I know that you do not want to give away the secrets behind "Joe's Secret Formula", but when might you want to use "the worst starter method" instead of Joe's secet fromula?

Baselines always get a little tricky... If in doubt, stick with Joe's Secret. It handles most leagues. In the beginning of DD (and VBD Spreadsheet), this was the only one available.

Some people are perpetual tweakers by nature... and wanted more control over the baselines. So we added the Worst Case Starter and the ability to create your own baselines.

The Worst Case Starter method is a standard one, so it is included with the DD. It is fairly easy for people to understand this one... but I'm not sure it is all that sound. Generally, if I'm going to use it... I use it as a starting point. For example, I'll enter the worst case starters into the "User Configurable by Position Rank" and then tweak the values up and down. For example, if you start 1QB, 2RB, and 3WR, etc. In a 12 team league, you would enter QB=12, RB=24, WR=36. Now if you wanted to increase the values of the RBs, you would increase this number, for example, RB=30. If you wanted to decrease the values of the QBs, then you would decrease this number, for example, QB=8.

One method that I've been playing around with this year (Yea, I'm a perpetual tweaker as well)... is setting the baselines by position rank based on how many players typically get drafted at that position. For example, if you start 1QB, 2RB, 3WR, 1TE, 1PK, and 1Def. and your league of 12 teams drafts 18 rounds. On average (you may have to look back at the past couple of years draft). You may find out that owners typically drafted 2QB, 5.5RB, 6.5WR, 1.5TE, 1PK, and 1.5Def (These numbers should be roughly equal to the number of rounds). In this case I mutiply each of the values by 12 (number of teams) and set the baselines as follows: QB=24, RB=66, WR=78, TE=18, PK=12, Def=18. With the limited number of samples that I've tried, the results look pretty good to me. I also like it because if you look at the VBD numbers... all of the draftable players have a positive number. If you start to draft a guy with a negative number... then that throws up a red flag... that you are drafting outside of the typical range.

There is also a discussion or two reqarding Baselines in the Shark Pool: Discussion of Baselines

Bruce,
I just calc'ed baseline for my league using the last for years. Here they are: RB 62, WR 76, QB 30, PK 21 and DEF 22. This makes things seem ok, except that DEF is in the top 10 picks. Would you expect this?

14-team, 15-round draft. 1/25 and 6 pts ru and Re. 1/50 and 4 pts Pa.

Def Scoring for Defensive Categories
DFR - Defensive/ST Fumble Recovered (ID/DT/DST) 1 point
DTD - Total Defensive and Special Teams TD 6 points
Int - Interceptions 1 point
PA - Points Against, Total Points Scored 0 - 0 PAs = 10 points
1 - 10 PA = 5 points

SACK - Sack 1 point
STY - Safety 2 points
YDS - Yards Allowed 0 - 99 YDSs = 9 points
100 - 199 YDSs = 6 points
200 - 299 YDSs = 3 points

No, that seems a little high for a Defense. We've had several people mention that the Defensive fantasy points seem a little out of whack, so we are revisiting all of the Defensive calculations in this next release.

Found this in the Shark Pool and made me want to ask what am I supposed to use for my auction league?


"I think it's also interesting to think about the difference in proper baseline for auction vs. draft leagues. "Joe's Secret Formula," based on how many players might be gone in the first 100 picks, isn't directly applicable to auction leagues."


I am entering my first auction in August and it is a dynasty. If I F this up I am going to be miserable for years!

I just looked back through the last 5-years of DEF scoring. THe top DEF scored between 175 and 200 points (3 of the years between 196 and 200). DD has the ravens listed at 248 points for my scoring system.

Is the 248 points with version 6.0c?

The Auction amounts should be pretty close. Have you tried setting the Baselines to "All draftable players" as mentioned above?

Was wondering the same thing.

6.0b. Have not had a chance to update yet.

Updated to 6.0c - still a little concerned. Ravens listed as the number one defense with 204 points scored. Ravens three year average is only 124, seems like an awfully big jump. I don't know if it will help, but here are the top three DST for the last three years:

2004 - Bills 196 - Patsies 178 - Steelers 171
2003 - Patsies 171 - Ravens 170 - Dolphins 146
2002 - Bucs 199 - Eagles 156 - Panthers 141
3 yr AVG - Bucs 153 - Patsies 149 - Dolphins 143 ......... Ravens 124
6.0c - Ravens 204 - Buff 198 - Patsies 192

The low end does not look much better. Our low score last year was 60 points (Raiders). The DD projected low score is 105.

Can you derate your scoring a little bit? For example subtract 1 from all the Points Against scoring categories and subtract 1 from all of the Yards against scoring categories and see if they get closer.

I'm intrigued by this, and I have a couple of years of draft history to use. We have 14 roster spots and keep two players per team; should I set the baselines to account for 14 roster spots or 12 draft rounds? (Or does it even matter?)

From what I understand, this baseline should be set for the total number of players selected. We keep one player, and I include those in my total. For you, I think it would be 14 rounds x 14 players (12 drafted and 2 keepers) = 196 total players. You would have to then determine what proportions of the players drafted fit into the various rosters spots (i.e. 30 QB, 65 RB, 90 WR etc.). If I understand correctly, this should set the VBD for the last players drafted to 0, rather than having negative numbers for players drafted after the "Starters" (i.e. QB=14 RB=28, WR=42 etc.).

You must be half right. 14 rounds/roster spots x 8 teams in the league = 112 total players.

Thanks!

Yea, try it based on 14 rounds and let us know how it turned out.

I have cut the scoring by more than half. My range is now 158 to 85. Seems too low on the high end, and too high at the low end. Yardage seems to make the biggest difference. Even cutting the scoring by as much as I did only reduced the low end by about 10 points.

I was reading this and the other thread, and I thought I picked up last year that Joe's formula used a baseline of first 100 players drafted. (I'm not trying to out the secret formula, but just assuming it is something like that for the moment...)

For those new to tweaking baselines, the important insight here is that Joe's formula is for first 100 (or whatever) players drafted in a "normal league," not necessarily in "yours." I've found that looking at players drafted in the first 100 picks in MY league works great as a baseline (though I'm thinking about reducing that, because that's about 8 rounds in a 12 team league, and I've usually abandoned VBD after round 7 at the latest and am picking on position needs.

Has anyone used a smaller than normal VBD baseline cutoff? (i.e., I'm considering using 84. I know that will create more negative numbers, which I'm comfortable with, since I'm ignoring VBD down in the 7th round and later.

However, does anyone have any thoughts about any anomalies that would result that you think I should avoid?

I've been reasonably happy using the positions taken in the top 100 baseline, but am always trying to improve things.

I have been playing around with the baselines and ran into some weird results compared to Joe's formula. The first thing I did was use the actual number of players taken in my draft last year (12 teams x 16 rnds=192). The results are similar to Joe's method but the most obvious difference is that Manning was moved from 5th (in joe's) to 1st. Can somebody explain the reasoning for this?

The second thing I tried was just putting the number of players taken in the first 8 rounds (96 in total). In the first 8 rounds of my draft there were no kickers taken so I put that in as 0. This resulted in the kickers all being shoved up. The first kicker is taken in the 7th spot and than every kicker in the league is taken over the next 4 rounds. Does anyone know what I am doing wrong?

I got the same thing. Change the kicker to 1 and then ignore that kicker in your drafting.

I think it must be like dividing by zero or something. (I guess if you divide Vinatieri's 148 points by the 1 BL player's points - V's 148 points, you get 1, which puts him way down. Divided by zero BL the app doesn't know what to do with.)