Thursday, March 22, 2007

crystal balls and pasta on walls

I think the empirical analysis Lee puts forth on SP strategy is hard to refute, but I'm going to give it a shot anyways.

There are two issues with targeting too many of the top 10-12 SPs:

1) The bidding & competition is too fierce. The prices get inflated. I love Santana but could only go to 15.5/3 yrs/NT on him in our league & the winner bid 17.5/3/NT. It is not as crazy as that when bidding for some of the other top SPs, but still in the 10-12M range, and I will only ever pay even that much for Santana.

2) How consistent is the group of top 10-12 SPs from year to year? Do the names stay the same, or does it vary? Aside from Santana, I don't think there is an SP who has been top 10 in GDR for the last 3 yrs though that is an off the cuff assertion (but you get the point). I loved to see the bidding on Zambrano this year as I think he'll drop to top 30 status.

So then the question becomes, is there a way to predict who might rise to the top 10 or 12, and can you get them cheaply (at least relatively speaking)? I think the answers are yes and yes. I also think it has to do with how scouting services like Baseball Prospectus and Baseball America project future performance. What they do is say "What are the CHANCES that pitcher X will put up a top 10 SP season?" We could define that as 18+ wins, 200+ IP, 200+ Ks, and an ERA in the 3's, and a WHIP of 1.2o or better.

So I budget a certain amount to spend on SPs, and if I get Santana, great, if not I don't usually bid on a top 10 guy from last year, avoiding the "red ocean" feeding frenzy. I'm looking for Blue Ocean here (read W. Chan Kim's book, good stuff). Then I have a list of 15-20 SPs that, by analysis, I think are the best bets to jump into the top 10 group. Some of them are known quantities coming off an off year or injury (think Felix, Rich Harden). Some of them are young up & comers with some risk attached (think Cole Hamels and Daniel Cabrera). What they all have in common is excellent peripherals & history (either in MLB or minors) which says they can put up those lofty numbers at some point in their career. It's up to me to decide if I think that year is coming now/soon.

So in the rest of my budget, I might get 2-3 SPs in the Harden/Felix mode, & pay around 5-7M for them, then I might get 2-3 more in the Hamels/Cabrera mode & pay 1-4M for them or even find a few, as Lee did below, in the FA market. Then, if I've been smart, I've also got at least 4-5 guys signed cheaply (maybe in the prospect draft, maybe not) who are on 2 or 3 yr deals and only cost .5-1M who may or may not give me ROI now but will likely do so for next year. I think that mix, if I judge talent & opportunity well, pretty much guarantees me 3 top 10 SPs and 5 top 30 SPs, or very close.

The other thing to keep in mind is having a stockpile of SPs gives you a lot of flexibility in matchups (as noted below) and also you can squeeze more productivity out of your rotation than the face value would indicate.

What I mean by the latter is, for instance, last year I bought Hamels & Liriano when they were relative unknowns. Liriano started in the pen, and Hamels was brought up later in the year & had a few bad starts before locking in. Neither of them finished in the top 10 SPs overall based on total pts, however, when I put those guys in my lineup, they averaged over 20 fpg for me and led me to the playoffs.

The combination of matchup efficiency, depth at a position where injuries often occur, and ability to optimize who has the "hot hand" can lead to a very high ROI for the "throw it at the wall and see what sticks" strategy, and a lot of its success is dependent on the abilities of the chef to read his pasta well...which is a little easier than crystal-ball reading, right? ;-)


leehaak said...

Great analysis Quick. I'm going to post back later about my 2006 top 20 projected pitchers, and how they actually did.

I believe forecasting SPs is more accurate than you've given it credit for, because most of the 'precursors' are factored into the GDR scoring. SPs with good peripherals tend to score better in GDR than in, say, rotis scoring.

I don't object to the strategy of purchasing low cost potential, but many of these pitchers (Hamels, Bonser, Wainwright, Hill, Garza etc) were fetching prices in the five or six million dollar range in several of my leagues. When I can routinely grab Sabathia for a couple of million more, I'm going to do it every time. This should be a lively, on-going debate, which makes it fun.

Qksilver said...

Yep, agree on the pricing. I saw Hamels going for too much in some cases, but usually Wainwright & Hill were in the 3-4M range for the leagues I was in. It's all supply & demand at that point. That said, I failed to get Sabathia in every league I'm in, he routinely went for $8M which is not horrible but I'd rather have to of the upside guys at $4M each in the end. As you said, really good debating & I think the answer will be there's more than one way to skin this cat... ;-)