QB Early -- What's wrong with this analysis? (1 Viewer)

[austinball]

Does the decision to go QB early not just boil down to this -- your early QB pick needs to outscore a later QB pick by MORE than your early stud WR (for example) would outscore a later WR pick? Let me illustrate.

Im drafting at the 10/11 in a 10 team league with 6 pts for passing TDs. Compare two draft scenarios. One i go QB with my second pick and the other I go Dez Bryant in the 2nd. In both drafts I fill the vacant spot in the 6th round with best available talent according to current ADP.

So Draft #1 goes:

Lynch RB

Brees QB

VJax WR

Ellington RB

Cameron TE

Harvin WR -or- D Jackson WR

Draft #2 goes:

Lynch RB

Dez WR

VJax WR

Ellington RB

Cameron TE

Foles QB -or- Brady QB

To determine if going QB early is a good or bad idea you have to compare how well Brees stacks up vs Foles/Brady choice AGAINST how well Dez stacks up vs Harvin/DJax choice. Everyone else on the rosters cancel each other out.

So the analysis is pretty much a wash actually, but its variance is highly dependent on the later QBs. According to FBG current projections for per game scoring in 2014:

Brees = 24.9

Harvin = 9.0

DJax = 8.8

Foles = 22.3

Brady = 20.8

Dez = 12.3

On a per game basis, Brees will outscore the later QBs by 2.6-4.1 and Dez will outscore later WRs by 3.3-3.5. So if everyone hits their projections by FBG, it pays to take Dez not Brees early if you can get Foles in the 6th round, but better to take Brees early not Dez if you get Brady or worse in the 6th.

So whats my point? I guess its that I hear from so many sources that going QB early will kill your depth at the RB or WR position. But looking at it THIS way, you really don't. You accept a lesser player at ONE roster slot (WR) for a better player at ONE roster slot (QB).

On a weekly basis you have to believe that your QB will outscore your opponent's QB by more than his WR1 outscores your WR2. Everything else is a wash.

What am I missing?? Because when i look at it this way, I think going QB early is superior. Id rather stockpile later WRs with upside to compete week in and week out with my opponent's #1 WR than draft multiple QBs to compete against Drew Brees every week.

 

[Concept Coop]

Non-PPR leagues in which you can only start 2 WRs are rare. If that's your format, your math look okay at first glance. But understand that most leagues start 2+ with flex spots and reward receptions.

 

[GregR]

You aren't missing anything.

A lot of people like to come up with "strategies" on how to draft. They take RBs early and often. Or take WRs early or won't take a RB until round 5. Or make sure they get a stud TE or QB. Or take players who have the same byes.

The best team is always going to be the result of taking the combination of players available at your picks who collectively score the most points. It's such a simple concept that it feels insulting to say it to someone. But notice that none of those strategies actually take steps to confirm whether they are getting you to that result. It should be integral to your draft process.

It's not all that hard to focus on putting together the maximum scoring team, without needing general overarching strategies. You just walked through the basic premise. Find the combination of players that you believe will be available at your picks that combine for the most points. If it's a simple case of "QB in 2nd and WR in 6th... or QB in 6th and WR in 2nd" then it's a very easy comparison to do.

Since for a full draft, you need to factor in all of the rounds and positions, and who you take this round can impact whether you need to take that position in a later round, figuring out your draft strategy does have a little more complexity than just that example. But it's still pretty simple. Focus on the starters since they will be your early picks.

1. Get projections together and rank the players using some sort of VBD system so you can equate players across different positions. You need total point projections to be able to check what a combination of players will score... and you need rankings as well to help identify players who are good values from your draft.

2. Come up with an ADP list appropriate for your league. That might be a major site's ADP list, or you might have to edit it to reflect your league's tendencies. But it should reflect what you think your league's draft will look like.

3. Compare the order the ADP list says the players are likely to be drafted, with the rankings you have on how you believe they actually should be drafted. Note the players who last longer than they should. Note especially the ones that look like they last to your specific picks each round. Note how many points VBD advantage they give you over players who should be taken at the spot they are falling to, as some players will be a bigger value than others. You can also note the bad values, players taken too early, so you don't try to work them into your plans unless necessary.

4. Generally your maximum team is going to use a lot of these players who were good values at the picks you have. Figure out the combination of players available at your picks that gets as many of these good value players as possible. For each combination, add up the total points for each starting player to confirm how many total points that will be for your team. Try different combinations to see which is best. That's the step where we actually confirm we're building the best team, and not just building a team that some strategy said to build.

5. Finally, start asking yourself questions on what you should do if things don't go how you expected. "What if Value RB X is not available in the 3rd where I think he will be? How will that change my draft? Do I now need to take my backup RB earlier as a result, and how does that impact other positions?" Again, actually put together potential teams you might select in that situation and add up the points and see what options are best. You are testing out alternate scenarios so if it happens in a draft you'll have already thought through the consequences and don't need to waste time figuring out the best solution.

You also want to balance out the rest of the draft with backups and high upside bench players and give thought where it impacts your earlier round decisions. Does taking a particular RB with a top handcuff (a Foster-Tate, Priest-LJ type scenario) mean you have to take the backup earlier to get him, and if not how does that impact your draft? Factor those kind of things in to your strategy as well, but they don't necessarily show up in the final "what does this combo of players score?" result. Sometimes you might decide, this combo scores slightly less than another, but the security of having my handcuff makes it a more desirable team.

 

[rjv]

Sounds good but IMO when you draft a QB early, rd1-2, you will be playing catchup the entire draft.

 

[GregR]

I'll add, you can sum a lot of that latter part as "do a lot of mocks and check the results".

The thing we're adding that helps us is that we're identifying all the players we think will be values where they are taken. With that in hand, we can make our time doing the mocks more effective since we know our best team is probably going to be based around several of those particular players. Rather than stumbling through early mocks to discover the names that keep showing up on our best teams, we start out already having identified them.

 

[kutta]

Does the decision to go QB early not just boil down to this -- your early QB pick needs to outscore a later QB pick by MORE than your early stud WR (for example) would outscore a later WR pick? Let me illustrate.

Im drafting at the 10/11 in a 10 team league with 6 pts for passing TDs. Compare two draft scenarios. One i go QB with my second pick and the other I go Dez Bryant in the 2nd. In both drafts I fill the vacant spot in the 6th round with best available talent according to current ADP.

So Draft #1 goes:

Lynch RB

Brees QB

VJax WR

Ellington RB

Cameron TE

Harvin WR -or- D Jackson WR

Draft #2 goes:

Lynch RB

Dez WR

VJax WR

Ellington RB

Cameron TE

Foles QB -or- Brady QB

To determine if going QB early is a good or bad idea you have to compare how well Brees stacks up vs Foles/Brady choice AGAINST how well Dez stacks up vs Harvin/DJax choice. Everyone else on the rosters cancel each other out.

So the analysis is pretty much a wash actually, but its variance is highly dependent on the later QBs. According to FBG current projections for per game scoring in 2014:

Brees = 24.9

Harvin = 9.0

DJax = 8.8

Foles = 22.3

Brady = 20.8

Dez = 12.3

On a per game basis, Brees will outscore the later QBs by 2.6-4.1 and Dez will outscore later WRs by 3.3-3.5. So if everyone hits their projections by FBG, it pays to take Dez not Brees early if you can get Foles in the 6th round, but better to take Brees early not Dez if you get Brady or worse in the 6th.

So whats my point? I guess its that I hear from so many sources that going QB early will kill your depth at the RB or WR position. But looking at it THIS way, you really don't. You accept a lesser player at ONE roster slot (WR) for a better player at ONE roster slot (QB).

On a weekly basis you have to believe that your QB will outscore your opponent's QB by more than his WR1 outscores your WR2. Everything else is a wash.

What am I missing?? Because when i look at it this way, I think going QB early is superior. Id rather stockpile later WRs with upside to compete week in and week out with my opponent's #1 WR than draft multiple QBs to compete against Drew Brees every week.

The red text is key. Of course if we knew what was going to happen, your math would be right. But we don't know what is going happen, so we have to draft based upon what is more likely to happen.

Some say the top 3 QB's are locks and are very consistent, so it's fine to pick them early because the risk of bust is low. Some say the same about early WR's. That's why there's so much debate - we really don't know what is going to happen so we have to use our opinion. And we all have different opinions.

 

[austinball]

Thanks to all for your opinions and feedback. Greg holy cow very thoughtful response.

Sounds good but IMO when you draft a QB early, rd1-2, you will be playing catchup the entire draft.

I guess this commonly used argument against drafting QB early is what Im really addressing. Because I disagree that you have to play catchup all draft long. The ONLY position you have to compensate for is the position you didn't take in order to take a QB early. If that's WR1, then yes you are facing a disadvantage against squads that went WR early. But that disadvantage is only for ONE WR position in the lineup. And you have the advantage at QB in that matchup.

And Kutta I agree that the only thing we know about projections is that they will be wrong. But if we are ranking players and drafting based on projections then that's all we can go on. And the data Ive seen says that the early elite QBs are as dependable if not moreso than the elite WRs to product consistent results.

So do you feel more confident that you can pick up later WRs that can deliver close to Elite production? Or do you feel that way about QBs? I think its more likely to find a later WR who will product to WR1 #s.

 

[Zyphros]

It's an interesting thought that I'll have to go through a couple mocks through the weekend to see what kind of teams I could end up with. I agree that you're only creating a "loss" at 1 other position (whatever you pass up for QB), and since there is more of a gap between elite and mediocre RB's then I feel that shouldn't be an option. Instead it should be WR that you "replace" with a QB early since you can find very good options in the 5th and later. There is just more depth at WR than any other position so in theory they are more replaceable. Just gotta pick the right ones.

 

[bucsbaby]

The best team is always going to be the result of taking the combination of players available at your picks who collectively score the most points.

/thread

 

[shadyridr]

No offense but youre just figuring this out now? This is the whole principle of VBD

 

[bagger]

I'll add, you can sum a lot of that latter part as "do a lot of mocks and check the results".

The thing we're adding that helps us is that we're identifying all the players we think will be values where they are taken. With that in hand, we can make our time doing the mocks more effective since we know our best team is probably going to be based around several of those particular players. Rather than stumbling through early mocks to discover the names that keep showing up on our best teams, we start out already having identified them.

this is exactly what I do.I have gone from mathematical VBD to finding value between my projections and adp and incessantly mocking to find the right blend of players to generally structure a team.

 

[TwinTurbo]

Every year is different and waiting on a QB only works if you pick the right QB late and if the RB/WR/TE that you pick early instead, doesn't end up as a bust.

The perfect draft article advocates waiting on a QB as usual, but I think grabbing Peyton or Brees in rounds 2 or 3 respectively is a solid strategy this year. Rodgers injury history and emergence of Lacy concerns me a bit. After QB3, there is a very strong argument to wait as long as possible and address other positions. The difference between QB4 and QB12 is not that great and it is a wide open field.

 

[FreeBaGeL]

I guess this commonly used argument against drafting QB early is what Im really addressing. Because I disagree that you have to play catchup all draft long. The ONLY position you have to compensate for is the position you didn't take in order to take a QB early. If that's WR1, then yes you are facing a disadvantage against squads that went WR early. But that disadvantage is only for ONE WR position in the lineup. And you have the advantage at QB in that matchup.

It's a bit of a mind twister because depending on which perspective you look at it from, it seems to give the opposite results. Your scenario makes sense logically, but so does the "catch up" scenario. Let's say you and another drafter start off drafting the same positions in every round, only you swap QB and WR from rounds 1 and 6. So it looks like...

WR1--QB1

RB1--RB1

WR2--WR1

TE1--TE1

WR3--WR2

QB1--WR3

In this case you're drafting your WR1 in round 3 while they get it in round 1, you're getting your WR2 in round 5 while they get it in round 3, and you're getting your WR3 in round 6 while they get it in round 5. So now when comparing across the board, you're actually giving up 3+ positions to him (WR1, WR2, WR3, etc). In that way you are kind of playing catch-up.

And Kutta I agree that the only thing we know about projections is that they will be wrong. But if we are ranking players and drafting based on projections then that's all we can go on. And the data Ive seen says that the early elite QBs are as dependable if not moreso than the elite WRs to product consistent results.

So do you feel more confident that you can pick up later WRs that can deliver close to Elite production? Or do you feel that way about QBs? I think its more likely to find a later WR who will product to WR1 #s.

I think ignoring the risk of the round 6 guy is one big thing that you're overlooking. While the projections of all the guys you listed may add up to similar numbers, there is a lot more risk in guys like Harvin and DJax than there is in a guy like Brady. If we're talking about four positions here (round 1 WR, round 1 QB, round 6 QB, round 6 WR) and only one of them carries extreme risk, wouldn't it make sense to make that the one you avoid? Round 1 QBs may be slightly less risky than round 1 WRs, but round 6 QBs are a LOT less risky than round 6 WRs.

Another thing I think you're overlooking is the notion that you can find a WR1 later in the draft by taking a bunch of shots without acknowledging the cost involved in doing that. Pigeonholing yourself into drafting a bunch of high upside WRs late in the draft is going to cost you picks that you'd otherwise be using elsewhere. They're not free.

Lastly, I've not been able to find many ADP lists with Percy/DJax available in the 6th round unless you're talking about 10 team leagues. In 12 team leagues fantasypros ADP (and most ADP listings I've seen) have guys like Torrey Smith, Julian Edelman, Jeremy Maclin, and Marques Colston who will likely not only unbalance your projection numbers but also carry a whole lot more risk than guys like Tom Brady and Matt Ryan. I don't think any of them have the upside to be a top 3 WR the way that Brady does at QB either.

ETA: Another thing to consider is that the risk in a round 1 QB is not only that he will be a bust, but simply that he won't be mega out of this world awesome. You used Brees in your example which is good, but he's really the only one that has done it consistently. Aside from Peyton's record breaking outlier his typical season lately has landed him around 4500/35 which is something that those round 6 QBs we're talking about have not only finished in striking distance of, but have actually matched several times themselves. In that sense, there's a pretty fair amount of risk in round 1 QBs. Anything less than 5000+ yards and 40+ TDs is a "bust" at 1st round value.

 

[Ilov80s]

Drafting a QB early puts pressure on RB and WR. Those positions require 2-3 correct gambles. QB only requires 1 good gamble. I just prefer the odds of gambling at QB.

 

[Tool]

i like the consistency of the top qb's and usually draft them higher than most people. Of course, a lot depends on your scoring system but your analysis seems sounds with a quick glance.

 

[We Tigers]

There's nothing too tricky to it, yeah. It's a simple concept in theory. For me, the main reason I generally pass on an early QB is to preserve flexibility. At positions where you only start 1 player, covering that position right away reduces your ability to capitalize on additional value later by one player category. In rounds 5-7, I generally like to be in a position where I can just take the best player available without having to worry too much about covering/duplicating a particular position.

That said, I throw a lot of that out the window if I'm close to either end of a snake draft. Keeping your options open works best from the middle; if there are 20 picks between your 4th and 5th rounders, chances are that "value picks" you see slipping won't slip all the way to you.