VBD Baselines with Flex Rosters? (1 Viewer)

[Seven Costanza]

Hoping you guys have some thoughts on this question.....the requirements for my starting lineup in my league are 1 RB, 3 WR (no discrepency is made between WR and TE in my league, we do not have to start a TE but can use a TE to fill a WR slot), 1 Flex (RB/WR).....

Would you guys still keep the RB and WR positions seperate and use a seperate baseline within each position to determine the X values? Or would you combine the 2 positions into one grouping and add the baselines together to determine the X values?

For example.....say I project 60 RBs and use RB 30 as my baseline to determine the X values for RBs and project 110 WRs and use WR 55 as my baseline to determine the X values for WRs. Should I keep the 2 groupings seperate to determine X values or combine all 170 players into one grouping amnd use 85 as my baseline?

My gut says keep the groupings seperate and when it is time to draft my flex player just pick the remaining RB or WR with the higher X value??

Thanks!

 

[GregR]

Keep the two positions separate. Figure out what the baseline would be if there was no flex. Then figure out after RB and WR starters, how many of RB and WR should start in the flex. Then increase your baseline for each position by that result to build in the players that should go to flex.

 

[Seven Costanza]

Greg could you clarify what you mean...?..

Thanks!

 

[GregR]

Greg could you clarify what you mean...?..

Thanks!

Let's say you just are going straight up last starter baselines. 2 RB, 3 WR, 12 teams would be RB24 and WR36. Now you add a RB/WR flex. Look at the next 12 highest scoring players from RB and WR. If 7 are RBs and 5 are WRs, then your RB baseline is now 24+7=RB31 and your WR baseline is 36+5 = WR41.

And with TE counting as a WR, just combined them in with WRs, they don't get their own baseline or be treated as their own position.

I'd generally go deeper than last starter to account for things like RB scarcity making 3rd RBs worth a little more... but you can set it wherever you think right for non-flex, then add in the extra spots for the flex

 

[Seven Costanza]

Awesome! That helped clear it up a ton! I go a little deeper with my baselines as well...I analyzed the top 100 picks from our last 2 drafts and have decided on a RB baseline of 28 and a combined WR/TE baseline of 53 (it is PPR, so the WR tend to go a little heavier than the RBs).

So once I identify the top 28 RBs and top 53 WR/TE ....I will look for the next 12 hihhest X values regardless of position and tease my baseline player in each category down? So using your example of 7 RBs and 5 WRs...my new RB baseline player will be RB 35 and my new WR baseline player will be 58???

Thanks for all your help!! I am using VBD for the first time and our league scoring/roster reqs dont allow for the basic VVD approach.

 

[CalBear]

Hoping you guys have some thoughts on this question.....the requirements for my starting lineup in my league are 1 RB, 3 WR (no discrepency is made between WR and TE in my league, we do not have to start a TE but can use a TE to fill a WR slot), 1 Flex (RB/WR).....

Would you guys still keep the RB and WR positions seperate and use a seperate baseline within each position to determine the X values? Or would you combine the 2 positions into one grouping and add the baselines together to determine the X values?

For example.....say I project 60 RBs and use RB 30 as my baseline to determine the X values for RBs and project 110 WRs and use WR 55 as my baseline to determine the X values for WRs. Should I keep the 2 groupings seperate to determine X values or combine all 170 players into one grouping amnd use 85 as my baseline?

My gut says keep the groupings seperate and when it is time to draft my flex player just pick the remaining RB or WR with the higher X value??

Thanks!

With just one flex spot, it's not worth combining the positions; your first three or four picks there will be more important than the fifth one, and your RB1 should be compared to other RBs.

One thing you can do rather than using number-of-starters baselines is to choose a point threshhold instead. That can make it simpler.

 

[TheCommish]

I go to our league draft pages each season to find out how many players at each position were drafted through 10 rounds. I then find the average for each position (usually going back no more than 3 years). Those are my baselines.

For example, through 10 rounds in 2010, 2011 and 2012, my league had drafted 42, 38 and 41 WRs. My 2013 baseline for the WR position is 40. If you do this, make sure you haven't had any major rule changes, or else the draft numbers could be skewed.

 

[Seven Costanza]

That is exactly what I have done commish....I used the first 100 picks, instead of 10 rounds.....I came up with a baseline for my RBs of 28 and 53 for my WR/TE.

 

[hotboyz]

That's what I do but since my draft is 14 Rds I use the 7th rd and funny thing the last 2 seasons the exact same amount of every position has been drafted. 12 QB 35 RB 30 WR and 7 TE so that will be my baseline

 

[CalBear]

I go to our league draft pages each season to find out how many players at each position were drafted through 10 rounds. I then find the average for each position (usually going back no more than 3 years). Those are my baselines.

For example, through 10 rounds in 2010, 2011 and 2012, my league had drafted 42, 38 and 41 WRs. My 2013 baseline for the WR position is 40. If you do this, make sure you haven't had any major rule changes, or else the draft numbers could be skewed.

I think setting your baselines based on your league's drafting tendencies is exactly the wrong thing to do. Unless you believe your league drafts perfectly. The goal of VBD is to identify value among the players on the board; if your league consistently over-values one position, setting your baseline based on your league's tendencies will also over-value that position.

 

[GregR]

I go to our league draft pages each season to find out how many players at each position were drafted through 10 rounds. I then find the average for each position (usually going back no more than 3 years). Those are my baselines.

For example, through 10 rounds in 2010, 2011 and 2012, my league had drafted 42, 38 and 41 WRs. My 2013 baseline for the WR position is 40. If you do this, make sure you haven't had any major rule changes, or else the draft numbers could be skewed.

I think setting your baselines based on your league's drafting tendencies is exactly the wrong thing to do. Unless you believe your league drafts perfectly. The goal of VBD is to identify value among the players on the board; if your league consistently over-values one position, setting your baseline based on your league's tendencies will also over-value that position.

I agree, and very well stated what the issue is with doing that.

The tendencies of how a league drafts players are indeed useful. But I don't think using them in the baseline is the place to apply them. Better to work with a baseline that highlights the value in the pool of players regardless of how the league drafts. Then take that result and compare it to how the league drafts to see where good and bad values exist.

 

[Seven Costanza]

Okay Greg and CalBear....great points! So what do you recommend? I looked at Joe Bryant's Principles of VBD drafting and he gives a few techniques. He mentions last starter as your baseline but says that isnt necessarily the best way. Then he lists a few other approaches, one was to review the first 100 picks of your draft...but I do see the potential flaw in that....if people draft stupidly, it would impact my baselines!

So what is a rookie VBDer to do??

 

[hotboyz]

I go to our league draft pages each season to find out how many players at each position were drafted through 10 rounds. I then find the average for each position (usually going back no more than 3 years). Those are my baselines.

For example, through 10 rounds in 2010, 2011 and 2012, my league had drafted 42, 38 and 41 WRs. My 2013 baseline for the WR position is 40. If you do this, make sure you haven't had any major rule changes, or else the draft numbers could be skewed.

I think setting your baselines based on your league's drafting tendencies is exactly the wrong thing to do. Unless you believe your league drafts perfectly. The goal of VBD is to identify value among the players on the board; if your league consistently over-values one position, setting your baseline based on your league's tendencies will also over-value that position.

I agree, and very well stated what the issue is with doing that.

The tendencies of how a league drafts players are indeed useful. But I don't think using them in the baseline is the place to apply them. Better to work with a baseline that highlights the value in the pool of players regardless of how the league drafts. Then take that result and compare it to how the league drafts to see where good and bad values exist.

I'm not quite sure I understand what you are saying here. How can you set baselines regardless of your league tendencies, especially one as difinitive as mine. So how would u determine baseline? I use the halfway point of my draft my draft is 14 rds so when I set my baseline at 7th rd I believe most teams starting lineups are set excluding kicker and defense 2 yrs in a row this league like clockwork drafts 12 Qb 35Rb 30Wr and 7 TE

 

[CalBear]

Okay Greg and CalBear....great points! So what do you recommend? I looked at Joe Bryant's Principles of VBD drafting and he gives a few techniques. He mentions last starter as your baseline but says that isnt necessarily the best way. Then he lists a few other approaches, one was to review the first 100 picks of your draft...but I do see the potential flaw in that....if people draft stupidly, it would impact my baselines!

So what is a rookie VBDer to do??

I think the best thing to do is set the baseline at "replacement-level player"; a level of production that you think you can replace at very little cost. Last-starter is a decent hand-wave at that, but differs for different positions; RB24 is harder to replace than TE12. I would look not at the drafting patterns, but the scoring patterns in your league. Eyeball the top 50 at each position in your league; at what point does it start to look like "a bunch of guys"? That's your point baseline. I prefer scoring-based baselines to position-based baselines, because position-based baselines are dependent on projections for low-ranked players, which tend to be fairly variable from projection set to projection set (moreso than the projections for high-ranked players).

 

[CalBear]

I go to our league draft pages each season to find out how many players at each position were drafted through 10 rounds. I then find the average for each position (usually going back no more than 3 years). Those are my baselines.

For example, through 10 rounds in 2010, 2011 and 2012, my league had drafted 42, 38 and 41 WRs. My 2013 baseline for the WR position is 40. If you do this, make sure you haven't had any major rule changes, or else the draft numbers could be skewed.

I think setting your baselines based on your league's drafting tendencies is exactly the wrong thing to do. Unless you believe your league drafts perfectly. The goal of VBD is to identify value among the players on the board; if your league consistently over-values one position, setting your baseline based on your league's tendencies will also over-value that position.

I agree, and very well stated what the issue is with doing that.

The tendencies of how a league drafts players are indeed useful. But I don't think using them in the baseline is the place to apply them. Better to work with a baseline that highlights the value in the pool of players regardless of how the league drafts. Then take that result and compare it to how the league drafts to see where good and bad values exist.

I'm not quite sure I understand what you are saying here. How can you set baselines regardless of your league tendencies, especially one as difinitive as mine. So how would u determine baseline? I use the halfway point of my draft my draft is 14 rds so when I set my baseline at 7th rd I believe most teams starting lineups are set excluding kicker and defense 2 yrs in a row this league like clockwork drafts 12 Qb 35Rb 30Wr and 7 TE

Here's a scenario to illustrate the point. Let's say your league is insane and consistently drafts nothing but RBs for the first seven rounds. If you set your baseline based on how many players are typically drafted by the seventh round, only RBs will show any value; everything else will be VBD 0. Obviously, your league isn't that crazy, but almost any league has inefficiencies.

Your league tendencies equally value QB12, RB35, WR30, and TE7. Do they really have equal value? That question can't be answered by looking at drafting tendencies. When you set VBD baselines, you are saying, "This is a set of players with equivalent value." You need to know that to compare the value of RB5 vs. WR1; how many extra points will they score relative to a set of baseline players of equivalent value? If your baselines aren't set to players of equivalent value (from a fantasy scoring perspective, not a league draft perspective), the VBD rankings for high-ranked players will be off.

 

[moleculo]

Okay Greg and CalBear....great points! So what do you recommend? I looked at Joe Bryant's Principles of VBD drafting and he gives a few techniques. He mentions last starter as your baseline but says that isnt necessarily the best way. Then he lists a few other approaches, one was to review the first 100 picks of your draft...but I do see the potential flaw in that....if people draft stupidly, it would impact my baselines!

So what is a rookie VBDer to do??

I think the best thing to do is set the baseline at "replacement-level player"; a level of production that you think you can replace at very little cost. Last-starter is a decent hand-wave at that, but differs for different positions; RB24 is harder to replace than TE12. I would look not at the drafting patterns, but the scoring patterns in your league. Eyeball the top 50 at each position in your league; at what point does it start to look like "a bunch of guys"? That's your point baseline. I prefer scoring-based baselines to position-based baselines, because position-based baselines are dependent on projections for low-ranked players, which tend to be fairly variable from projection set to projection set (moreso than the projections for high-ranked players).

I was just gonna post about "replacement level player". My current thought is that replacement level = who you can replace player X with after the draft - and that, obviously, is the highest rated undrafted player.

In my 12 team, 20 man roster league, there were 30 QB's drafted last year. Therefore, the baseline for QB is QB30 or so. When I put it all together, this does seem to overvalue QB's, but IMO that makes sense in that QB's are usually the toughest to find on the WW, especially early.

I understand what was said above about not trying to follow the herd - if your league overvalues a position, you don't want to follow suit. I don't think I agree with that. What you need to do is acknowledge league trends so you can get ahead of them. If your league overdrafts RB's, it becomes even more important to get your RB early because there won't be any late.

I am still vetting this really low baseline, not sure I completely buy in just yet. It's much lower than anything I have done before. Still trying to understand how it works. One thing which is nice - in my static VBD sheet, the really low baseline means that there are very few negative numbers, which is nice.

ETA: one other nice thing about the low VBD is that it has more meaning deeper into the draft. Drafting is easy at the top - it's the mid to late ranges where decisions get difficult and that's where traditional VBD breaks down.

 

[moleculo]

I go to our league draft pages each season to find out how many players at each position were drafted through 10 rounds. I then find the average for each position (usually going back no more than 3 years). Those are my baselines.

For example, through 10 rounds in 2010, 2011 and 2012, my league had drafted 42, 38 and 41 WRs. My 2013 baseline for the WR position is 40. If you do this, make sure you haven't had any major rule changes, or else the draft numbers could be skewed.

I think setting your baselines based on your league's drafting tendencies is exactly the wrong thing to do. Unless you believe your league drafts perfectly. The goal of VBD is to identify value among the players on the board; if your league consistently over-values one position, setting your baseline based on your league's tendencies will also over-value that position.

I agree, and very well stated what the issue is with doing that.

The tendencies of how a league drafts players are indeed useful. But I don't think using them in the baseline is the place to apply them. Better to work with a baseline that highlights the value in the pool of players regardless of how the league drafts. Then take that result and compare it to how the league drafts to see where good and bad values exist.

I'm not quite sure I understand what you are saying here. How can you set baselines regardless of your league tendencies, especially one as difinitive as mine. So how would u determine baseline? I use the halfway point of my draft my draft is 14 rds so when I set my baseline at 7th rd I believe most teams starting lineups are set excluding kicker and defense 2 yrs in a row this league like clockwork drafts 12 Qb 35Rb 30Wr and 7 TE

Here's a scenario to illustrate the point. Let's say your league is insane and consistently drafts nothing but RBs for the first seven rounds. If you set your baseline based on how many players are typically drafted by the seventh round, only RBs will show any value; everything else will be VBD 0. Obviously, your league isn't that crazy, but almost any league has inefficiencies.

Your league tendencies equally value QB12, RB35, WR30, and TE7. Do they really have equal value? That question can't be answered by looking at drafting tendencies. When you set VBD baselines, you are saying, "This is a set of players with equivalent value." You need to know that to compare the value of RB5 vs. WR1; how many extra points will they score relative to a set of baseline players of equivalent value? If your baselines aren't set to players of equivalent value (from a fantasy scoring perspective, not a league draft perspective), the VBD rankings for high-ranked players will be off.

if you know in advance that it will be RB heavy early, it becomes even more imperative to get RB's early - WR's, QB's and TE will be available later. In this situation, you want RB's valued highly, IMO.

 

[CalBear]

if you know in advance that it will be RB heavy early, it becomes even more imperative to get RB's early - WR's, QB's and TE will be available later. In this situation, you want RB's valued highly, IMO.

I totally disagree. Let's say you're in a kicker-crazy league; 24 kickers will go in the first five rounds. Do you need to take a kicker early to make sure you get one? No, because you can get better production from other positions in every round. You can take K32 at the end of the draft, or pick up one on waivers, and kick you league's butt because you got additional production at other positions in every round.

 

[Seven Costanza]

I guess I could run it both ways.....use a draft based baseline and look at it from a scoring based baseline and compare the two and see what sort of inconsistencies I get.

 

[CalBear]

I guess I could run it both ways.....use a draft based baseline and look at it from a scoring based baseline and compare the two and see what sort of inconsistencies I get.

That would be a good exercise which could suggest inefficiencies in your league that you could take advantage of.

 

[moleculo]

if you know in advance that it will be RB heavy early, it becomes even more imperative to get RB's early - WR's, QB's and TE will be available later. In this situation, you want RB's valued highly, IMO.

I totally disagree. Let's say you're in a kicker-crazy league; 24 kickers will go in the first five rounds. Do you need to take a kicker early to make sure you get one? No, because you can get better production from other positions in every round. You can take K32 at the end of the draft, or pick up one on waivers, and kick you league's butt because you got additional production at other positions in every round.

The above statement was not an absolute rule, it's a rebuttal that you should ignore league trends. I pose you need to know league trends to properly set up a static VBD baseline.

let's check the math. The difference between K1 and K25 is roughly 25 points. Per my projections, that's about the same as the difference between RB1 and RB5. You would be insane to want to get ahead of the kicker trend.

Here's a hypothetical: 12 team league. Start 1 QB, 2 RB, 3 WR, 1 TE. your particular VBD ranking is pretty traditional - RB's, WR, and QB all sprinkled together nicely. Lets assume all of your league-mates will be picking RB exclusively in rounds 3 and 4. You are drafting in, say, the 9th spot.

let's assume the first two rounds go mostly by ADP - roughly 17 RB's, 5 WRs, 1 QB, 1 TE. Now, we know that after 4 rounds, there will be 41 RB's, 6 WR's, 1 QB, and 1 TE.

What should you do in the first two rounds? Per ADP and (coincidentally) your VBD sheet, you take calvin Johnson at 1.09 and AJ Green 2.03. Drafting at 3.09, the best available RB will be Ahmand Bradshaw (RB26). That sucks, so instead you buck the trend and take Brees. At 4.03, the top RB will be Shane Vareen (RB31). If you are true to your VBD board, you take Demaryius Thomas (WR6) instead. By the time you get to 5.09, the top RB on the board is Vick Ballard (RB42), and that is who you will be forced to take. at 6.03, you are drafting Isiah Pead as a starting RB.

Had your VBD accounted for this trend, you would have taken Richardson (RB10) at 1.09 and Ridley at 2.04. In the middle of the crazy RB run, you still would have been able to snag Brees ( 3.09), Demaryius Thomas (4.03), and followed up with Harvin at 5.09, and Jordy Nelson at 6.03.

Getting caught by the crazy run:

1.09: Calvin Johnson

2.03: AJ Green

3.09: Drew Brees

4.03: Demaryius Thomas

5.09: Vick Ballard

6.03: Isaiah Pead

being ahead of the run:

1.09: Trent Richardson

2.03: Stevan Ridley

3.09ew Brees

4.03: Demaryius Thomas

5.09: Percy Harvin

6.03: Jordy Nelson

I think the second team is far superior. The way you get that is with advance knowledge of how your league will draft. If your VBD static baseline were set up to capture this, RB value would be increased and this would have lead you to take Richardson+Ridley instead of the more sexy Calvin/AJ.

 

[TheCommish]

I go to our league draft pages each season to find out how many players at each position were drafted through 10 rounds. I then find the average for each position (usually going back no more than 3 years). Those are my baselines.

For example, through 10 rounds in 2010, 2011 and 2012, my league had drafted 42, 38 and 41 WRs. My 2013 baseline for the WR position is 40. If you do this, make sure you haven't had any major rule changes, or else the draft numbers could be skewed.

I think setting your baselines based on your league's drafting tendencies is exactly the wrong thing to do. Unless you believe your league drafts perfectly. The goal of VBD is to identify value among the players on the board; if your league consistently over-values one position, setting your baseline based on your league's tendencies will also over-value that position.

I know exactly what you are saying, but if I were to draft using the VBD Excel App spreadsheet or the Draft Dominator, I would end up with too many RBs too early. In most leagues, having a ton of RBs is fine, but receivers are highly valued in our scoring system, and I sometimes end up loaded at RB and scrambling all season long at WR. Personally, I think it's tougher to find a viable WR than a RB on the waiver wire.

I think that this web site's drafting systems place far too much value on RBs. Which is why I have created my own spreadsheet that links to the VBD Excel App for certain things that I deem relevant. And my league's historical drafts are remarkably consistent (since 2007, we have had 39, 40, 40, 42, 38 and 41 WRs drafted through 10 rounds). So, when I use historical numbers as my baselines, I have a very good idea about how many players at each position will be drafted in the next couple of rounds. I can use that to my advantage when anticipating a run at a certain position. For example, if we just finished the 6th round and 30 WRs are off the board (5-year average for my league), I can refer to my spreadsheet and see that about 5 WRs have historically been drafted in the 7th and 8th rounds combined. If I happen to value the next 8 WRs on my board equally, I can feel relatively safe that I can draft other positions in the 6th and 7th rounds and still get a WR that I like in the 9th round.

All positions are still tiered by value, but this system has more relevance to me and my league's drafting tendencies.

EDITED TO ADD: My league is an IDP league, which is why so few WRs are drafted in the 7th and 8th rounds. The defensive players start going heavy by these rounds.

 

[GregR]

Okay Greg and CalBear....great points! So what do you recommend? I looked at Joe Bryant's Principles of VBD drafting and he gives a few techniques. He mentions last starter as your baseline but says that isnt necessarily the best way. Then he lists a few other approaches, one was to review the first 100 picks of your draft...but I do see the potential flaw in that....if people draft stupidly, it would impact my baselines!

So what is a rookie VBDer to do??

I think the best thing to do is set the baseline at "replacement-level player"; a level of production that you think you can replace at very little cost. Last-starter is a decent hand-wave at that, but differs for different positions; RB24 is harder to replace than TE12. I would look not at the drafting patterns, but the scoring patterns in your league. Eyeball the top 50 at each position in your league; at what point does it start to look like "a bunch of guys"? That's your point baseline. I prefer scoring-based baselines to position-based baselines, because position-based baselines are dependent on projections for low-ranked players, which tend to be fairly variable from projection set to projection set (moreso than the projections for high-ranked players).

I mostly agree with this, though I would warn to read "replacement level player" to be judged by fantasy starter significance rather than by who is available on waivers. That is, if you take a league with an adequate amount of bench space, if you suddenly increase the amount of bench space, the best guys on waivers change but I don't think the VBD value of the pool of player changes much at all.

For a QB in a 12 team, start 1 QB league, someone mentioned their league carries 30. I probably wouldn't set it that deep. There might be 15-17 or so QBs who are fantasy significant, who will see enough starter time I think should probably be included in the players above 0 VBD because they will be expected to provide value. After I get past the 3-4 QBs who will be in a fantasy QBBC, that is probably where I'd put my baseline. Because RBs tend to get injured a bit more and you have 2 spots and more people will play matchups with the last spot, I'd go deeper on the RBs than I do on the QBs.

 

[Saint]

Maybe I am missing the point here, but I always used a custom ADP based on my league tendencies to predict the runs. For baselines, you can do it several ways. Here's one of the older posts on baselines that I still use: http://forums.footballguys.com/forum/index.php?showtopic=482478&st=0&p=10698382&hl=+setup%20+draft%20+dominator&fromsearch=1entry10698382

 

[GregR]

if you know in advance that it will be RB heavy early, it becomes even more imperative to get RB's early - WR's, QB's and TE will be available later. In this situation, you want RB's valued highly, IMO.

I totally disagree. Let's say you're in a kicker-crazy league; 24 kickers will go in the first five rounds. Do you need to take a kicker early to make sure you get one? No, because you can get better production from other positions in every round. You can take K32 at the end of the draft, or pick up one on waivers, and kick you league's butt because you got additional production at other positions in every round.

I think the disagreement here comes down to how to incorporate different aspects of value into a draft strategy. There is an aspect of value that is completely dependent on the pool of players this year (their projections) and on league rules. It is how a league "should draft" in general, and you can capture it and visualize it well when you use VBD with a baseline in the fashion Cal and I are talking about.

There is also an aspect of value like moleculo is talking about based on how a league drafts. The further that a league drafts away from an ideal draft (as shown by VBD), the more that those drafting strategies will change what you should do to end up with an optimized team. This element of value is well captured by a dynamic VBD approach, where you look at the drop off in players between who is available at each position at my pick this round and who is available at my pick next round.

What I'm saying is, the best way to use VBD is to use it to do the first paragraph set of work. Then once you have that, you compare the "correct" draft order to what your league is like to draft. Doing so you will identify players that are poor values because they are taken too early, and good values because they are taken too late. Now you'll have some guidance when you use a dyanamic VBD look ahead at what players will be available at future picks because you know they will be available at picks they shouldn't if everyone draft optimally.

If you want the best team possible, you can't just count on a single ordered list of players like VBD produces. I think the disagreement between Cal and moleculo can probably be summed up as, Cal is advising to use VBD only for what it's good for and not try to force other aspects of value into it's final list since it isn't well suited to represent them.

So once you have VBD, then you need to look at your draft picks and the collection of players available at each set of picks to figure out what the optimum strategy is for your combination of picks, in light of the value that exists in the player pool (the VBD result) and the value changes that occur because of how your league drafts. You walk through a draft, mocking it with your league's draft tendencies to see what decisions you are faced with each round. And most importantly in doing so, you have the time to focus on a decision like "what happens to my team if I wait to pick up my RB3 until round X". You can test and see what happens if you take him, say, by round 4 or if you wait until round 5 or 6. Maybe you find that you there is a drop off between those points where you feel the need to take a RB4 much earlier and that puts you at risk of missing out on your QB or TE who you think will be available later than he should and was a good value.

Ultimately, reaching the optimal team that can be taken with your draft picks is a very hard equation to solve. Even though we have points available and can make a prediction of players available, you just about have to brute force and test every possibility as it is a very complex mathematical problem to solve fully even without including things like extra value of handcuffs, injury concerns, risks of losing starting job, etc.

 

[GregR]

Couple of other thoughts.... an auction league is a great way to illustrate this concept of not trying to build our league's tendencies into a VBD value prediction.

If we want to optimize our team in an auction, we need a set of values that express the optimal value for each player. You need to have that so that as the auction goes on, you can identify players that are costing too much, and players that cost too little and so are good values.

Now just knowing that doesn't always lead you to the best team possible. If a league overspends for RBs, for example, it can often be that your optimal team will also require you to overspend on RBs. The thing is your optimal team would overspend on RBs less than everyone else did. Why would your optimal team need to overspend on them? Because the salaries imposed by the other team's poorly-priced bids can make it so if you wait too long to get starters at an overpriced position, that the ones you get are so bad it really is very difficult to make up the values with extra quality in players at other positions.

So now compare that back to our draft league. In our auction league, we want our "list" to be the true value of the players and to express that value. And to optimize our team we still need to take into account the league's tendencies. But the best way to do that is not to try to auction off of a list built off league values. We want to auction off our "true value" list that we get creating prices from VBD, and then be well informed in other fashion how the differences between that true value and actual auction cost will happen.

Same with the draft. We want to see first what is the value in the pool of players. THEN we compared it with our league's draft history. We don't try to build the two in together. We use them as separate tools that can be used together to help map our way through a draft to the ultimate team.

 

[Saint]

Couple of other thoughts.... an auction league is a great way to illustrate this concept of not trying to build our league's tendencies into a VBD value prediction.

If we want to optimize our team in an auction, we need a set of values that express the optimal value for each player. You need to have that so that as the auction goes on, you can identify players that are costing too much, and players that cost too little and so are good values.

Now just knowing that doesn't always lead you to the best team possible. If a league overspends for RBs, for example, it can often be that your optimal team will also require you to overspend on RBs. The thing is your optimal team would overspend on RBs less than everyone else did. Why would your optimal team need to overspend on them? Because the salaries imposed by the other team's poorly-priced bids can make it so if you wait too long to get starters at an overpriced position, that the ones you get are so bad it really is very difficult to make up the values with extra quality in players at other positions.

So now compare that back to our draft league. In our auction league, we want our "list" to be the true value of the players and to express that value. And to optimize our team we still need to take into account the league's tendencies. But the best way to do that is not to try to auction off of a list built off league values. We want to auction off our "true value" list that we get creating prices from VBD, and then be well informed in other fashion how the differences between that true value and actual auction cost will happen.

Same with the draft. We want to see first what is the value in the pool of players. THEN we compared it with our league's draft history. We don't try to build the two in together. We use them as separate tools that can be used together to help map our way through a draft to the ultimate team.

Spot on here Greg.

 

[CalBear]

I go to our league draft pages each season to find out how many players at each position were drafted through 10 rounds. I then find the average for each position (usually going back no more than 3 years). Those are my baselines.

For example, through 10 rounds in 2010, 2011 and 2012, my league had drafted 42, 38 and 41 WRs. My 2013 baseline for the WR position is 40. If you do this, make sure you haven't had any major rule changes, or else the draft numbers could be skewed.

I think setting your baselines based on your league's drafting tendencies is exactly the wrong thing to do. Unless you believe your league drafts perfectly. The goal of VBD is to identify value among the players on the board; if your league consistently over-values one position, setting your baseline based on your league's tendencies will also over-value that position.

I know exactly what you are saying, but if I were to draft using the VBD Excel App spreadsheet or the Draft Dominator, I would end up with too many RBs too early. In most leagues, having a ton of RBs is fine, but receivers are highly valued in our scoring system, and I sometimes end up loaded at RB and scrambling all season long at WR. Personally, I think it's tougher to find a viable WR than a RB on the waiver wire.

I think that this web site's drafting systems place far too much value on RBs. Which is why I have created my own spreadsheet that links to the VBD Excel App for certain things that I deem relevant. And my league's historical drafts are remarkably consistent (since 2007, we have had 39, 40, 40, 42, 38 and 41 WRs drafted through 10 rounds). So, when I use historical numbers as my baselines, I have a very good idea about how many players at each position will be drafted in the next couple of rounds. I can use that to my advantage when anticipating a run at a certain position. For example, if we just finished the 6th round and 30 WRs are off the board (5-year average for my league), I can refer to my spreadsheet and see that about 5 WRs have historically been drafted in the 7th and 8th rounds combined. If I happen to value the next 8 WRs on my board equally, I can feel relatively safe that I can draft other positions in the 6th and 7th rounds and still get a WR that I like in the 9th round.

All positions are still tiered by value, but this system has more relevance to me and my league's drafting tendencies.

EDITED TO ADD: My league is an IDP league, which is why so few WRs are drafted in the 7th and 8th rounds. The defensive players start going heavy by these rounds.

I agree that the VBD Excel Spreadsheet and the default configuration of the Draft Dominator over-value RBs. That's precisely because they use "Joe's 'Secret' Formula" for setting baselines, and Joe's Secret Formula sets baselines based on expected players taken in the first 100 picks, which is heavily loaded with RBs. I think Joe's formula is the wrong approach, especially if there's anything different about your league than the old-school 1QB/2RB/2WR/1TE/no flex setup. In fact I would say that if your league is different than that it is complete wrong to use Joe's Secret Formula for your baselines. Using your league's tendencies is better than Joe's Secret Formula but it's the wrong approach for the same reason; with VBD you're trying to identify how to get the most points in your starting lineup, not predict how your leaguemates will draft.

 

[CalBear]

if you know in advance that it will be RB heavy early, it becomes even more imperative to get RB's early - WR's, QB's and TE will be available later. In this situation, you want RB's valued highly, IMO.

I totally disagree. Let's say you're in a kicker-crazy league; 24 kickers will go in the first five rounds. Do you need to take a kicker early to make sure you get one? No, because you can get better production from other positions in every round. You can take K32 at the end of the draft, or pick up one on waivers, and kick you league's butt because you got additional production at other positions in every round.

The above statement was not an absolute rule, it's a rebuttal that you should ignore league trends. I pose you need to know league trends to properly set up a static VBD baseline.

let's check the math. The difference between K1 and K25 is roughly 25 points. Per my projections, that's about the same as the difference between RB1 and RB5. You would be insane to want to get ahead of the kicker trend.

Here's a hypothetical: 12 team league. Start 1 QB, 2 RB, 3 WR, 1 TE. your particular VBD ranking is pretty traditional - RB's, WR, and QB all sprinkled together nicely. Lets assume all of your league-mates will be picking RB exclusively in rounds 3 and 4. You are drafting in, say, the 9th spot.

let's assume the first two rounds go mostly by ADP - roughly 17 RB's, 5 WRs, 1 QB, 1 TE. Now, we know that after 4 rounds, there will be 41 RB's, 6 WR's, 1 QB, and 1 TE.

What should you do in the first two rounds? Per ADP and (coincidentally) your VBD sheet, you take calvin Johnson at 1.09 and AJ Green 2.03. Drafting at 3.09, the best available RB will be Ahmand Bradshaw (RB26). That sucks, so instead you buck the trend and take Brees. At 4.03, the top RB will be Shane Vareen (RB31). If you are true to your VBD board, you take Demaryius Thomas (WR6) instead. By the time you get to 5.09, the top RB on the board is Vick Ballard (RB42), and that is who you will be forced to take. at 6.03, you are drafting Isiah Pead as a starting RB.

Had your VBD accounted for this trend, you would have taken Richardson (RB10) at 1.09 and Ridley at 2.04. In the middle of the crazy RB run, you still would have been able to snag Brees ( 3.09), Demaryius Thomas (4.03), and followed up with Harvin at 5.09, and Jordy Nelson at 6.03.

Getting caught by the crazy run:

1.09: Calvin Johnson

2.03: AJ Green

3.09: Drew Brees

4.03: Demaryius Thomas

5.09: Vick Ballard

6.03: Isaiah Pead

being ahead of the run:

1.09: Trent Richardson

2.03: Stevan Ridley

3.09ew Brees

4.03: Demaryius Thomas

5.09: Percy Harvin

6.03: Jordy Nelson

I think the second team is far superior. The way you get that is with advance knowledge of how your league will draft. If your VBD static baseline were set up to capture this, RB value would be increased and this would have lead you to take Richardson+Ridley instead of the more sexy Calvin/AJ.

Your hypothetical is (obviously) contrived. The principle of VBD is that you will get more scoring in your starting lineup by choosing players who will score more than replacement players, at whatever position they are. A stud WR paired with a weak RB will score more than a mediocre RB paired with a mediocre WR. To use your specific example, FBG is currently projecting Calvin Johnson for 108 VBD points, and Trent Richardson for 68, based on worst-starter baselines. AJ Green is projected for 77, Stevan Ridley for 29. If you give up those 185 VBD points in the first two rounds because you "need a RB", you will never get them back, because the WRs available later in the draft don't have the scoring potential of the stud WRs to differentiate them from the replacement-level RBs. If you invent an absurd scenario where you can get the #13 and #14 WRs (by FBG projections) in the fifth and sixth rounds, sure, drafting RBs would work, but even in your contrived scenario those guys wouldn't be available.

 

[moleculo]

if you know in advance that it will be RB heavy early, it becomes even more imperative to get RB's early - WR's, QB's and TE will be available later. In this situation, you want RB's valued highly, IMO.

I totally disagree. Let's say you're in a kicker-crazy league; 24 kickers will go in the first five rounds. Do you need to take a kicker early to make sure you get one? No, because you can get better production from other positions in every round. You can take K32 at the end of the draft, or pick up one on waivers, and kick you league's butt because you got additional production at other positions in every round.

The above statement was not an absolute rule, it's a rebuttal that you should ignore league trends. I pose you need to know league trends to properly set up a static VBD baseline.

let's check the math. The difference between K1 and K25 is roughly 25 points. Per my projections, that's about the same as the difference between RB1 and RB5. You would be insane to want to get ahead of the kicker trend.

Here's a hypothetical: 12 team league. Start 1 QB, 2 RB, 3 WR, 1 TE. your particular VBD ranking is pretty traditional - RB's, WR, and QB all sprinkled together nicely. Lets assume all of your league-mates will be picking RB exclusively in rounds 3 and 4. You are drafting in, say, the 9th spot.

let's assume the first two rounds go mostly by ADP - roughly 17 RB's, 5 WRs, 1 QB, 1 TE. Now, we know that after 4 rounds, there will be 41 RB's, 6 WR's, 1 QB, and 1 TE.

What should you do in the first two rounds? Per ADP and (coincidentally) your VBD sheet, you take calvin Johnson at 1.09 and AJ Green 2.03. Drafting at 3.09, the best available RB will be Ahmand Bradshaw (RB26). That sucks, so instead you buck the trend and take Brees. At 4.03, the top RB will be Shane Vareen (RB31). If you are true to your VBD board, you take Demaryius Thomas (WR6) instead. By the time you get to 5.09, the top RB on the board is Vick Ballard (RB42), and that is who you will be forced to take. at 6.03, you are drafting Isiah Pead as a starting RB.

Had your VBD accounted for this trend, you would have taken Richardson (RB10) at 1.09 and Ridley at 2.04. In the middle of the crazy RB run, you still would have been able to snag Brees ( 3.09), Demaryius Thomas (4.03), and followed up with Harvin at 5.09, and Jordy Nelson at 6.03.

Getting caught by the crazy run:

1.09: Calvin Johnson

2.03: AJ Green

3.09: Drew Brees

4.03: Demaryius Thomas

5.09: Vick Ballard

6.03: Isaiah Pead

being ahead of the run:

1.09: Trent Richardson

2.03: Stevan Ridley

3.09ew Brees

4.03: Demaryius Thomas

5.09: Percy Harvin

6.03: Jordy Nelson

I think the second team is far superior. The way you get that is with advance knowledge of how your league will draft. If your VBD static baseline were set up to capture this, RB value would be increased and this would have lead you to take Richardson+Ridley instead of the more sexy Calvin/AJ.

Your hypothetical is (obviously) contrived. The principle of VBD is that you will get more scoring in your starting lineup by choosing players who will score more than replacement players, at whatever position they are. A stud WR paired with a weak RB will score more than a mediocre RB paired with a mediocre WR. To use your specific example, FBG is currently projecting Calvin Johnson for 108 VBD points, and Trent Richardson for 68, based on worst-starter baselines. AJ Green is projected for 77, Stevan Ridley for 29. If you give up those 185 VBD points in the first two rounds because you "need a RB", you will never get them back, because the WRs available later in the draft don't have the scoring potential of the stud WRs to differentiate them from the replacement-level RBs. If you invent an absurd scenario where you can get the #13 and #14 WRs (by FBG projections) in the fifth and sixth rounds, sure, drafting RBs would work, but even in your contrived scenario those guys wouldn't be available.

Thanks, i don't have FBG projections handy. I don't care for Ridley that high; I was just pulling the next guy from the ADP list. what is the differential in terms of actual projected FF points vs VBD points?

VBD aside, I simply would not feel comfortable starting the season with RB42 as my best back, regardless of how good my two WR's are. depending on league size, the RB WW gets pretty bare. there are maybe 45 RB's worth having. If there are 12 teams hoarding 4 backs each, it will be near impossible to find starting quality backs on the WW.

 

[hotboyz]

I go to our league draft pages each season to find out how many players at each position were drafted through 10 rounds. I then find the average for each position (usually going back no more than 3 years). Those are my baselines.

For example, through 10 rounds in 2010, 2011 and 2012, my league had drafted 42, 38 and 41 WRs. My 2013 baseline for the WR position is 40. If you do this, make sure you haven't had any major rule changes, or else the draft numbers could be skewed.

I think setting your baselines based on your league's drafting tendencies is exactly the wrong thing to do. Unless you believe your league drafts perfectly. The goal of VBD is to identify value among the players on the board; if your league consistently over-values one position, setting your baseline based on your league's tendencies will also over-value that position.

I know exactly what you are saying, but if I were to draft using the VBD Excel App spreadsheet or the Draft Dominator, I would end up with too many RBs too early. In most leagues, having a ton of RBs is fine, but receivers are highly valued in our scoring system, and I sometimes end up loaded at RB and scrambling all season long at WR. Personally, I think it's tougher to find a viable WR than a RB on the waiver wire.I think that this web site's drafting systems place far too much value on RBs. Which is why I have created my own spreadsheet that links to the VBD Excel App for certain things that I deem relevant. And my league's historical drafts are remarkably consistent (since 2007, we have had 39, 40, 40, 42, 38 and 41 WRs drafted through 10 rounds). So, when I use historical numbers as my baselines, I have a very good idea about how many players at each position will be drafted in the next couple of rounds. I can use that to my advantage when anticipating a run at a certain position. For example, if we just finished the 6th round and 30 WRs are off the board (5-year average for my league), I can refer to my spreadsheet and see that about 5 WRs have historically been drafted in the 7th and 8th rounds combined. If I happen to value the next 8 WRs on my board equally, I can feel relatively safe that I can draft other positions in the 6th and 7th rounds and still get a WR that I like in the 9th round.

All positions are still tiered by value, but this system has more relevance to me and my league's drafting tendencies.

EDITED TO ADD: My league is an IDP league, which is why so few WRs are drafted in the 7th and 8th rounds. The defensive players start going heavy by these rounds.

I agree that the VBD Excel Spreadsheet and the default configuration of the Draft Dominator over-value RBs. That's precisely because they use "Joe's 'Secret' Formula" for setting baselines, and Joe's Secret Formula sets baselines based on expected players taken in the first 100 picks, which is heavily loaded with RBs. I think Joe's formula is the wrong approach, especially if there's anything different about your league than the old-school 1QB/2RB/2WR/1TE/no flex setup. In fact I would say that if your league is different than that it is complete wrong to use Joe's Secret Formula for your baselines. Using your league's tendencies is better than Joe's Secret Formula but it's the wrong approach for the same reason; with VBD you're trying to identify how to get the most points in your starting lineup, not predict how your leaguemates will draft.

I don't agree that setting ya baseline according to ya league tendencies is not getting the most points in ya lineup I'm not predicting what they gonna draft I'm setting a baseline knowing what positions go by certain point in the draft thus maximizing value. The hardest thing I deal with in setting my bsseline is my lineup requirements We go 1qb 1rb1wr 1 TE 3 flex RB/WR

 

[GregR]

I go to our league draft pages each season to find out how many players at each position were drafted through 10 rounds. I then find the average for each position (usually going back no more than 3 years). Those are my baselines.

For example, through 10 rounds in 2010, 2011 and 2012, my league had drafted 42, 38 and 41 WRs. My 2013 baseline for the WR position is 40. If you do this, make sure you haven't had any major rule changes, or else the draft numbers could be skewed.

I think setting your baselines based on your league's drafting tendencies is exactly the wrong thing to do. Unless you believe your league drafts perfectly. The goal of VBD is to identify value among the players on the board; if your league consistently over-values one position, setting your baseline based on your league's tendencies will also over-value that position.

I agree, and very well stated what the issue is with doing that.

The tendencies of how a league drafts players are indeed useful. But I don't think using them in the baseline is the place to apply them. Better to work with a baseline that highlights the value in the pool of players regardless of how the league drafts. Then take that result and compare it to how the league drafts to see where good and bad values exist.

I'm not quite sure I understand what you are saying here. How can you set baselines regardless of your league tendencies, especially one as difinitive as mine. So how would u determine baseline? I use the halfway point of my draft my draft is 14 rds so when I set my baseline at 7th rd I believe most teams starting lineups are set excluding kicker and defense 2 yrs in a row this league like clockwork drafts 12 Qb 35Rb 30Wr and 7 TE

Haven't forgot your post. Have been trying to think of a good way to express a few things here.

 

[CalBear]

I don't agree that setting ya baseline according to ya league tendencies is not getting the most points in ya lineup I'm not predicting what they gonna draft I'm setting a baseline knowing what positions go by certain point in the draft thus maximizing value. The hardest thing I deal with in setting my bsseline is my lineup requirementsWe go 1qb 1rb1wr 1 TE 3 flex RB/WR

Taking a less-valuable player because the guy after you would take him if you didn't take him first isn't "maximizing value"; quite the opposite. Knowing when positions will go in the draft is only useful for maximizing value if you know when they should go.

 

[Biabreakable]

I don't agree that setting ya baseline according to ya league tendencies is not getting the most points in ya lineup I'm not predicting what they gonna draft I'm setting a baseline knowing what positions go by certain point in the draft thus maximizing value. The hardest thing I deal with in setting my bsseline is my lineup requirements

We go 1qb 1rb1wr 1 TE 3 flex RB/WR

For worst starter your baselines would be (assuming 12 teams?) QB12 TE12 RB 30 WR 30. While teams may opt to start more than 30 RB so perhaps your baseline there should be slightly higher than at WR instead of them being even. This is where knowing league tendencies can help. If more lineups have 3RB than 3WR? How many more? Is that correct? Should more teams be starting WR? The answer to these questions for your league will help you find inefficiencies and gain an edge on value against your competition.

 

[Binky The Doormat]

Jeebus mommy i can't follow you math guys. How about this ...I know with a decent amount of certainty that our league will draft (12 team league - 16 rds/0.5 ppr 1/2/2/1/1/):

by end of 3rd rd: 8QBs/20RBs/7WRs/1TEs

by the end of 4th rd: 10QBs/27RBs/11WRs/2TEs

by the end of the 5th rd: 10QBs/30RBs/18WRs/2TEs

by the end of the 6th rd: 10QBs/32RBs/26WRs/4TEs

by the end of the 7th rd: 12QBs/33RBs/31WRs/8TEs

by the end of the 8th rd: 13QBs/35RBs/38WRs/10TEs

by the end of the 9th rd: 13QBs/41RBs/42WRs/12TEs

by the end of the 10th rd: 16QBs/42RBs/48WRs/13TEs

That being given - what should I set the flex up as? I play with it quite a bit (mostly the 2.3/2.7 default) here and some other sites to which I subscribe. Help!!

 

[hotboyz]

I don't agree that setting ya baseline according to ya league tendencies is not getting the most points in ya lineup I'm not predicting what they gonna draft I'm setting a baseline knowing what positions go by certain point in the draft thus maximizing value. The hardest thing I deal with in setting my bsseline is my lineup requirements

We go 1qb 1rb1wr 1 TE 3 flex RB/WR

Taking a less-valuable player because the guy after you would take him if you didn't take him first isn't "maximizing value"; quite the opposite. Knowing when positions will go in the draft is only useful for maximizing value if you know when they should go.

Why do you think this would lead to taking a lesser value player? That's not how it works for me I kno this goes against the grain but I don't move my baseline during da draft I come with my value # throw them in the pile sorted by value #. I don't draft by this alone some common sense comes into play and sometimes a gut feel of the way a draft is flowing. So if its 4th Rd and my vbd list is saying take a WR but I have a RB with similar value and I got a feel by knowing my league tendencies that there is gonna be a RB run coming up I may take that RB with the hope that WR comes rd again. We all kno drafts never go according to plan that's why I think Mock Drafts are completely use less

 

[CalBear]

I don't agree that setting ya baseline according to ya league tendencies is not getting the most points in ya lineup I'm not predicting what they gonna draft I'm setting a baseline knowing what positions go by certain point in the draft thus maximizing value. The hardest thing I deal with in setting my bsseline is my lineup requirements

We go 1qb 1rb1wr 1 TE 3 flex RB/WR

Taking a less-valuable player because the guy after you would take him if you didn't take him first isn't "maximizing value"; quite the opposite. Knowing when positions will go in the draft is only useful for maximizing value if you know when they should go.

Why do you think this would lead to taking a lesser value player? That's not how it works for me I kno this goes against the grain but I don't move my baseline during da draft I come with my value # throw them in the pile sorted by value #. I don't draft by this alone some common sense comes into play and sometimes a gut feel of the way a draft is flowing. So if its 4th Rd and my vbd list is saying take a WR but I have a RB with similar value and I got a feel by knowing my league tendencies that there is gonna be a RB run coming up I may take that RB with the hope that WR comes rd again. We all kno drafts never go according to plan that's why I think Mock Drafts are completely use less

Setting your baseline based on your league tendencies will result in you taking lesser value players, unless your league always drafts perfectly. If your league overvalues or undervalues one position, you will wind up following the herd in over-choosing overvalued players and failing to choose undervalued players. The goal of VBD is to model how valuable players are in fantasy football, not to predict when they'll be taken in the draft.

 

[hotboyz]

So Cal bear if that is the case how do you determine baseline? You have to have something no measure against right? Do if you not using what u expect to be drafted by a certain point what do you use? If you throw out some random number like 100 you are essentially doing the same thing that's a random number! Whatever number u set is gonna alter value or X number. When you use draft tendencies that also factors in lineup requirements no. By me using the halfway point and knowing that pretty much most if my league has there starters by this point I don't see how that can be wrong?

 

[hotboyz]

When u say it will result in me taken lesser value thats lesser value compared to the rest of the fantasy world but its maximum value compared to my specific league. That's all you care about is how players relate to your league and ya leagues requirements and scoring

 

[Maurile Tremblay]

Let's say you're in a kicker-crazy league; 24 kickers will go in the first five rounds.

Your hypothetical is (obviously) contrived.

 

[Maurile Tremblay]

For what it's worth, I don't think there's any such thing as a uniquely correct baseline.

Worst starter, average backup, best undrafted player, the top 100 overall, and so on -- all of them have their merits and their demerits. Each is good in certain contexts and problematic in others, even within the same draft.

 

[CalBear]

So Cal bear if that is the case how do you determine baseline? You have to have something no measure against right? Do if you not using what u expect to be drafted by a certain point what do you use? If you throw out some random number like 100 you are essentially doing the same thing that's a random number! Whatever number u set is gonna alter value or X number. When you use draft tendencies that also factors in lineup requirements no. By me using the halfway point and knowing that pretty much most if my league has there starters by this point I don't see how that can be wrong?

VBD based on the idea of Value Over Replacement Player (VORP) popularized by the baseball quants. VORP starts by defining a baseline level of production, which is what we should be doing with VBD. Choosing worst-starter, or first-100-picks as a baseline is a hand wave at setting a baseline level of production, but there are issues with either method in terms of finding what really represents a baseline level of production.

Quants use the Mendoza Line (.200 BA), or something similar, to set the baseline for a replacement level player. Ideally in fantasy football we'd be setting baselines based on projected points at each position, which you can do by looking at historical scoring patterns based on your league's scoring system.

 

[Biabreakable]

For what it's worth, I don't think there's any such thing as a uniquely correct baseline.

Worst starter, average backup, best undrafted player, the top 100 overall, and so on -- all of them have their merits and their demerits. Each is good in certain contexts and problematic in others, even within the same draft.

Something I try to pay attention to when setting baselines is how close the total projected points are at each position and if needed to move my baseline so that these are closer to the same total points for each position.

I noticed this when going over the last 11 years of data by VBD the baseline WR scores 120-150 total points so I set the baseline of the RB in a similar range of 120-150 total points. The TE position is harder to pair with these because the higher number of starters at the RB/WR position. So for TE I set the baseline at about 75-80 total points. I just dont want the total points at any position to ever be more than twice as much as another for comparative purposes.

 

[Gottabesweet]

Also having a hard time with this. I'm in a 12 team league PPR 6 per TD etc.

We only start

QB

RB

WR

TE

2 Flex

K

D

ONLY 4 Bench spots.

Now when I plug in data it shows Graham at 16 overall. I'd assume he should be much higher. When I delete the flex and swing it to 1RB 3WR starters Graham jumps to 5th justifying me picking him 1.9

The first 10 players are RBs otherwise with the potential of starting 1,2 or 3.

How would you set up VDB

My league I'd assume 4 QBs will go In the first 18 picks.

 

[cstu]

if you know in advance that it will be RB heavy early, it becomes even more imperative to get RB's early - WR's, QB's and TE will be available later. In this situation, you want RB's valued highly, IMO.

I totally disagree. Let's say you're in a kicker-crazy league; 24 kickers will go in the first five rounds. Do you need to take a kicker early to make sure you get one? No, because you can get better production from other positions in every round. You can take K32 at the end of the draft, or pick up one on waivers, and kick you league's butt because you got additional production at other positions in every round.

You still need a RB though and while you're mostly correct it would lead to you having a very poor RB at your starter and tougher to compete. However, if you use your 1st on RB* and let everyone else pick over RB's you can get a lot of value at the other positions.

*This assumes you don't draft #12 and 11 RB's are already off the board.