Thursday, February 12, 2009

Dumping Steals

Eugene still doesn't like the idea of dumping steals:

If you went for those guys- excluding Quentin and Huff and Ludwick on the respective lists b/c they weren't predictable, you would likely also be sacrificing BAvg. Dumping steals and going middle of the road in BAvg at best isn't going to do much good on offense.

Thome, Cust, and Giambi all hurt BAvg quite a bit. Howard, Dunn, and Burrell do that to you on the other side as well. That's 1/3 of the list. It's hard to buy the three category studs on the list and still compete in pitching.
I think that's a fair point (and suspected that Eugene or someone else was going to bring it up). I would add as well that by cherry picking the best hitters, you are going to come up with outliers like Quentin, Huff and Ludwick - guys who were great last year but certainly couldn't be counted on to automatically be Top 10 finishers in the category.

What is worth doing is going back and looking at the top teams in each non-SB category in your league and trying to ascertain the stats you would have needed to buy.

In my A.L. last year, 228 HR, 970 RBI and a .2817 BA won the non-SB categories. Per hitter, this came out to 16 HR and 69 RBI per hitter. Now you're not looking at Carlos Quentin or Ryan Ludwick anymore. You're looking at Lyle Overbay.

Overbay's average salary last year was $13 and he turned a $1 profit. That's also a reasonable baseline for expectations, and once again eliminates guys like Quentin or Huff.

Eugene is correct in one very important area. You will need players like Huff or Quentin to make a category dump work - either through your freezes your through your auction. The paradox of a category dump is that you're dumping categories in the first place because you don't have enough value in your freezes, but you're deciding which categories to dump based on the freezes you have.

Make sense? If not, let me try explaining it another way.

Let's say you have 10 freezes at $80 of salary and $120 worth of value in a league with a 20% inflation rate. The $180 you have left to spend would give you $150 of return in the auction. That's good for a $270 team.

The team with the best freeze list has $120 worth of salary and $185 worth of value. The $140 he has to spend would give him a $117 rate of return and a $302 team.

Now Eugene would argue that this is all academic anyway.

dumping saves, b/c they are more concentrated and not necessarily correlated with any other skill (like ERA/Ratio), can work out a little better.
(Actually, there is a strong correlation in Roto between the top save guys and the value they return in ERA/WHIP, but that's a subject for another day.)

I've never argued this point. Saves is the first category I look to dump if I don't have the closers to compete.

But here's another Roto paradox. Because Eugene and I certainly aren't the only owners who have figured this out, closers go for less money in auctions, set-up men who eventually become closers are severely undervalued, and there are plenty of cheap closers floating around on Freeze Day.

If I'm in the position where I have two cheap closers, I'm not going to dump saves and will look to dump another category. The next most logical category is steals.

In 2008, 35 hitters who were bought in A.L.-only auctions stole 10 or more bases. I probably wouldn't eliminate all of these hitters from consideration (Nick Markakis and his 20/87/10/.306 would have definitely remained on my list because - as Eugene also correctly points out - average is a big part of my non-SB strategy), but even if I did, that would still eliminate only 20% of the hitting population.

By comparison, not bidding on closers is eliminating bids on 13% of the pitching population. Not as much, but still quite a bit.

I'd agree with Eugene that dumping saves is a more logical strategy if you're going to dump a category. But dumping steals can and does work in competitive leagues.

1 comment:

Roll2 said...

If you are going to dump steals, you better nominate and go get the valuable MI who don't steal much, because there are only a few of them, and once your league mates figure out what you are doing, they likely will make them pay a premium for them.