Wednesday, March 18, 2009

Starters vs. Relievers

Eugene Freedman is lucky enough to be in a league with no IP requirement, and is pursuing a no starting pitcher strategy. But what about the rest of us who have to get to a certain number of innings or lose all of our ERA/WHIP points?

Part of the allure of the no SP strategy in no IP-requirement leagues is that relievers are better in ERA and WHIP than starting pitchers. The idea is to blow off wins, hope that one of your cheap relievers turns into Brian Fuentes and/or Brandon Lyon so that you can compete in saves, and win ERA/WHIP by a lot. This type of pitching staff would be super cheap (maybe $30 tops if you bought a middling closer) and you could then spend $230 or more on your hitters and clean up in the hitting categories.

In most leagues I've played in, there has been a minimum IP requirement of at least 1,000. That makes grabbing nine middle relievers impossible, since you'd buy 650-700 IP in a best case scenario, and trading for an additional 300-350 IP would mostly nullify your offensive advantage.

A strategy that I've liked over the years that comes close to this is a low wins/saves strategy, which I've also called the three SP strategy. The idea is to buy two or three starters, six or seven middle relievers, and hope that you can finish middle of the pack in wins through league attrition via dumping and luck into a mid-season closer or two with one of your middle relievers. It's worked well for me on more than one occasion.

All of the theory above doesn't take into account how much more reliable relievers are in ERA/WHIP versus starting pitchers. How did they do in 2008?

2008 Auction Rosters: Starting Pitchers
American League:
9,348 IP, 586 W, 2 SV, 4.21 ERA, 1.345 WHIP.
National League: 10836 1/3 IP, 660 W, 1 SV, 4.15 ERA, 1.351 WHIP.

2008 Auction Rosters: Relief Pitchers
American League:
2257 1/3 IP, 109 W, 495 SV, 3.77 ERA, 1.323 WHIP.
National League: 2236 1/3 IP, 119 W, 529 SV, 3.92 ERA, 1.337 WHIP.

Those numbers surprised me when I computed them from the N.L. and A.L.-only leagues that I tracked during the 2008 season. I went back and double checked my formulas in Excel just to make sure I wasn't miscalculating.

The only significant difference was in the A.L. between SP and RP ERA. All of the other qualitative pitching categories were a virtual wash. The WHIP for relievers is certainly a little better, but not enough to make up the gap if a top team lucked into a Cliff Lee or a Ricky Nolasco last year.

This doesn't necessarily nullify a low IP strategy. But you do have to be just as careful with the relievers you buy as with the starting pitchers you buy. And this means that you might have to spend more money to get the relievers you want - depending on your league's spending patterns.

Don't assume that buying mostly relievers is going to automatically buy you a strong ERA/WHIP team. Last year it might not have.

8 comments:

Anonymous said...

Last season, I was killed with this strategy. I think if you're going to reasonably pull this off, you'll need two ace-type pitchers who can essentially ensure good WHIPs and ERAs to offset any stumbles you might have from your No. 3 or from a middle-reliever who suddenly can't find the strike zone. But, I guess if you went with two stud starters, that probably negates the money-saving strategy.

As a side note, I once was in a draft league without an innings-pitched requirement. I went with all closers and middle relievers, and I won the league by 10 points because I focused entirely on offense--and traded away closers for hitters once my lead was impossible to catch.

Thanks for the great posts.

Gypsy Soul said...

Say hey, Mike. How goes it? Just wondering your opinion re playing in an NL only, 5x5, with just 12 teams. How does this effect bid prices? Do you just reduce the prices by the percentage of fewer teams compared with a 13 team league? ie about 8%. Or do you allocate more money to the top players? Thank you.

Dr. Hibbert said...

I play in a 12-team AL league with a relatively low (900) IP minimum. With one exception, every champ over the last 10 years has executed a low IP ERA/Ratio run with some saves thrown in (essentially a LIMA plan). With a low IP minimum, 2-3 starters are all you need to draft--you can pick up extra innings along the way via FAAB or trade. I find that in most leagues not only are middle relievers VASTLY undervalued but also readily available in the free agent pool.

bkharr said...

I am thinking of trying this in 2 leagues. League 1 pitching freezes Lee 3, Litsch 3, bj ryan 17. League 2 freezes Litsch 8, Devine 1, Ray 5, Rodney 1. If I try this should I pursue a balanced attack or studs & duds?

Toz said...

Interesting discussion, and one that my league-mates and I, and Mike and I, have discussed on numerous occasions.

In my own mind, the low IP strategy works from a salary/budget perspective if you have significant value locked in at closer or starter (or, if you are really lucky, both). For example, league 2 that BK puts out there has, assuming Ryan grabs his job back after the first month or so, two excellent closer freezes. Add a stud pitcher to the mix, grab a third cheapy and fill in, and spend the balance on your offense.

As for studs and duds versus a balanced attack, take a look at my next post on Fantasy Baseball Advice.

Anonymous said...

Eugene balked at trying his strategy out, and traded me Sherrill (I have a cheap Ray) for Buehrle at 10 and Saunders at 1. Hi Eugene. On this topic, about 7 years ago I was involved in the startup of a NL 5x5 keeper league, 12 teams with a 7-man reserve. The guy who won punted wins and K's. Not only that, but he told another owner, who at that point ran a popular fantasy baseball site (known mainly because of its very active forum) and played in expert leagues, what he was doing, and the response was that it was impossible to win doing so in 5x5. After he had won he offered to write an article for the site, but the owner, who finished in the middle of the pack after drafting a pedestrian team and did't return the following year, did not take him up on his offer. This guy contends that the strategy of going all reliever in 5X5 leagues is only viable in redraft or the first-year keeper leagues.

Anonymous said...

its bkharr-prob w/my google password thanks for the input, guys. In response to your questions, both leagues 4x4 old school. League 1, mature, moderate inflation, hitting freezes Hamilton 31 Aviles 7, R Sweeney 5 possibly Lowrie 5. League 2, third year, immature, low to moderate inflation, hitting freezes Barajas 1, R Sweeney 5.

Anonymous said...

IP requriement 1000 both leagues