Updating Cheatsheets mid-draft

[ohiostfb 0]

Mike,

 

In the update for version 1.1 letter, you state the following regarding your changes to DB:

 

"In Draft Buddy include option to clear (if checked) or keep (not checked) the keepers and draft picks already input when running the Draft Buddy Setup. This is important if you are mid-draft but want to update cheatsheets from the Compiler or change Buddy settings without losing the keepers and draft picks you’ve already tracked in Buddy."

 

 

Now, first of all, I guess that this could be in a situation of a "slow" draft which takes a few days or more and then info (projections/depth charts) changes. Obviously in this case you would want to change the cheatsheets and update DB. For instance if I were in the 3rd round of a draft when the Travis Henry deal broke, then I can certainly see why you would need to update a cheatsheet mid-draft.

 

My question is in regards to whether you're doing this during "fast" drafts.....mid-stream.

 

I know that this is probably a personal preference and this is something that you found out during a draft so I don't want to ask you all your secrets, but is updating the cheatsheets during your draft (and thus Draft Buddy) something that you're doing.....or better yet something that you think we should look into? If so, I'm guessing that you are adjusting your baselines due to how many players have been picked by position, which would be moving towards dynamic VBD (as "created" over at FBGs??).

 

Anyways, I guess I just want to know if you see any reason to update the cheatsheets during the middle of a "fast" live draft.....which as I see it, would be dynamic VBD?

 

 

For your info, I enjoy the dynamic VBD principles, but at this time I don't care for FBGs product in this area and just really enjoyed the Draft Buddy last year. I'm just looking for ways to incorporate the things that I would like to try, and wanted to know if you or anyone out there is using the Draft Buddy to accomplish this.

[Mike MacGregor 20]

Hi ohiostfb. When I made that comment I was actually just thinking a slow draft. I never update the Compiler to Buddy during a live draft. I don't think there is enough time to do so, and I don't think it is a simple process to achieve this dVBD concept you're referring to. I understand the basic concepts of what dVBD is trying to accomplish. Honestly, using the Compiler overall rankings model to start the draft, switching up to positional rankings for mid to late draft, and utilizing tiering will be more effective than a dVBD model at the end of the day. Plus it is more straight forward to understand and apply, utilizing fewer assumptions.

[ohiostfb 0]

Thanks for the response, Mike. I've done much thought on the subject and have always viewed dVBD as basically a sexier form of tiering, but you're probably right in that simply using tiering is more straight forward.

 

And when I'm on the clock or my pick is approaching, that's what I want.

 

 

Congrats on the baby boy. Keep up the great work.

[ephford 0]

Hi ohiostfb. When I made that comment I was actually just thinking a slow draft. I never update the Compiler to Buddy during a live draft. I don't think there is enough time to do so, and I don't think it is a simple process to achieve this dVBD concept you're referring to. I understand the basic concepts of what dVBD is trying to accomplish. Honestly, using the Compiler overall rankings model to start the draft, switching up to positional rankings for mid to late draft, and utilizing tiering will be more effective than a dVBD model at the end of the day. Plus it is more straight forward to understand and apply, utilizing fewer assumptions.

 

All of this assumes auction draft, and is in regards to dynamic Value Based Drafting:

 

Mike,

 

Thanks for the great product. I was searching around to see if anyone asked about dVBD stuff becuase I had presumed it would appear in the draft buddy. I didn't realize it was still somewhat isolated as a FBG's thing.

 

I think my verdict is still out on the use of dVBD since last year my "home-grown" dVBD software kept adjusting the suggested max bid value downward as people kept overpaying for their players (by my estiamtion). Of course, the studs left still went for a premium, closer to where they were originally estimated for the auction.

 

What I was wondering is if you can set an option for a second column with dVDB values. That way we can choose whether to pay attention to it and recognize the impact of people overpaying through the course of the draft.

 

That's my suggestion! I have another one that I may put under a new thread if I can't find the answer.

 

Ephraim

[Mike MacGregor 20]

Hi Ephraim. It sounds good in theory to adjust the projected auction values to account for over/under paying in the early going, but how would the calculations work for that info? That is where it breaks down for me. Plus the time required and computer power to process the calcs after each pick.

 

We have a baseball version of the Compiler and Buddy which just this past season teamed with RotoChamps (www.rotochamps.com) to provide player projections and insight on valuing players for fantasy baseball. Since auctions are a lot more common in baseball, I hope to earn some valuable insight into valuing players for auctions when I hook up with RC to plan out the baseball Compiler for 2006. Then the next step will be to apply that knowledge to the football Compiler.

[ephford 0]

Hi Ephraim. It sounds good in theory to adjust the projected auction values to account for over/under paying in the early going, but how would the calculations work for that info? That is where it breaks down for me. Plus the time required and computer power to process the calcs after each pick.

 

We have a baseball version of the Compiler and Buddy which just this past season teamed with RotoChamps (www.rotochamps.com) to provide player projections and insight on valuing players for fantasy baseball. Since auctions are a lot more common in baseball, I hope to earn some valuable insight into valuing players for auctions when I hook up with RC to plan out the baseball Compiler for 2006. Then the next step will be to apply that knowledge to the football Compiler.

Hmm, I don't have my file handy but I'll look into it. I think I got that right.

 

 

Basically, you have to agree with the VBD philosophy.

 

To calculate that we find our baselines, come up with the new fantasy points (Peyton minus the baseline is 154pts), and divide the total number of dollars to be spent in the entire draft by the entire number of baseline fantasy points available. This assigns a $ per point above baseline value. Then when you multiply your player's VBD fantasy point value by this figure you get the long-sought "Max bid" or calulated auction worth.

 

My league, ignoring keepers, starts out with 100,000,000 in money and whatever the amount of VBD fantasy points may be (since everyone's projections/baselines can be different). My crappy little spreadsheet had fixed worksheets with my projections in them and was very dynamic; Every calculated VBD number (peyton = +154, Charlie Batch = -25) would change as soon as I tweaked one scoring parameter or one projected stat that influenced the calculation. I am not familiar with a spreadsheet that runs off of macros (like yours), so I don't know how hard it is to re-calculate, especially when you are referencing the compiler as an external source of info.

 

So, since mine changed as soon as a player was picked and paid for I could simply get Excel to subtract that dollar amount from the original total I started out with. As soon as that number changed, the relative worth of the each fantasy point above baseline changed again, and I had new dollar amounts calculated for the remaining players.

 

In one sense, it would be no different if I could somehow change the salary cap total in the draft buddy setup page after each pick and then rerun it without clearing keepers and players already drafted.

 

Make sense? I need to figure out how to break your protection and I may put this in myself.