my method for determining auction value (1 Viewer)

[schu]

Here's how I've figured $/VBD over the years.

1st I establish the baseline at each position by looking at past auction history and determining which players have value -- that is...the positional rank of 1st player at each position that was bought for minimum $. For my league (off the top of my head) this came out to QB18, RB33, WR32, TE15, K11, D13.

Using the baseline points scored at these positional ranks I then go to the VBD application and subtract those baselines from each player's points scored formula on the projection data tabs. Next I sum positive points at each position and sum these totals. So TOTAL VBD = sum VBD (QB, RB, WR, etc...)

Once I know how many VBD points are up for grabs I determine the total $ available to spend on VBD points. This is calculated by roster size and baselines. If you have a 12 team league and 14 roster spots then you know that there are 168 players that will be paid for at the end of the auction. Add up your "value" (above baseline) players. For the example of my league the number of "value" players would be 116 (17+32+31+14+10+12). Therefore I expect 52 players (168-116) to go for minimum value. If you have $200 cap and a $1 min bid you would have $2348 (12*$200-$52) in cap expected to be spent on "value" players.

Take the "value" players $ available figure and divide by TOTAL VBD points to get your rate of $/VBD pt.

Finally I add a column on the VBD application next to the modified points figures and multiply by this rate to determine the player's value in terms of $.

This method obviously works best with stable leagues with auction history available. I've found that my league owners behave very predictably in how many minimum $ players they will roster. This makes the calculations of value relatively simple as you can see above. If you're in a league with stable ownership, give it a try and let me know what you think of the values you get.

 

[Ariakis]

I just did a fudge factor to establish rough high/lows and then a 3rd column for total points. At the actual auction I used this as a tool to help make my own judgements and it seemed to work pretty well. I also added a few tiers but since players get nominated the strategy you intend to employ can change quickly.

 

[schu]

...but since players get nominated the strategy you intend to employ can change quickly.

Sorry I don't follow you here.

 

[Rocco]

I have been doing auction football for 16 years. The best method I have come up with is this:

1) Establish my VBD, rankings tiers etc.

2) Take the prices paid by position in rank order high to low from last year and apply them to my current year rank.

This gives me a rough idea of how i can expect bidding to fall out. Yes, sometimes there is a seismic shift in bidding, but generally the patterns fall the same. This works well, obviously, in a stable league.

After that, its the flow of the draft and finding that sweet spot of when the big guys are gone and people are not quite sure where the dollars are going next. Always a player or two that falls in there.

One nominating strategy that can work well is right away, i mean the first couple of bids, nominating the 3rd or 4th best RB. People tend to shy away from getting into a bidding war before a ceiling has been established. can save 10% or more.

 

[schu]

Rocco-

Since you have an established league you should try my method. The only assumption you have to make is that the baseline for each position is the same as past years. Then is will give you real VBD based $ projections for all the players above those baselines instead of you having to base the $ on last years position rank values.

-schu

 

[CalBear]

This method obviously works best with stable leagues with auction history available. I've found that my league owners behave very predictably in how many minimum $ players they will roster. This makes the calculations of value relatively simple as you can see above. If you're in a league with stable ownership, give it a try and let me know what you think of the values you get.

-schu

Note that this method gives you a prediction of auction cost, not auction value. That is, if everyone in the league tends to overpay for QBs, this method will tell you to overpay for a QB. I think it's more important to identify inefficiencies in spending; it's good to know what people are likely to spend money on, but I care more about what I should spend money on.

 

[Zasada]

This method obviously works best with stable leagues with auction history available. I've found that my league owners behave very predictably in how many minimum $ players they will roster. This makes the calculations of value relatively simple as you can see above. If you're in a league with stable ownership, give it a try and let me know what you think of the values you get.

-schu

Note that this method gives you a prediction of auction cost, not auction value. That is, if everyone in the league tends to overpay for QBs, this method will tell you to overpay for a QB. I think it's more important to identify inefficiencies in spending; it's good to know what people are likely to spend money on, but I care more about what I should spend money on.

Exactly. My league-mates always overpay for QBs. My model points that out quite well and I know where my potential bargains are going to be at other positions.

 

[Rocco]

Rocco-Since you have an established league you should try my method. The only assumption you have to make is that the baseline for each position is the same as past years. Then is will give you real VBD based $ projections for all the players above those baselines instead of you having to base the $ on last years position rank values.-schu

fair enough - i will post some results of my math later. Really should leave work first though, I suppose....

 

[schu]

Note that this method gives you a prediction of auction cost, not auction value. That is, if everyone in the league tends to overpay for QBs, this method will tell you to overpay for a QB.

I think it's more important to identify inefficiencies in spending; it's good to know what people are likely to spend money on, but I care more about what I should spend money on.

Value in an auction is determined by what is gained (in points projected differential) over the player that can be had for minimum cap as the baseline. Understanding who is the best player available for minimum cap is key to determining that. I believe that although this method returns value specific to your league the baselines can probably safely be estimated for new leagues based on number of teams and roster sizes. The "should spend money on" question is answered dynamically during the auction because this method gives VBD based accurate cost valuations.That doesn't mean I don't target players. I look for potential inefficiencies going into the auction based on ADP, owner history, and league specific rules. My point is that is DOES matter if a league tends to overpay for QBs because you will start a QB. Therefore you have to find relative value and this method gives you cost numbers that you can compare across positions.

 

[Dickie Dunn]

My league-mates always overpay for QBs. My model points that out quite well and I know where my potential bargains are going to be at other positions.

I've discovered the same in my league, and used the same method as Rocco to determine where I think the value is going to fall in my league (i.e. RB4-RB14; WR12-WR23, QBs overpriced). I'll see how it works in my auction this weekend.I've been toying with my nomination strategy, knowing that I would like to a.) flush out as much money as possible on QBs, and b.) target 2-3 RBs in that range I noted. I'm thinking of throwing out one of those RBs first, then hopefully get the QB spending going later and wait for value to drop to me. Key will be sitting close to last year's toilet bowl champ (first nomination).

 

[Rocco]

Schu,

Worked it through, including "inventing" some numbers for Pass, Run and Special Teams, which we draft independent of each other.

Of $350, 12 team 18 man roster:

Peyton comes in at $80

L Johnson at $122

T Holt at $54

I am confident Peyton will go for more, Johnson for less, Holt for probably a bit more. But, i like this. It has some rational and intuitive feel, and i will use it along side my spent last year by rank at the draft.

Thanks for the post and info.

Rocco

 

[schu]

Bumping this from last year.

This is my method for determining auction value above baseline based on VBD principles.

Will help answer any questions pertaining.