Page 1 of 1

CVRC Question

Posted: January 12th, 2013, 11:41 am
by aburt19
I am in a 10 team AL only 5x5 league. When I entered all the information into the league set up,
it showed a value for ERA of 4.052. The ERA for each of the 10 teams last year was

1st 3.452
2nd 3.590
3rd 3.682
4th 3.764
5th 4.122
6th 4.386
7th 4.420
8th 4.475
9th 4.646
10th 4.685

When I first started playing this game 20 years ago, I read that any pitcher who produced an
ERA lower than the last place team's ERA, that pitcher is producing positive value. But I'm
bothered by the big difference between what the CVRC shows and what the highest ERA last
season was. In the last 6 years, the lowest the last place ERA ever got was 4.52. Choosing
the number to use as a baseline for the ERA, WHIP and BA has always been difficult.

My question is what ERA should be used as a baseline? The same question applies to WHIP,
although the difference is smaller.

Thanks.

ETA: Sorry, I just noticed that topic was discussed in a previous post. You can ignore this.

Re: CVRC Question

Posted: January 12th, 2013, 12:20 pm
by Todd Zola
That's a good question -- it has been some time since I looked into it.

Regardless what is used, there will be a replacement adjustment, but the baseline still matters.

Due to this replacement, it's not necessarily cut and dried that worse than the baseline gets negative and better gets positive value - it depends where the baseline falls.

Also, leagues fluctuate AROUND that number -- though the ERA did look further off than normal variance.

For now, I'd plug in the last place and see what happens.

After I finish with this update, I'll find some 2012 leagues and see what the last place numbers were and do some runs as well.

Re: CVRC Question

Posted: January 12th, 2013, 12:32 pm
by Todd Zola
OK -- VERY preliminary as this piqued my curiosity

Tout Wars went from 3.1 to 4.4. A 12 tean AL only league I run went from 3.7 to 4.7.

That doesn't make a lot of sense, but it's what happened.

So I used a baseline of 4.6 for AL and the values barely changed (which is the important thing).

Re: CVRC Question

Posted: January 26th, 2013, 11:14 pm
by aburt19
I know that you are out of town this weekend, so there is no need to answer this until you have
time.

I noticed when I downloaded the hitting CVRC for the first time this year that the category weights
showed even weighting. I didn't think much about that because sometimes it comes up with
mixed league, 12 team, 15 team, etc. Today I read that you think the category weights should
be evenly weighted among the categories. It says that you think that it reflects the values
that players are going for this year.

That is a very dramatic shift from 2012 when the weighting IIRC was 1.25 for home runs,
1.10 for RBI, with BA and SB being weighted less to compensate.

I was curious as to what convinced you to make the shift. I think in the past it was based on
research that winning teams tended to do well in HR and RBI. Has that tendency changed? I
thought that the weighting last year was a little heavy toward HR and RBI and used 1.2 for HR
and 1 for RBI, with BA going to .95 and stolen bases gaining the rest.

I have had good luck with weighting it some without going to far. I'm trying to figure out
whether the same market forces that changed your mind are going to change players value
in an established league with a low rate of turnover among the owners.

Thanks.

Re: CVRC Question

Posted: January 26th, 2013, 11:25 pm
by Todd Zola
I'm actually back in town for the weekend but heading to the WWL in Bristol CT next week for the Fantasy Baseball Summit.

Winners still kill it in the HR category and worry the least about steals.

The category distribution for SB is still different than the others (lower threshold, higher percentage on top) but the curve has flattened some.

But, perhaps because there are more steals now, people are paying market value for them.

The CER (Category Efficiency Ratings, driving force for the weightings) were as much market driven as anything. Players were always "worth" the value with weighting being equal - it just didn't cost that much to get them steals so if we kept them at 1 across the board, our users would be well short in power and overloaded in speed unless they made the adjustment -- so we made it for them.

That adjustment is no longer necessary.