VBD app vs. self calculated VBD (1 Viewer)

[Jutz]

When I first started playing FF I read Joe's article about the basic concept of VBD... back in the cheatsheets days when the site was completely free. Being an engineer / math geek kinda person... I thought it was awesome.

My first few years after hearing about VBD... I would get the last 'free' version of the projections from footballguys... screen scrape them into excel... and basically calculate my own VBD. I was relatively successful. I would usually make the playoffs, I made a couple SB appearences, I often placed and won enough $ to pay for my league dues the next year. Each year I was a significant team in the league.

As the years progressed, I got jobs that were more demanding, had a family, started hangin' out more in the FFA... and wasn't as interested in FF. I got lazy... but I was making more $... so I just payed the subscription and downloaded the VBD app from footballguys. Plug in the scoring rules... and [Levitar] BAM [/Levitar] I got my spreadsheet in seconds. I thought it was awesome... however; I was not nearly as successful. I did OK... but have have had several years where did not make the playoffs at all... and have not sniffed the SB.

This year I decided... I am going to pay for the website subscription... get the most recent projections... and create my own VBD to draft off off. I had some very unconventional drafts. In the 2 leagues I am in I drafted from the 5 spot and took Brady in the 1st round. In both leagues I did not draft my first RB until the 4th. In both leagues I drafted Jason Whitten in round 2 or 3... but... even somewhat surprisingly to me... I am having good success so far this year. I supposed the season still has time to fall apart for me... but I feel alot better about this years chances then in seasons past.

So my question to you... when you see that the VBD app has 'super secret formulas' and aparently (unknown to me before this year) combines the VBD draft strategy with some other strategy... do you think the app outhinks itself? Is it possible that straight up VBD with no formula trickery could perform better? Do we have paralasis through analysis with all the 'adjustments' in the app?

Or... is it just coincidence. Maybe I made better waiver moves this year? Maybe I have more time to dedicate? What say you?

 

[CalBear]

So my question to you... when you see that the VBD app has 'super secret formulas' and aparently (unknown to me before this year) combines the VBD draft strategy with some other strategy... do you think the app outhinks itself? Is it possible that straight up VBD with no formula trickery could perform better? Do we have paralasis through analysis with all the 'adjustments' in the app?

I think "Joe's Secret Formula" overstates the values of RBs in most league formats, so you're better off using a different baseline. Of course, you can do that within the VBD app, you don't need to write your own.

 

[Jutz]

So my question to you... when you see that the VBD app has 'super secret formulas' and aparently (unknown to me before this year) combines the VBD draft strategy with some other strategy... do you think the app outhinks itself? Is it possible that straight up VBD with no formula trickery could perform better? Do we have paralasis through analysis with all the 'adjustments' in the app?

I think "Joe's Secret Formula" overstates the values of RBs in most league formats, so you're better off using a different baseline. Of course, you can do that within the VBD app, you don't need to write your own.

As I was writing the OP... I was thinking the same thing... the RBs get all the love in the app. I hear what you are saying that you can make an adjustment within the app... but I feel like once I know that something doesn't seem right... it makes me wonder what else is in the 'secret formulas' and I feel like I need to do the work myself... or I don't have the same level of trust.

 

[CalBear]

So my question to you... when you see that the VBD app has 'super secret formulas' and aparently (unknown to me before this year) combines the VBD draft strategy with some other strategy... do you think the app outhinks itself? Is it possible that straight up VBD with no formula trickery could perform better? Do we have paralasis through analysis with all the 'adjustments' in the app?

I think "Joe's Secret Formula" overstates the values of RBs in most league formats, so you're better off using a different baseline. Of course, you can do that within the VBD app, you don't need to write your own.

As I was writing the OP... I was thinking the same thing... the RBs get all the love in the app. I hear what you are saying that you can make an adjustment within the app... but I feel like once I know that something doesn't seem right... it makes me wonder what else is in the 'secret formulas' and I feel like I need to do the work myself... or I don't have the same level of trust.

It's possible there are some other tweaks in there. Using the Draft Dominator for auction values, the values weren't directly aligned with VBD--I found out that there's a hidden multiplier based on expected number of games played, that you have to change by editing a text file in the Draft Dominator directory. (I leave that multiplier as > 1, but I nudge it down from the default settings).

 

[senor peso 2006]

I'd ask the op, where are you losing it in your leagues based on the strength of your teams? The VBD is good for the first 50 or 60 picks after that you are on your own. My guess is the first 4 or 5 rounds of your drafts are okay; you have decent value and production to set the base of your team (it could be better and it could be worse). From my experience, the real strenght of my teams comes from the late mid round picks to the late picks. If I do not get good value there my teams struggle. When I find value there, my teams are highly competive. So it makes me wonder if becuase life is getting busy and other things take priority (a good thing usally), that you don't have time to sniff out those late round picks that make the differnce (at least for me). To pull this together, is it the VBD defaults that are coming up short or maybe just not having the same amount of time as you have had in the past to prepare?

 

[moleculo]

projections that you can trust are the foundation that everything else is based on. If you have garbage projections, everything else falls apart.

What I did last year was make my own VBD ranking, and for my baseline, I averaged the projected scores of 1.5* expected # of league starters - that is, in a 12 man league, the baseline value was the average score of the top 18 QB's. Using an average makes the baseline less subjective to individual projections which is nice. I came up with 1.5 just by playing around with some #'s, and it seemed to work out for me...

Where my drafts fell apart was using projections that were whacky. A couple of bad calls by whomever does your projections can really throw your whole draft off.

 

[two_dollars]

I get the average draft positions off MDC, break them into tiers, move my sleepers/must haves up one or 2 tiers, organize them inside the tier based on team strengths and weaknesses (not neccessarily player strengths), and keep it handy while running mock drafts. I run about 20-30 mocks, then put together a basic draft strategy for my draft position vs roster requirements. At the actual draft I just keep an eye on what positions other players have filled up and run a dynamic VBD in my head. It never fails. I have never missed the playoffs in 6 years, 2 leagues.

 

[Jutz]

I'd ask the op, where are you losing it in your leagues based on the strength of your teams? The VBD is good for the first 50 or 60 picks after that you are on your own. My guess is the first 4 or 5 rounds of your drafts are okay; you have decent value and production to set the base of your team (it could be better and it could be worse). From my experience, the real strenght of my teams comes from the late mid round picks to the late picks. If I do not get good value there my teams struggle. When I find value there, my teams are highly competive. So it makes me wonder if becuase life is getting busy and other things take priority (a good thing usally), that you don't have time to sniff out those late round picks that make the differnce (at least for me). To pull this together, is it the VBD defaults that are coming up short or maybe just not having the same amount of time as you have had in the past to prepare?

Interestingly enough... I think I did better in my late round pics with my own self calculated VBD. As I mentioned... I drafted Witten based on VBD in round 2 for one league... and in round 3 in the other. He has not produced to that level of VBD at all. I drafted Carolina Steve Smith in round 2 in one of my leagues this year... and he has time to turn the season around... but has for the most part been a bust.The spot where I made the best move in both leagues was round 4 with Ray Rice as my first RB selected. He has helped me out tremendously.Some other later round values I got were Bowe and Cotchery at WR. I drafted Fred Jackson late in one league who was tremendous in ppr leagues for the first 4 games.... but am now struggling to fill in. I do have to admit though that my waiver wire pics... and having enough time to seek them out has probably helped me the most... and really... that may be the answer moreso then anthing else. I picked up the likes of Knoshawn, Mendenhall, MSW, and Collie... even got the Saints D off the wire... and Nate Kaedeing off the wire.I still believe that the VBD app does pump up RBs a little too much. In a way... I think since my self calculated VBD kept me away from the RBs early is what may have helped the most. Although Whitten has not been a stud... he is very consistent... and I think has a good chance to have a solid second half.

 

[CalBear]

I'd ask the op, where are you losing it in your leagues based on the strength of your teams? The VBD is good for the first 50 or 60 picks after that you are on your own. My guess is the first 4 or 5 rounds of your drafts are okay; you have decent value and production to set the base of your team (it could be better and it could be worse). From my experience, the real strenght of my teams comes from the late mid round picks to the late picks. If I do not get good value there my teams struggle. When I find value there, my teams are highly competive. So it makes me wonder if becuase life is getting busy and other things take priority (a good thing usally), that you don't have time to sniff out those late round picks that make the differnce (at least for me). To pull this together, is it the VBD defaults that are coming up short or maybe just not having the same amount of time as you have had in the past to prepare?

Interestingly enough... I think I did better in my late round pics with my own self calculated VBD. As I mentioned... I drafted Witten based on VBD in round 2 for one league... and in round 3 in the other. He has not produced to that level of VBD at all. I drafted Carolina Steve Smith in round 2 in one of my leagues this year... and he has time to turn the season around... but has for the most part been a bust.

That doesn't say anything about VBD at all; it says something about the projections you were using. VBD just tells you who is projected to score more points relative to others at their position. If you drafted Chad Ochocinco and Tony Gonzalez based on VBD, you'd be doing great.

 

[LittlePhatty]

Or... is it just coincidence. Maybe I made better waiver moves this year? Maybe I have more time to dedicate? What say you?

I'd guess it is more coincidence/luck and waiver pickups.You're slotted into your drafting position and you have no control over who is taken by other owners.

With the exact same VBD list, but different draft positions, you could have gotten:

Drafting at #4

Matt Forte

Steve Smith (CAR)

Clinton Portis

OR

Drafting at #7

Chris Johnson

Reggie Wayne

Ronnie Brown

I think VBD is the best way to go, but there will always be some luck involved.

 

[Rod]

When I first started playing FF I read Joe's article about the basic concept of VBD... back in the cheatsheets days when the site was completely free. Being an engineer / math geek kinda person... I thought it was awesome.My first few years after hearing about VBD... I would get the last 'free' version of the projections from footballguys... screen scrape them into excel... and basically calculate my own VBD. I was relatively successful. I would usually make the playoffs, I made a couple SB appearences, I often placed and won enough $ to pay for my league dues the next year. Each year I was a significant team in the league.As the years progressed, I got jobs that were more demanding, had a family, started hangin' out more in the FFA... and wasn't as interested in FF. I got lazy... but I was making more $... so I just payed the subscription and downloaded the VBD app from footballguys. Plug in the scoring rules... and [Levitar] BAM [/Levitar] I got my spreadsheet in seconds. I thought it was awesome... however; I was not nearly as successful. I did OK... but have have had several years where did not make the playoffs at all... and have not sniffed the SB.This year I decided... I am going to pay for the website subscription... get the most recent projections... and create my own VBD to draft off off. I had some very unconventional drafts. In the 2 leagues I am in I drafted from the 5 spot and took Brady in the 1st round. In both leagues I did not draft my first RB until the 4th. In both leagues I drafted Jason Whitten in round 2 or 3... but... even somewhat surprisingly to me... I am having good success so far this year. I supposed the season still has time to fall apart for me... but I feel alot better about this years chances then in seasons past.So my question to you... when you see that the VBD app has 'super secret formulas' and aparently (unknown to me before this year) combines the VBD draft strategy with some other strategy... do you think the app outhinks itself? Is it possible that straight up VBD with no formula trickery could perform better? Do we have paralasis through analysis with all the 'adjustments' in the app?Or... is it just coincidence. Maybe I made better waiver moves this year? Maybe I have more time to dedicate? What say you?

If you have past data for your league, I think you should use that.Instead of using the FBG's projections and trying to calculate how many points a player will score in the year, I use the actual score averages in my league going back eight years. In those eight years, I know the top QB has averaged 376 points, number two has averaged 350 points, the top RB has averaged 340 points and so on. Then I use the FBGs player rankings, plus some other sources and determine a ranking I agree with for each position. I assign the average points for each position to the players in my ranking. Now for the baseline. I know, based on our leagues past drafts, on average, how many positions will be taken a a given point in the draft. After analyzing our past drafts I determined that my baseline should be right around the point where a defense and a kicker are taken. For my league, it is right around 83. At this point in the draft 11 QB, 32 RB, 32 WR, 6 TE, 1 PK, 1 Def are taken and the players sort very close to their value, based on the past averages. One other thing I'll tell you is how to get a very accurate ADP for your players. I use MFL draft data. I only use actual drafts after August 25th for the most accurate information. I'm not going to go into a lot of detail but I'll give you the basics. In my league, we draft 24 QBs, 48 RBs, 60 WRs, 24 TEs, 24 PKs, and 24 Defs. That's all I'm interested in is the top X amount of players for each position. I'll tweak the percentages until I get the results as close to what I need. For example 55% of drafts will give me 63 WRs. I'll copy and paste the top 60 and their ADP into a spreadsheet. Do this for all the positions you need and sort by ADP. Renumber them, 1 - 204 in my case. Now this is great but I then compare the results from MFL with our leagues actual past draft results and I do have it set up to adjust the MFL numbers. It gets a little complicated but I hope you get the idea. The ending result ends up being very close to what happens on draft day. BTW, even after all this preparation for the draft, I'm doing about the same as the guy that grabbed a fantasy football rag and used those rankings. Go figure.I hope this helps.Rod

 

[Uruk-Hai]

Instead of using the FBG's projections and trying to calculate how many points a player will score in the year, I use the actual score averages in my league going back eight years. In those eight years, I know the top QB has averaged 376 points, number two has averaged 350 points, the top RB has averaged 340 points and so on. Then I use the FBGs player rankings, plus some other sources and determine a ranking I agree with for each position. I assign the average points for each position to the players in my ranking.

I think the Bat-Signal just went up for Maurile.Good, well-thought-out post though Rod

 

[inca911]

The reality is that if we take the exact same projections and you use the VBD app to generate auction values while I use my own calcs, my team will beat yours if we spend the same total amount. With that basic fact, it is clear that the app has some issues. I wrote an article a few years back that exposed a lot of the issues and proposed a two-line approach to fix a lot of the issues:

http://subscribers.footballguys.com/2006/0...e_johnson01.php

I've since adapted my approach to a one-line fit with a customizable baseline, and it's been a big advantage over other league-mates who I know just download the latest VBD and run with it. It just flat works and relies only on math without any tweaking. The tweaking should be used to generate your projections, not to further change values once projections have been set. Regardless of a two-line or one-line fit approach, the current VBD app is just plain flawed for generating auction values. Try this: go into the app and give AP another 15 TDs on top of his projections. What you'll find is the VBD auction price won't increase nearly enough to reflect the value of his additional 15 TDs. Make it 30 extra TDs and see what happens.

 

[⚡DEADHEAD⚡]

Some thoughts...

Projections don't have a very high certainty of actually being accurate, statistically speaking (say within 95% probability that the projection is accurate). I'd have much more faith in predictions that have a range.

The importance of fantasy production (and projections) should not be based on the number of points that a player is projected to get, but the relative performance compared to the position players ahead and behind him. That's why I like Dodds cheatsheets. I don't care how many points you think MJD is going to get. I do care that he is "projected" to be the number two RB in a PPR system. Ok, that's an easy one. But if you have Calvin Johnson, Bowe, Welker, MSW and Desean and can start three, I want to know who the top three are for any given week.

Finally, I don't believe for a second that there is such a thing as a fantasy expert. There are people who know things about fantasy football, there are fantasy football analysts, there are people who are better than others at fantasy football, but there are no experts. Don't buy into it.

 

[gheemony]

I don't agree with Joe's Secret Formula. This tweaked to put players where he wants to take them and is not a true relative comparison. It forces RBs up and Ks and Ds down. I agree that taking a K or D early is bad strategy (always take with my last 2 picks), but your baseline shouldn't be adjusted to force them down.

I use a variation of Maurile's auction method to set my VBD. Mostly, this is because my most important league is auction league. But I use the same general formula to create VBD values for draft leagues. Maurile's method is the most fundamentally sound. Read his article and create your own formula based on his theories. If you adjust to your league, it's great.

Also, the auction values in DD are horrible unless you switch to Maurile's method, but even then they are off. In general, auction values is a big weakness of this site.