|ESPN.com: Draft Kit||[Print without images]|
Ask the average person to measure a pitcher's skill with a singular statistic. How many would spit his earned run average, or ERA, back at you?
Even in today's numbers-heavy game, long-standing measures such as ERA remain in the forefront of many people's minds; familiarity exercises much influence.
Here's the problem: ERA might be an accurate measure of what has transpired during a specific, past period of time -- most frequently a single season -- but it is one of the most misleading tools for forecasting future success. And forecasting future success is all that matters in fantasy baseball.
ERA's value is in how it accounts for the two most critical skills of a pitcher: recording outs and preventing runs. It is also one of the five primary Rotisserie categories, necessitating an ERA projection for each pitcher.
At the same time, ERA's failing is its inability to account for external factors that influence the final number. Even the Wikipedia page for earned run average hints at its misleading nature, saying that "because of the dependence of ERA on factors over which a pitcher has little control, forecasting future ERAs on the basis of the past ERAs of a given pitcher is not very reliable and can be improved if analysts rely on other performance indicators such as strikeout rates and walk rates."
It's for that reason that when formulating projections, we must consider the context of individuals' ERAs. For example, Kyle Lohse had the majors' eighth-best ERA in 2012 (2.86), but due to his uncertain status for 2013, he's projected for a 3.63 ERA, representing one of the largest increases for any individual this year. We do this accounting for a variety of factors detailed below: Team defense, left-on-base percentage and ballpark factors are the ones most relevant in Lohse's case.
And those are not the only three factors that influence the ERA category. Here's a closer look at those and some others you should consider:
Perhaps there is no greater influence on a pitcher's ERA than the defense behind him. That doesn't refer only to earned versus unearned runs due to errors and passed balls; that includes range, or a defense's ability to convert batted balls into outs. A team with poor range -- hello, Cleveland Indians -- will afford more hits and thus result in pitchers affording more earned runs than one stocked with players who cover considerable ground.
Sure enough, of the five teams that arguably sported the worst defenses in 2012, using statistics such as FanGraphs' Ultimate Zone Rating and Baseball Info Solutions' Defensive Runs Saved -- the Indians, Colorado Rockies, Houston Astros, Detroit Tigers and New York Mets -- four ranked among the 11 worst in team ERA, and those teams had five of the 16 worst individual ERA qualifiers in the game. The Tigers, thanks primarily to immensely skilled starting pitchers as well as a staff that struck out the fifth most hitters in the game (1,318), were the outlier: Their 3.77 team ERA was 10th best. Having Justin Verlander sure helps!
Let's return to the Indians for a moment. A casual onlooker might note that they, as a team, sported the majors' ninth-best fielding percentage (.984) and ninth-fewest errors (96) last season, giving them an appearance of a middle-of-the-road-to-above defense. However, upon further inspection, the Indians had the game's worst UZR (minus-57), and their minus-51 Defensive Runs Saved -- a positive number is the goal in that department -- was third worst.
Future prospectors, however, need account for the Indians' winter moves: The team unloaded its worst defender, Shin-Soo Choo (minus-17 UZR, minus-12 DRS in 2012), while acquiring two of the game's better performers in the field, Michael Bourn (22.4 UZR, 24 DRS) and Drew Stubbs (6.8 UZR, 2 DRS), and moving Michael Brantley to a more suitable home in left field. The Indians' outfield defense -- also accounting for the addition of the at-least-league-average Nick Swisher in right field -- should be substantially improved, which in turn might benefit the team's ERA.
It's for that reason that when evaluating individual pitchers, consider using one of several statistics that measure skill independent of defensive factors. There are many: dERA (Defense-independent ERA), CERA (Component ERA), DICE (Defense Independent Component ERA), SIERA (Skill Interactive ERA) but among the most popular are FIP (Fielding Independent Pitching score, on an ERA scale) and xFIP (Expected FIP), easily found on such websites as FanGraphs.
FIP is a strong measure of a pitcher's performance regardless of how his fielders fared behind him, counting only numbers upon which he himself exercises influence: home runs, walks, strikeouts and hit batsmen. The xFIP, meanwhile, "normalizes" home runs -- remember that such factors as ballpark and weather can influence that -- calculating the formula using the league average of home runs allowed per fly ball.
Pitchers whose FIP or xFIP varies greatly from their ERA might have either benefited from particularly good or bad defense, or particularly good or bad luck, or even both. That's why it's important to examine how the pitcher got to his final ERA; use FIP and xFIP to see if there's an obvious explanation.
Be sure to consider a pitcher's history in terms of FIP/xFIP, however, as a select few individuals have exhibited an ability to succeed in ERA despite a poor track record in FIP/xFIP. Jeremy Hellickson, for example, has now posted back-to-back seasons of FIPs greater than 4.40 but ERAs of 3.10 or better. Sure, the Tampa Bay Rays' strong defense behind him has contributed, and should continue to do so in 2013, but it's also possible that Hellickson is one of those rare examples of the pitcher who routinely outperforms his peripherals.
These were the 2012 leaders and trailers in both FIP and xFIP:
Adam Wainwright's ERA-FIP differential, at +0.84 the third highest among ERA qualifiers, is the most glaringly obvious example in the "fluky" category. In his first full season since recovering from Tommy John surgery, the noted curveballer struggled with his command initially before making major strides during the second half of the year. To that end, Wainwright's FIP after the All-Star break was 2.82, almost identical to the 2.86 number he posted during his career year of 2010 (20 wins, 2.42 ERA, 213 strikeouts). There is therefore an excellent chance that his 2012 Rotisserie statistics were an aberration and he's due to improve in them in 2013.
Hinted at in the previous section, and somewhat related in fact, unearned runs actually have zero impact upon ERA in the literal sense. But unearned runs are the product of shabby defense -- they are the result of errors and passed balls -- and while a pitcher doesn't get charged for many of the runs he allows after such a mistake is made, for evaluation purposes perhaps he should. (To be fair, perhaps he should also be credited for the outs he'd have recorded if not for the error.)
Remember, recording outs is the A-number-one priority for a pitcher, and his skills should be honed enough that he's capable of recording four -- sometimes more -- outs in a single inning when needed. Pitchers who allow themselves to be unraveled easily by errors might possess flawed skills; put them in front of a faulty defense and those mental lapses might become further exposed.
These were the 2012 leaders in unearned runs, the chart including what percentage of the team's unearned runs that pitcher allowed, as well as the number of times opposing batters reached base via an error:
McAllister's unearned runs/reached on error splits are most distressing; be aware that a whopping seven of his 2012 total unearned runs occurred in a single game, last Aug. 6 against the Minnesota Twins. On that day, he surrendered three hits and a walk after his Indians committed a two-out error, a performance far more deserving of criticism than his ERA for the day indicated. A candidate to open the year as the Indians' No. 4 starter, McAllister's performance above further illustrates his mediocre talent, meaning that any gains he experiences as a result of the team's defensive improvements might be negated by his limited skills.
Sometimes called "Strand Rate," such as in Ron Shandler's Baseball Forecaster, this measures a pitcher's ability to prevent runners he allows to reach base from scoring. Along with BABIP and home run/fly ball percentage, left-on-base percentage is one of the most popular tools fantasy owners use to measure "luck." Typically, a major league-average left-on-base percentage is 70-72 percent, but last season's 72.5 percent league rate was the highest since the turn of the century, and it represented the fifth consecutive year in which the number increased.
Here's where left-on-base percentage comes into play as an ERA influencer: It is as much an indicator of a few lucky -- or unlucky -- bounces when a pitcher had runners on as it is that a pitcher's skills are greater either working out of the set position (typically with men on) or the windup (typically bases empty). Make sure to examine a pitcher's such splits before reading too much into this category.
These were the 2012 leaders and trailers in left-on-base percentage, the chart including each pitcher's OPS differential between when he was pitching with the bases empty and pitching with men on base:
Hellickson has now turned in back-to-back seasons of at least an 80 percent left-on-base rate, and a quick glance at his minor league performance in the category reveals a similar such pattern throughout his professional career. One of the reasons he's so successful in that department is his outstanding performance pitching out of the set position; his .225/.271/.352 triple-slash rates allowed with men on base results in a .623 OPS that ranks 14th best out of 144 qualified starting pitchers the past two seasons combined. Couple that with the Rays' outstanding defense and Hellickson looks a lot less "fluky" than people often claim.
The bullpen behind a starting pitcher can also influence his ERA more than you might think. You'll see a starter's final stat line in the box score, but what you might not see was that he left the game with two outs and the bases loaded, only to have the reliever who replaced him serve up a grand slam on the very next pitch. Those runs are charged to the starter, significantly hurting his ERA, and while they're somewhat deserved -- he did allow those baserunners, after all -- the greater the volume of those allowed, the more misleading the starter's seasonal ERA.
Baseball-Reference.com tallies a statistic it calls "Bequeathed Runners," or BQR, which refers to runners a starter leaves on base when he exits. These were the 2012 leaders and trailers in terms of percentage of bequeathed runners allowed to score, minimum 10 bequeathed runners:
Two pitchers in particular stand out here. One is Mike Minor, whose 63.2 percent rate of bequeathed runners allowed to score is strangely high for a pitcher on a team that had a 30.3 percent overall rate in the category as well as a bullpen that managed the majors' second-best ERA (2.76) and fourth-best rate of saves plus holds per opportunity (91 percent). That the Braves added further relief help in the form of Jordan Walden suggests that Minor is due greater fortune in this department, and it might neutralize some of the ERA regression he'd likely be due as a result of the correction he'll receive to his .252 BABIP.
The other is Anibal Sanchez, who, after watching only nine of his 62 combined bequeathed runners score in his first six big league seasons, saw 13 of his 21 bequeathed runners cross the plate in Year 7. Granted, Sanchez was rather fortunate in those previous six years, and it's not like his Tigers have one of the majors' best bullpens, but even a slightly better percentage in that department might help him in terms of ERA in 2013.
This one's as obvious as they come. A pitcher's ERA is impacted by his home ballpark -- or any ballpark in which he's pitching on a given day -- as a smaller ballpark leads to untimely home runs that inflate the number, while more spacious venues might afford a pitcher more leeway in terms of homers. This is one ERA influencer that fantasy owners have always accounted for; there is still a statistic that helps you do the work more quickly: Adjusted ERA, or ERA+. ERA+ calculates a pitcher's ERA assuming he worked all his games in a neutral ballpark, with anything over 100 considered good, anything beneath it bad.
These were the 2012 leaders and trailers in ERA+:
Four of the top six performers in terms of ERA and ERA+ were the same, the only difference being Matt Cain (sixth in ERA, 2.79, but 16th in ERA+, 125) and Chris Sale (11th in ERA, 3.05, but fifth in ERA+, 142), and naturally, Cain called a pitching-friendly ballpark and Sale a hitting-friendly venue his home.
Zack Greinke is a pitcher who stands out, as one of the stronger performers in terms of ERA+ (27th, with 114) and an individual who moves to a pitching-friendly ballpark for 2013. He'll also get a decent number of road games in other pitching-friendly venues within the National League West, including San Diego's Petco Park and San Francisco's AT&T Park, meaning that he's a strong bet to improve 2012's 3.48 ERA.
In conclusion, this isn't to say that any of the five influences described in detail above should be expected to radically shift a pitcher's ERA, but rather they can do so. When doing your 2013 draft preparation, therefore, examine what went into an individual's ERA, specifically these five things. Keep in mind that even one earned run can mean one-twentieth (or 0.05) of a run in ERA for a typical starting pitcher; five earned runs can mean as much as a quarter of a run (or 0.25).
That's why, when evaluating pitchers, I place ERA lower on my priority list than categories like WHIP, strikeout and walk rates. To return to the column's opening question, if you asked me, I'd nominate FIP -- not even a traditional Rotisserie stat!