CDG Pitcher Values Wonky

[vegascoop]

Something's broke or I'm doing something wrong. I put in the following categories and values:

settings.jpg

And I get the following top pitchers:

results.jpg

Magnuson,Trystan for $66?? Jenks,Bobby $65??

[loddar]

Choosing "Hits Allowed", "Walks Allowed" and "Earned Runs" as fully weighted pitching categories will do this.

[vegascoop]

Since my draft is fast approaching, which catagory(s) should I remove or re-weigh to create the best draft list? If I should re-weigh them, what's the best settings?

Thanks.

Also is this a bug or expected behavior?

[RAY@HQ]

The approach you're using doesn't work. If you have aggregate negative categories like walks, hits allowed, etc., the pitchers who are going to score well in that system are those that have low totals in those categories, obviously... but those are generally pitchers that have few innings.

Please describe your league's scoring system so we can answer the question re: how best to use the tool.

[vegascoop]

So I should make those three categories -1? Here is the scoring systems for my Yahoo league:

Scoring Type: Rotisserie

Roster Positions:	C, 1B, 2B, 3B, SS, OF, OF, OF, Util, Util, SP, SP, RP, RP, P, P, BN, BN, BN, DL
Batters Stat Categories:	Runs (R), Hits (H), Home Runs (HR), Runs Batted In (RBI), Stolen Bases (SB), Walks (BB), Hit By Pitch (HBP), Total Bases (TB), Batting Average (AVG), On-base + Slugging Percentage (OPS), Outfield Assists (OFA)
Pitchers Stat Categories:	Innings Pitched (IP), Wins (W), Saves (SV), Hits (H), Earned Runs (ER), Walks (BB), Strikeouts (K), Holds (HLD), Total Bases Allowed (TB), Earned Run Average (ERA), (Walks + Hits)/ Innings Pitched (WHIP)

[RAY@HQ]

No, you don't want to make them -1. We already know more hits are bad. You're going to want to cut the weights, as suggested above. Try making the negative categories .33 instead of 1, something like that. Play with it a bit... but that should get you closer to the outputs you expect.

[vegascoop]

OK, I can do that.

And please consider the following two items "constructive" criticism:

1. The default still seems pretty messed up. I mean Magnuson is OK for the three negative categories, but is absolutely awful for the SIX other categories. And yet CDG prices him at $66. Why are those three categories outweighing the other other SIX?

2. While its great you can tweak the CDG, by default you should be able to just put a check-box by the categories in a particular league and get a decent list. If the default for something like earned runs should be .33, then put the default as .33, not 1. You're asking me to tweak it until the output is what I "expect." I "expect" these tools to give me a good base list before I do any tweaking.

[RAY@HQ]

There is no elegant way to create an accurate default for your scoring system, sorry. In fact I get so many questions about these "negative" categories that I'm strongly tempted to go in the other direction and remove the support for them from the tool.

The reason that 3 categories outweigh 6 is a reason of degrees. Magnuson is ridiculously awesome in the 3, for no reason other than playing time.

[loddar]

And please consider the following two items "constructive" criticism:

1. The default still seems pretty messed up. I mean Magnuson is OK for the three negative categories, but is absolutely awful for the SIX other categories. And yet CDG prices him at $66. Why are those three categories outweighing the other other SIX?

2. While its great you can tweak the CDG, by default you should be able to just put a check-box by the categories in a particular league and get a decent list. If the default for something like earned runs should be .33, then put the default as .33, not 1. You're asking me to tweak it until the output is what I "expect." I "expect" these tools to give me a good base list before I do any tweaking.

I think you are misunderstanding the correct use of the CDG. You really shouldn't be married to the numbers that it spits out. This is a classic case of GIGO ("garbage in, garbage out"). HQ, rightly imo, often suggests to tweak numbers so that useful data goes in and useful data comes out.

Take Magnuson. The reason why those 3 categories weigh heavily is because he is not just "OK" in them. You would have to consider him outstanding in hits allowed, walks allowed or earned runs and therefore miles ahead in valuation terms when compared to any decent SP with 100+IP because that SP could not make up the valuation gap in the other six cats (and it's probably only 4 because I wouldn't describe him as "absolutely awful" in WHIP or ERA).

I also disagree with making the default weight any other number than 1. Ray probably suggested .33 so as to make the three negative categories count as one. But .25 or even .10 possibly could be more useful in your league setup. You're the only one who would know that when you run the numbers.

[loddar]

In fact I get so many questions about these "negative" categories that I'm strongly tempted to go in the other direction and remove the support for them from the tool.

I agree. The confusion they create outweigh any benefits derived from them.

[vegascoop]

OK-- I'm starting to get it. It's a somewhat of a defect in the way my league is keeping score rather than with the tool. (Since there is NO IP requirement it could be a good strategy to start pitchers who never pitch.) I'll play around with it. I may just get a baseline without including the negative categories at all.

Thanks for all of the help.