up your mind that you don't like the tool, so feel free not to use it.
I hope others choose differently and let me know what they think about their results.
Last edited by a moderator:
GregR_2
Footballguy
No, it isn't that I have something against the calculator I just don't agree with the methodology I see in it. If I misunderstood your use of the NFL draft pick chart from going over the article, definite apologies, my bad.But I just don't get this line of thinking, and yes, will argue against it.
You say this like it is a bad thing. Isn't getting at the ACTUAL RIGHT answer, or as close as possible, the goal of doing something like this? Shouldn't we strive to make a calculator that in the months leading up to the Reggie Bush year, reflect the Reggie Bush year and don't give the same answer as the following year when as you point out, things have changed?The same goes for the redraft calculator. Do you really think the vast majority of use it will see between now and the start of the season is not for players that we have projections for already that can be built into this year's version?I can tell you got upset if I misunderstood how you used the NFL chart, for which I'll openly apologize again. But I don't understand what the motivation is to not want to include projections for the year after the time they start existing. Short of having total precognition that's about the best set of input data we could hope for, isn't it? Why would we not want to include that? Would we not have gotten better results in the preseason before Peyton's record-setting year if the values for that year's redraft included the fact that history suggested we could expect around a 10% surge in QB scoring due to the rule change?It isn't about liking it or not liking it, it's about using the best information to feed into it that we can.
