Custom Draft Guide - What happened?!

[cardboardbox]

A week or two ago I generated custom dollar values for my league using the custom draft guide and they are much different than what I'm getting today in preparation for this weekend's auction. My league is 5x5 mixed 14 team 280 cap. 14 batters, 9 pitchers, 7 bench. 69/31 split. Top players, fractional, do not force positions. I usually go with 0 reserves just for the draft guide.

Two weeks ago I got 35.60 for Pujols. Today he's 60.96? HUH?!

I think I'm going to restart my computer... I think the draft guide was screwy last year too and I had to restart to get it to work right.

[RobR]

We released an update to dollar values on Wednesday night. The highlights are:
--The salaries of the number of players to be drafted will sum to the league budget
--The number of players above replacement level will total to the number of players to be
drafted
--Elite players in averaging stats will see a value increase. Elite players in lumpy
counting stats (SV, SB) will see a value decrease.

[cardboardbox]

so its expected that Pujols would go from 35 to 60 using the same projections and the same league parameters?

[RobR]

I expect Pujols to move a lot. A $15 move with your league parameters doesn't surprise me. When I run him on the old system, using our stats from Tuesday I get $44 in your league, so he'd already moved $9. Are you sure the $36 is a Top Players number?

I'm not saying a $15 move is small, but Pujols is the number one mover under most scoring systems. Here's some reasons why.

1) His AVG is 15 points above the #2 projection and 20 points above #3. A 20 point advantage in AVG over full season's AB is valuable and we were undervaluing it before.

2) We've lowered the value of SB guys and that's the only category Pujols isn't elite in.

3) You're using top players which magnifies the changes of elite players.

4) Your league is relatively shallow which raises the replacement level. Elite players will move more the higher the replacement level.

Do you have output from an older valuation run? It could help me explain.

[cardboardbox]

I expect Pujols to move a lot. A $15 move with your league parameters doesn't surprise me. When I run him on the old system, using our stats from Tuesday I get $44 in your league, so he'd already moved $9. Are you sure the $36 is a Top Players number?

I cant say with absolute certainty. I am fairly certain I was because when I didnt use top players the values were even further from what is realistic in my league. But if I did not, the current balanced value for Pujols is 51 which is still $15 above what it was previously.

I'm not saying a $15 move is small, but Pujols is the number one mover under most scoring systems. Here's some reasons why.

1) His AVG is 15 points above the #2 projection and 20 points above #3. A 20 point advantage in AVG over full season's AB is valuable and we were undervaluing it before.

2) We've lowered the value of SB guys and that's the only category Pujols isn't elite in.

maybe I missed it, but could you point me in the direction of an explanation for this change in SB value?

3) You're using top players which magnifies the changes of elite players.

4) Your league is relatively shallow which raises the replacement level. Elite players will move more the higher the replacement level.

Do you have output from an older valuation run? It could help me explain.

I have a spreadsheet from a few weeks ago. A few things that I notice now about the old values... The top 196 batters are worth $90 less than they should be and there are an additional 45 players with value above 0. Because I am not 100% certain of the parameters I used 2 weeks ago, it would be a waste of time to research it further. But it is clear to me and others that there was a big change and we arent sure what to make of it.

[RobR]

There was a big change for about 10 players per league. We believe all the new values are a better representation of player value.

In the old system, we didn't always accurately select the proper number of players to have value above replacement level. In lumpy categories like SB or Sv, this lead to too high of a replacement level being selected. An increase in replacement level increases the value of players who are above average in a category. It can have a large effect on elite players. When we corrected the player selection, we corrected the replacement levels, which lowers the values of above average players in SB.

Of course the missing $90 also contributes. It would have been distributed over all players, but concentrated in the elite players. Not $15 to Pujols of course. Also it's worth noting that in the old system under your settings, there were three players above $0 and under $2. There should be more and that would shift dollars to top players.

It's possible that if you used to use top players because the elite were undervalued by us for your league that you no longer need to use it.

[teddyballgame]

[RobR]

We recently added a replacement salary input and defaulted it to $1. That is likely the source of the change (and why it was important to add the option). We used to use a replacement salary of $0, which overstated marginal salary for most leagues.

As for your second question, we have two inputs to address those. If your budget per team is $130, you should enter that instead of $260. If your minimum salary is $0.50 instead of $1, you should enter that as replacement salary.

When I try your exercise, I get 21.87 for Pujols. 21.87 x 2 = 43.47. Would you mind double checking?

[teddyballgame]

Rob, checked and verified, perfect.

Thank you, and also thanks for answering a follow up question I hadn't even asked about what that "replacement salary" spot was all about.

BTW, HQ should put that info in the CDG "explanation page."