Welcome to the Casual Sabermetrician. If you are anything like me, you’re interested in advanced baseball statistics but all the formulas and your lack of an advanced degree in mathematics leave your head spinning. The good news is, these Saber stats are always being tweaked and new ones are always being invented, so it’s not too late to get in on the ground floor on some and to use this knowledge to your advantage on Draft Day. Let’s dig in.
Our first example will be to look at starting pitching. In most, if not all leagues, hitters are valued over pitchers. One major reason is that pitchers are much less consistent and harder to predict than hitters. Why is this? It’s simple, once bat meets ball a lot of what happens next is out of the pitcher’s control. The statistics we have to address this are grouped under the heading DIPS (Defense Independent Pitching Statistics), these have been around since 1999 and were invented by Voros McCracken. If this sounds familiar, it’s because you read about it in Moneyball. Stats that have been used the past decade include Defense-Independent Component ERA (DICE) and the more widely used Fielding Independent Pitching (FIP) invented by Tango Tiger. The Hardball Times has their own version of FIP, but it is Tango Tiger’s that is used on most sites, most notably Fangraphs.
But the stat I want to look at more closely in this post is tERA – True Runs Allowed (also known as TRA). This is a much newer stat invented by Graham MacAree of Stat Corner. It seeks to improve upon FIP and xFIP (Expected FIP) by taking into account the performance of balls in play.
From Stat Corner:
The hope for tRA then was to construct a metric which takes into account every action a pitcher is responsible for and turns those numbers into runs and outs based around a highly logical and transparent mathematical framework.
For fantasy purposes – how can we use tERA to help us in our upcoming draft? By looking at the delta between ERA and tERA and looking for outliers. This is a simple exercise and you can do it at home. On FanGraphs – go to Leaderboards for Pitchers and select Advanced. Export this chart to Excel. Open the file and add a column and call it ERA-tERA. Create a formula to subtract the tERA column from ERA and then sort. Let’s see what stands out…
Justin Masterson (0.89); Kyle Davies (0.88); Chris Narveson (0.78); James Shields (0.78); Jeremy Bonderman (0.72)
Clay Bucholz (-1.59); Trevor Cahill (-1.04); Tim Hudson (-1.03); Rodrigo Lopez (-1.01); Jaime Garcia (-0.93)
Let’s analyze the above. Masterson looks to regress to the mean more than most but is useful only in the deepest or AL only leagues with his extreme BB/9 rate. Davies is in the same boat as Masterson. Narveson is interesting and could be a good sleeper pick in deep leagues on an improved Brewers team that is all in this year, he is going undrafted in most mockdrafts. Shields is where this list gets interesting. Big Game James has struggled since his and the Rays breakout season in ’08. Shields has suffered from a lot of bad luck the past 2 seasons including an unsightly .341 BABIP in ’10, but his peripherals are all good – his K/BB ranked 10th in the majors in 2010. The Rays don’t look to be contenders, but look for Shields to improve in 2011. His current ADP is 160 (43rd SP) so he could also come as a big time bargain in later rounds. Bonderman – eh, he’s undraftable. Another pitcher of note is Zach Greinke – his delta is (0.66) and moving to the aforementioned Brewers and the NL, it’s easy to imagine Zach rekindling his ’09 mastery. His ADP is 47 (8th SP) so you might want to reach to ensure you get him, but if he slips to the 4th or 5th round in your draft you better pounce.
Buchholz’s delta jumps way off the page at a whopping -1.59 runs. His 2010 peripherals were below league average so he is due for a major dose of reality in 2011. He is being drafted 91st (25th SP) ahead of guys like Roy Oswalt and Shawn Marcum – I want no part of Buchholz’s return to reality and neither do you – let the uninformed over pay for him. Trevor Cahill is another good young SP who is due for a major regression – his incredible league lowest .236 BABIP is just not sustainable. He’s being drafted right after Buchholz – pass. Hudson’s low K/9 rate and .249 BABIP also signal a regression in ’11. At ADP 140 (38th) he’s not quite as risky but look at picking up Shields instead. Lopez is undraftable, Garcia’s high walk rate is cause for concern and his impressive rookie campaign will yield to the inevitable sophomore slump, he’s only worth a late round flyer in deep or NL leagues.
I hope this first installment of the Causal Sabermetrician was helpful, I’ll see you in the next episode where we’ll turn to the offense.