Getting to Know Stats: Pitching WAR

Washington Nationals v New York Mets

By Darren Helley

Regular statistics in baseball can only tell half the story. As much as home runs, earned run averages and runs batted in can wow us on the stat sheet, they don’t tell us the extent or value of the contribution of a certain player towards his respective team. The last ten years of baseball has seen the rise of non-standardised sabermetric statistics that aim towards judging a player’s overall value, one in which lies well beyond the regular numbers.

Arguably the most popular and most recognised player value statistic in baseball, WAR calculates a player’s amount of additional wins for his team above the number of expected team wins in which that particular player is substituted by a replacement player. The perfect statistic to identity value for money within any given player in baseball. The 2014 NL MVP and Cy Young Award Winner Clayton Kershaw scored an impressive 8.0 WAR to lead the majors, meaning that the Dodgers won nearly 8 more games with Kershaw on their team than if he were to have been replaced.

WAR is also the chessboard for the face off between respective baseball statistical giants. Baseball-Reference and Fangraph both offer different variations of the increasingly popular statistic; comparing the two together does generate some very interesting, glaring differences as they estimate offensive, defensive and pitching values differently. Felix Hernandez scored a 6.7 WAR in 2014 through the eyes of Baseball-Reference, but that value plummets to a mere 6.1 with Fangraph.

The primary factor for Pitching WAR is FIP due to its neutral nature of the statistic. Fielding Independent Pitching takes away the influence of defence towards a Pitcher’s value, thereby giving a more honest reflection of his ability. But to transition FIP into the WAR formula, we must turn a pitcher’s FIP value into a win total through a percentage and generate both a league and replacement average of the same nature for it to be compared against: not the most simplest of tasks. The transition can be broken down into four steps:

.Replacement Level: The replacement level-average for wins percentage always stays at .380 for Starting Pitchers and .470 for Relief Pitchers. The average is higher for Relief Pitchers due to their short-inning nature, thus having a higher expectation to deliver every outing compared to a Starter. The FIP-average varies every season, which is connected to the replacement level average for wins percentage and appropriately increased to match. Therefore, if the league average is at 4.4 FIP in the American League, the replacement level-average for FIP would be set above that mark for a Starting Pitcher with a .380 win percentage (approx. 5.6 FIP). It’s worth noting that this would be the replacement average for that league and not for the entire major leagues.

.Run Scale: FIP is converted into a runs scale, simply by dividing the value by 0.92. The main reason for this is to take the value of a Pitcher’s FIP and attempt to simulate how many runs allowed he commits, rather than simply match it with his ERA (which is what FIP is designed for).

.Park Adjustments: As previously warned, WAR is park adjusted. This means that the replacement level-FIP average for a Starting Pitcher will dictate based on what ballpark he pitches in. Fangraph is arguably the best system of WAR in terms to providing a justified and fair park adjustment scheme, for it uses a five-year average regressed park factor rather than relying on a potential one-year fluke.

If the FIP for the replacement Starting Pitcher was at 4.8 and a Pitcher’s respective ballpark depressed offence by approximately 2% over a five-year period, then the replacement level-FIP for that ballpark would be 4.705. However, it is important to remember that a Starting Pitcher will play approximately half of their games at home, so they would only need their FIPs adjusted by only half of their home park factor.

.Run Environment: Probably the most troublesome and bitter stage of the Pitching WAR calculation, a Pitcher’s Run Environment can be influenced by both how well he pitches, but also how well his offence produces on every 5th day. Note: such calculation eliminates “clutch” from a Pitcher’s performance, therefore neutralising it.

The Seattle Mariners currently have a Runs Scored-average of just 3.68, yet Felix Hernandez is 4-0 and sporting an ERA of 1.82. Simply put, if an elite pitcher is on the mound, his team will not need as many runs as with a replacement Pitcher in order to win games. Hence, a Pitcher can influence his own run environment parley a lower Runs/Win conversion rate. The typical average for Runs/Wins is 10 for a replacement Pitcher, so a Pitcher like Felix Hernandez will turn in a lower average given his run environment. The conversion formula is as such:

((League RA + Pitcher’s RA)/2)+2)*1.5 

These four stages, from generating a replacement level-FIP to converting it into a run scale, influencing it via ballpark factor and then discovering how well a Pitcher’s Runs/Win scale is finally leads us to the overall value of his WAR.

Let’s use King Felix as an example once more; his 4.28 FIP is approximately 0.5 runs better than the league average. We need to translate this into a win%. Dividing the difference from the league average by the Runs/Wins average (influenced by Ballpark factor) will give us the additional amount to add-on to the average-Pitcher win%. 0.5 divided by 10 is 0.05, therefore we add 0.05 to .500 (this would be the average-Pitcher win%, not the replacement level). Felix’s win% would be .550, so we now subtract this amount from the replacement level-average (.380) to calculate how many more wins by percentage he fairs better than the replacement: 0.17.

The final calculation is the wins percentage above replacement (0.17) multiplied by the total from the formula ‘innings pitched/9‘. The King would pitch 200.67 innings in this scenario. Thus, the final calculation looks like this for Felix Hernandez:

.170 x (200.67 / 9)

This calculation produces a total- Felix’s Pitching WAR- of 3.8: his WAR value from the 2008 campaign. This breakdown merely covers the main highlights of a Pitcher’s WAR and the stages of constructing the final formula, which revolves on achieving a Pitcher’s wins% above replacement.

Part Two of this WAR coverage will see a more in-depth look at how positional players’ WAR totals are calculated.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s