Monday, September 07, 2009

Building A Better Pitching Staff

posted by Jon Williams
In the old days of Fantasy Baseball you couldn't pick up a Fantasy Magazine or Guide of any kind without reading that pitching was completely unpredictable. If that was ever true, it isn't any longer --at least not to the extent that it used to be. Using newer statistics and advanced indicators has given us the ability to make better informed decisions when it comes to pitchers but in particular starting pitchers. The more information we are able to gather the better our decisions. Unfortunately for most owners, the strategies being used have not changed with the times. I think it is very much the right time to do so.

We can now nail who the really successful starters are and separate them from the flukes and illusions that used to confound fantasy owners of the eighties and early nineties. I am not saying we'll nail their category totals but we can hit within an acceptable range enough of the time that predicting which pitchers we can safely bid full price to acquire is fairly easy to do if still a bit time consuming. Where owners still mess up is in predicting sleepers. It is still very difficult to discern when young pitchers are taking a step up in production (Edwin Jackson), when a veteran's new pitch or attitude will result in a incredible turn around (Joel Pineiro, Cliff Lee) from the young starters who had a little lucky streak at just the right time (Manny Parra) or from the veteran who had a fluky good season (Kyle Lohse). It's difficult but not impossible. The real question is whether it is actually worth investigating and investing in players like this.

Finding hitters to make low cost investments in is considerably easier. An increase in playing time is often enough to turn scrubs like Jake Fox and Drew Sutton into the next Russell Branyan and Ben Zobrist. Does anyone dispute that it is easier to predict Branyan and Zobrist than Jackson and Pineiro? Not many will I am certain. So, why do we fantasy owners still prefer to invest in cheap pitching sleepers more so than the equivalent hitters? We are all trying to fill out our lineups with as many of the Carlos Lee, Adrian Gonzalez, and Dustin Pedroia class of player as possible and then devoting less than a third of our budgets on pitchers. I suppose some will argue that pitchers are more at risk to become injured. That's true but a great hitter is often just as difficult to replace as a top pitcher, perhaps even more difficult. There are always safe relief pitchers available in even the deepest of leagues.

A huge percentage of fantasy owners have adopted LIMA-like strategies that call for most of an owner's draft-day budget to be spent on hitters with a minimal amount (usually $60 of a $260 budget) spent on pitching. If you intend to purchase saves this strategy basically forces you to buy the Pineiro/Parra/Lohse types and hope for the best. I have decided that this is a road to ruin. It requires too much luck especially with half the owners in your leagues chasing the same types of players because they are utilizing essentially the same strategy. You are far better off chasing the top pitchers (try to get at least three of them for around $25 each) and at least one closer (as cheaply as possible) and looking for your sleepers among the position players. If you fill out the rest of your staff with veteran, highly skilled set-up men for just a few dollars each you will be far better off than buying the high risk starters that qualify as sleepers.

The top starters you buy have to be very carefully vetted. You want to eliminate any candidates that are showing signs of decline, injury, or of becoming a trade target. Don't let huge names become your targets. Jake Peavy is a big name but now that he's in Chicago I want no part of him. Roy Oswalt is showing signs of decline in his stats? Avoid him. Scott Kazmir is plagued by minor injuries? Let someone else draft him. Pick experienced starters with high K/BB, K9, and GB% and low BB9, LD% and HR/FB rates and that play on good solid teams (it doesn't need to be the best team) with good solid defense (it doesn't need to be full of gold glove candidates, just avoid the disaster defenses. Be very strict when you make out your list of candiates. It should be a very short list. Now, if you find a pitcher that fits the above criteria who for some reason has not had the expected success you can add him to your list but make him a low priority that you'll draft only if he goes for the right price.

Now, with $100-110 spent on your pitching staff you'll need to work harder to fill out your lineup but you'll find that with hitters there are far fewer absolute statistical disasters. Spend your money on at-bats. I try to draft a stolen base anchor in the infield and a 30/30 outfielder. Then fill out your lineup with the cheapest at-bats you can find. This year easy to project at-bats came at a low price from players like Aaron Hill, Russell Branyan, Brett Gardner, Ben Zobrist, Travis Hafner, Nelson Cruz, Marcos Scutaro, Cesar Izturis, Miguel Montero, Everth Cabrera, Scott Hairston, Jerry Hairston, and Josh Willingham just to name a few. It can be done if you are willing to put the work into it.

I received this e-mail (edited for brevity) this week from a reader named DeWitt that asks some very relevant questions and raises some good points:
A couple of weeks ago you sent us a link to an initial analysis someone at Elite had done with its fresh-off-the-presses all-leagues database. That person's first-glance deductions were 1/ active owners do better, and 2/ 'streaming' doesn't work. The first seems self-evident. The second, though, contained a leap of logic I can't let rest.

Specifically, it was pitching that he looked at. Trying to catch the hot-hand/playing match-ups/churn-and-pray --- it doesn't work, he reasoned, because those teams which seem to have engaged in it end up with bad numbers --- high ERA, WHIP, etc. His conclusion may well be spot-on. His reasoning isn't.

Streaming causes bad numbers, he says. I say the reverse is every bit as true: bad numbers cause streaming.

Think about it. You start the season with six weeks of disastrous pitching, fall to 13th in a league of 12, savor your 6.78 ERA for a second and, buddy, you're splashing around the free agent pool fishing for Pelfreys. And Mr- I've-Got-LIncecum-and-Greinke? He's dating super-models, not working the waivers.

You see my point. Early success discourages streaming. Early failure may require it. It's self-fulfilling. Elite's data-dude has, I'm afraid, confused cause and effect.

It's like saying people who aren't born into money are less successful at accumulating it. Blind people are more likely to shop for underwear while walking a dog. Or, because women tend to get pregnant after sex, women wanting to reproduce should seek out partners with high sperm counts.

Cause and correlation. Two entirely different things.

More productive, perhaps, would be to separate streamers into two sets --- those who employ it as a purposeful strategy from Opening Day, and those who begin later, out of desperation. Which group is larger? Is either successful? One more than the other?

Among the latter group, is it possible to compare the panicky (streamers) to the patient (active non-streamers) and recommend one strategy over the other? Or, among the purposeful streamers, is there anything to be deduced from those who employ the strategy well?
I think the reader has a very good point about streaming. Most of those doing it are acting out of desperation rather than desire. If you had a great pitching staff already, constantly churning it to squeeze out a few extra wins and strikeouts is not going to do anything good to your numbers in most leagues. Because the best pitchers are already rostered. Those who go into the season planning to stream pitchers are usually trying to get by with a low cost pitching staff. Whatever streaming does for your Wins and Strikeouts it usually makes up for by destroying your ratios. Starting the Brett Tomko's and Parra's of your league just because they're facing the Padres in San Diego is not going to help anyone but your opponents. Maybe the example is a bit extreme, but how many great pitchers are available for streaming purposes? Undeniably, there are some who stream their way to success but I do not believe that makes it a strategy worth emulating unless you are in an extremely shallow league. If you're in anything deeper than a standard 12-team mixed, I would avoid the strategy. To be clear, there is a huge difference between streaming and maximizing your current pitching staff by effectively using your bench options.

My primary point is that you can avoid the need for streaming by drafting a better initial pitching staff. I don't have much of an opinion on whether streaming is ethical or not. I think if you don't like it you should put rules in place to prevent it. For more opinions on streaming you can check out these articles while I move on to other things...

Roto Arcade: Streaming Ethics

True Guru: Streaming Pitchers
TGIF: The Ethics of Streaming Pitchers
Todd Zola: Strategy of Streaming

Under Not Fantasy Sports but Still Awesome!

Dan Brown, author of the Di Vinci Code and Angels and Demons, has a new book coming out called The Lost Symbol. The new book also features the historian detective, Robert Langdon. Dan brown writes books that are almost impossible to put down. Everyone you know will be talking about it very soon. The even greater news is that Amazon is offering an insane discount on the book right off the bat. You can get the Hardcover edition for just $16.17 which is 46 percent off the cover price. If you buy through one of my Amazon links you'll also be helping out Advanced Fantasy Baseball. In fact, if you buy stuff from Amazon.com you can always do so through the portals on this site and help me out while doing your shopping as usual.

Labels: , ,

IF YOU LIKE IT, SHARE IT!

0 Comments:

Post a Comment

Subscribe to Post Comments [Atom]

Links to this post:

Create a Link

<< Home