Quality Starts vs. Wins - Which statistic more accurately reflects a pitcher's effectiveness?

Quality Starts vs. Wins

Since Daniel Okrent’s memorable plane flight in 1980 when he created the idea of Rotisserie baseball, the statistics which fantasy baseball players use to measure the success of pitchers has mostly unchanged. Wins, Saves, ERA and WHIP have been used as the main four categories to judge a pitcher’s value to their respective fantasy team. Although those four categories covered enough initially when fantasy baseball was created, due to the rise of sabermetricians, or people who attempt to measure success in baseball through objective evidence, other statistics have been created to measure a pitcher’s overall “worth” to a fantasy team, the most notable being quality starts (QS). The statistic was developed by sportswriter John Lowe in 1985 while writing for the Philadelphia Inquirer, and is as "a game in which the pitcher completes at least six innings and permits no more than three earned runs." 

Cases can be made for both statistics in terms of their viability of use for a fantasy league. Generally speaking, the players with the most quality starts are also those that obtain the most wins on a yearly basis. So what is the benefit to using one more than the other?

The case for W over QS:

Technically speaking, a pitcher who throws six innings and allows three earned runs every start would be a valuable commodity in a league which uses QS as a category. This pitcher in question however would also be sporting a 4.50 ERA, which in the eyes of most analysts is entirely too high to be considered effective (generally speaking an ERA of under 3.20 is considered the ideal). Due to the fact that a pitchers ERA can be much higher than what is considered to be a reasonable standard and still provide worth, some experts believe that a QS is too easy to obtain, and argue that the definition of a QS should be changed to 7 innings rather than 6, or 2ER rather than 3. Even though I personally agree that allowing an earned run every other inning isn’t particularly effective per se, since the statistic has been around for almost 30 years, I doubt that it will change overnight.

QS’s are also much easier to obtain than wins. As an example, ESPN’s 2012 pitching projections forecasted that 14 starting pitchers would accumulate more than 24 QS during the baseball season. Those same projections didn’t place any pitcher above 19 wins (Roy Halladay and Justin Verlander were tied at 19). Since QS’s are easier to obtain than W’s, it is also easier to find a pitcher off of the waiver wire who may help with that category. This in theory would make it easier for a fantasy manager to build a successful staff of effective starting pitchers, while allowing them to focus more specifically on hitting categories.The amount of wins that one MLB team’s pitching staff may accrue for a baseball team is a finite number (generally speaking 40-60, depending on the success of the team. Phillies had ~ 65). Since a pitcher can be credited for a QS even in the event of a loss, there will naturally be a higher number in the market.

Also, the QS category somewhat works against pitchers who tend to go deeper into games but may allow an additional earned run. From a statistics viewpoint, a pitcher who throws a complete game (9 innings) but only allows 4 earned runs has an ERA of 4.00, lower than the 4.50 standard for a QS, but they will not earn a point in the category due to allowing too many earned runs. This counteracts the notion that a QS is an accurate assessment of a pitcher's effectiveness.

The case for QS over W:

The easiest argument against using W as a category is fairly straightforward. A pitcher can perform brilliantly and not be credited with a win, and can be absolutely dreadful and squeak out a win. Case in point? Opening night of 2012 Felix Hernandez pitched against the might of the Oakland Athletics lineup in Japan (yes, that was sarcasm). King Felix finished with the following stat line:

8.0 IP 5H 1ER 0BB 6K

Essentially he held the other team to one earned run all game, but didn’t get credit for a win because Seattle’s offense is one of the worst in the league. On the other hand, we can take a look at the flip-side of the coin. Ivan Nova had a recent start against the Los Angeles Angels of Anaheim where he posted the following stat line:

6.0IP 8H 4ER 2BB 8K

Nova went on to get credit for the win, despite allowing 3 more ER in 2 less innings and not pitching nearly as well.

Since a pitcher can theoretically throw very poorly and still accrue a W, (after all, as long as you give up less runs than the other pitcher, you’re in line for a W if you pitch more than 5 innings), it isn’t a very good measure of a pitcher’s overall effectiveness.

Baseball writers have also started to shy away from using W as the be-all, end-all pitching statistic. The easiest example of this would be examining the Cy Young voting. Over the course of 2001-2008, the average number of W for a Cy Young winner in the American League was 20.75. In 2009 the trend bucked downwards when Zach Grienke of the Kansas City Royals won it with only 16 W, but his other statistics (2.16 ERA, 242 Ks) were incredibly dominant. In 2010 Felix Hernandez won the Cy Young with only 13 wins, but with a 2.27 ERA and 232 Ks. Due to the fact that fans and writers have begun to acknowledge that W’s are (for the most part) out of the pitcher’s control the majority of the time, the practice of solely using W’s to measure a pitcher’s effectiveness is being discarded.

In my estimation, Wins more effectively measures the prowess of a team’s hitters rather than the pitchers itself. Quality Starts more accurately measures the skills of the pitcher. To provide an example, if a team wins 11-10 and at the end of the game both pitchers go the distance, the “winning” pitcher could have a 9.0IP 10ER statline. Is that truly effective? Not necessarily. QS more accurately measures what the pitcher actually allows to happen due to him either performing well or poorly.

Quality Starts also takes into account a team’s defense as well, since it is only based off of earned runs. Unearned runs won’t count against a pitcher. Not all teams are blessed to have a stellar defense like the Tampa Bay Rays, some have to put up with bonehead outfielders like Logan Morrison behind them who turn routine “can of corn” fly balls into an adventure. Since errors are something which the pitcher has no inherent control over (other than his own errors, of course) this is another reason why QS’s more reflect the pitcher specifically, rather than the team as a whole with W’s.

Additionally, a pitcher may be unlucky enough to be going against a fellow ace like Halladay or Verlander when they are locked in. Trying to pitch against someone who goes 8.0 IP and only allows one run or two isn't the fault of the pitcher, its just dumb luck. The opposition has an effect on a W, but not a QS, which more measures the pitcher.

Conclusion

Bottom line? Although it is easier to accrue a quality start than a win, and although the definition of what is truly “quality” or not could use some tweaking, using quality starts instead of wins is a more precise means of interpreting value and success for starting pitchers effectiveness in fantasy baseball. Wins are a highly random endeavour dependent upon luck, defense, a team’s offense and the opposing pitcher’s outing. Quality starts measure purely how many runs a pitcher allowed over the course of an outing, which is a true measurement of how effective they were in that outing, and in a season.