clock menu more-arrow no yes

Filed under:

There are different ways to calculate WAR, and Blake Snell is noteworthy in all of them

New, comments

And what makes Blake Snell a noteworthy pitcher.

Kansas City Royals v Tampa Bay Rays Photo by Mike Ehrmann/Getty Images

During the game on Wednesday night Brian Anderson and Dewayne Staats started discussing Blake Snell’s season and where he ranked by WAR and there being different WAR calculations.

Anderson and Staats admitted they were unaware of the differences and how they are calculated. This is ok and something that many people don’t understand including some who throw around the numbers on a daily basis.

The three most commonly used WAR are provided by Fangraphs (fWAR), Baseball Reference (rWAR), and Baseball Prospectus (WARP).

For position players there usually isn’t a wide gap between the three metrics, because the offensive contributions of players are similar, although they use different defensive metrics. Fangraphs uses Ultimate Zone Rating (UZR) for all position players outside of pitchers and catchers which they use Defensive Runs Saved (DRS). Baseball Reference (rWAR) uses DRS for all positions. Baseball Prospectus uses Field Runs Above Average (FRAA). For catchers Baseball Prospectus’s FRAA has the most comprehensive defensive metric including pitch framing, but outside catcher there’s close alignment.

With pitchers, however, the methodology is completely different for all three.

Fangraphs (fWAR and RA9 WAR)

At Fangraphs fWAR is calculated based on Fielding Independent Pitching (FIP). FIP attempts to estimate the results the pitcher would get without giving credit or blame to the defense that plays behind them. So at it’s core it’s strictly a formula that is based on strikeouts, walks, hit by pitches, and homers allowed.

FIP = (13 x HR + 3 x (BB + HBP) - 2 K)/IP + “FIP constant”

The “constant” changes every season based on the run scoring environment in order to have FIP approximate the appearance of ERA, for easier comparison.

In the Fangraphs methodology, results on balls in play don’t matter unless it is a homerun. Due to this fWAR takes out things like the stat BABIP and other elements of defense in order try to compare only what the pitcher can control.

This version of WAR is the one quoted by Anderson and Staats, and here they found it hard to believe Snell is the 10th best pitcher in the American League. As of this morning Snell finds himself tied for tenth in the league at 3.0 fWAR. Chris Sale leads the AL with 6.1 fWAR.

With a sterling 2.07 ERA, 10th is not an undeserved placement: Snell’s 3.30 BB/9 is ahead of Charlie Morton’s 3.48 BB/9 of the top 10 pitchers, and his 10.36 K/9 ranks seventh among the 11 pitchers overall. Additionally, Snell’s 88.1% Left On Base Percentage (LOB%) of 88.1% is the highest and one of only three who has stranded 80% or more of base runners (Chris Sale and Justin Verlander).

But in terms of the WAR metric, Snell’s 3.29 FIP is the culprit here. Snell isn’t given credit for the .236 Batting Average on Balls In Play (BABIP) which is by far the lowest in this group (with Corey Kluber’s .262 BABIP coming closest) subtracting that element of team defense.

Snell’s 2.07 ERA can be attributed to the high LOB% and low BABIP referenced above. By comparison, Sale’s 1.97 ERA is the only qualified starter with a lower ERA.

It’s worth noting Snell also comes in at 139.0 innings, pitched which is the fewest of the group and at it’s core WAR is a counting stat. The disabled list stint and limited workload since his return will cost him innings and thus WAR.

But wait, there’s more...

Fangraphs also provides a Runs Allowed Per 9 Innings WAR (RA9 WAR) in the value tab, but that isn’t what is shown on the primary dashboard. RA9 WAR is pretty straight forward, and was discussed earlier today by Jared Ward. The metric reflects total runs allowed (both earned and unearned) per nine innings. Snell’s RA9 is 2.20 once you include the unearned runs.

Overall, Snell’s 5.7 RA9 WAR ranks him third behind Chris Sale (6.6) and Trevor Bauer (6.1). Strictly by runs allowed Snell is in the conversation for best pitcher in the American League this year, but volume prevents him from the top spot despite Sale and Bauer going to the disabled list recently.

Snell is only 7.0 IP behind Sale, and could pass him if the Cy Young frontrunner misses more than one more start on his current disabled list stint. Trevor Bauer is 27.0 IP ahead of Snell, but is going to miss most if not all of the regular remaining games this season. A lot of contenders for the Cy Young are missing time this year, which will only push Snell higher up the fWAR leaderboard.

89th MLB All-Star Game, presented by MasterCard - Red Carpet Photo by Patrick Smith/Getty Images

Baseball Reference (rWAR)

Baseball Reference uses runs allowed at it’s core. They integrate park adjustments for the pitching environment and adjust for the runs the defense they pitched in front of either saved or added.

Snell’s 5.6 rWAR ranks third behind Sale (6.1) and Bauer (5.9). These numbers are very similar to RA9 WAR at Fangraphs, because they start with runs allowed and the minor difference is due to different park factors (which are a separate calculation) and that B-Ref does give some credit or blame to the defenses they pitch in front of. Both elements help Snell in this case.

Baseball Prospectus (WARP)

Baseball Prospectus’s WARP is based of Deserved Run Average (DRA). DRA at it’s core is most closely related to fWAR, but it tries to take batted ball quality into account instead of just assuming that all BIPs that aren’t homers are created equally, giving their stat a more nuanced approach.

DRA is the brain child of Jonathan Judge, Harry Pavlidis, and Dan Turkenkopf. In 2015 they discussed their methodology in a piece at Baseball Prospectus.

These are the factors that DRA attempts to control for:

  • The overall friendliness of the stadium to run-scoring, accounting for handedness of the batter (using our park factors here at Baseball Prospectus);
  • The Identity of the opposing batter;
  • The identity of the catcher and umpire;
  • The effect of the catcher, umpire, and batter on the likelihood of a called strike (e.g., framing / umpire strike zone, from 1988 onward);
  • The handedness of the batter;
  • The number of runners on base and the number of outs at the time of the event;
  • The run differential between the two teams at the time of the event;
  • The inning and also the half of the inning during which the even is occurring;
  • The quality of the defense on the field for each individual play (assessed through BP’s FRAA metric);
  • Whether the defense is playing in their home stadium or on the road;
  • Whether the pitcher is pitching at home or away;
  • Whether the pitcher started the game or is a reliever; and
  • The temperature of the game at opening pitch (from 1998 onward).

Given these factors Snell has a 2.63 DRA. Based on this he has put up 4.28 WARP which ranks behind Justin Verlander (5.65), Chris Sale (5.50), Gerrit Cole (5.46), Trevor Bauer (5.46), Corey Kluber (5.15), and Luis Severino (4.63) in the American League.

Quantity once again limits Snell’s ranking here. With the exception of Sale (146.0 IP) and Snell (139.0 IP) all the pitchers above him have thrown 159.1 or more innings. In the world of Baseball Prospectus, that hurts your score, and for good reason.

In the end, WAR is a counting stat.

Whatever WAR you are using innings thrown matter. So while Snell’s ERA looks very shiny he has thrown the 24th most innings so far this season at 139.0 IP. You can’t earn WAR by not playing.

So depending on what you are looking to measure Snell has been the tenth most valuable starting pitcher in the American League if you want it to be measured by fWAR, third most valuable by RA9 WAR and rWAR, or seventh most valuable by WARP.