T10B Busted: WAR (Wins Above Replacement)

Synopsis: Sabermetics has taken over how baseball teams are constructed, how games are managed, and how players are evaluated. Whereas stats like HRs, RBIs and BA used to delineate players, MLB general managers now seem to be obsessed with stats like OBP, OPS, and WAR. I fully appreciate the importance of statistics when evaluating players, but I question the current over-reliance on them. In this post, I hope to discount the importance of wins above replacement (WAR) when evaluating greatness.  


T10B BUSTED: WAR (WINS ABOVE REPLACEMENT)

At the 1/4 mark of the 2018 season in late May, I read an article by ESPN.com’s Sam Miller titled, “Mike Trout is on pace for the greatest season in MLB history.” No offense to the author, but he provides a telling example of when the numbers fail the eye test. As a caveat, I believe statistics serve a purpose because they help us avoid a subjective bias. In this case, however, his numbers are misleading.

Miller wrote the following paragraph in his article.
 
The game’s best player [Trout] was on pace to produce 14.2 wins above replacement. It’s almost unimaginable — nobody has cracked 12 WAR in a half-century, and no active player has ever WAR’d higher than 2016 Trout’s 10.5 — but it’s time to take seriously the possibility that we’re watching the greatest season of all time. 
 
I agree with Miller’s assessment that Trout is the best player in the game today.  However, it seems that the ESPN.com reporter got overly excited when he argued that Trout might be having the “greatest season of all time.”
 
Initially, I dismissed Miller’s argument as suffering from a sample size fallacy. While 40 games encompasses a meaningful portion of the season, it’s unfair to extrapolate a full year of production based on only 25% of games played. Upon further review, I realized Miller’s argument fell short more so because of an over-reliance on WAR as the basis to define greatness. 
 
As support, I offer Trout’s extrapolated 2018 season in comparison to the two other candidates for the greatest season ever. Specifically, I offer Babe Ruth’s 1921 season and Barry Bonds’ 2001 season. 
 
GREATEST MLB SEASON – (FOR A POSITION PLAYER)
 
Player Year Games Hits HRs Runs RBIs BA
Ruth 1921 152 204 59 177 168 .378
Bonds 2001 153 156 73 129 137 .328
Trout 2018 154 157 46 125 92 .294
               
Player Year OBP SLG OPS OPS+ OWAR WAR
Ruth 1921 .512 .846 1.359 238 12.2 12.9
Bonds 2001 .515 .863 1.379 259 12.4 11.9
Trout 2018 .440 .632 1.072 193 10.7 13.4
Source: Baseball-reference.com
 
Notes:
 
  • The bold numbers reflect the best performance out of the three players. Ruth and/or Bonds dominated Trout in every category but WAR. Apparently, the hypothetical replacement player is much worse today than either 20 or 100 years ago. I fully understand that Bonds’ season deserves an asterisk given the impact of “The Cream” on his numbers. Then again, WAR presumably accounts for PEDs by comparing players from the steroid era with juiced up replacement players.  
  • I prepared this table about 1 week after Miller’s article. Trout didn’t have a good week, which could account for the extrapolated WAR of 13.4 versus the 14.2 Miller referenced. Then again, I found at least three different calculations for WAR so he may have been using a different source altogether. I don’t know about you, but I have an issue with stats that are not purely objective.
Assuming Trout had maintained the same level of production, he would have finished the 2018 season with a .294 average, 125 runs, 92 RBIs, and 46 home runs. While those numbers are impressive, they don’t jump off the page as the “greatest season ever.” Yet, somehow Trout would have had the highest WAR ever.
 
As further support for my concern with WAR, I offer the following passage from Miller’s article 
 
May 12: 0-for-2. Probably my favorite hitless Trout game of the season so far. He drew four walks, but the sequence I love is in the 11th inning. With Zack Cozart on first as the potential winning run, Trout grounded it to the third baseman, who threw out Cozart. That’s a failure. But Trout’s power had the third baseman playing deep against him, which, combined with Trout’s speed, made it impossible for the Twins to turn the double play. Trout beat the throw to first. That’s a tiny, tiny, tiny bump to his WAR.

 

Let me recap in case it’s not obvious. Trout’s WAR increased because he grounded into a fielder’s choice instead of a double play. I understand that causing one out is a better outcome than causing two. Still, it’s a negative result that should have hurt, not helped his WAR. The statistic apparently assumes the hypothetical replacement player would have hit into a double play.

2018 MLB SEASON: MID-AUGUST

Now that the 2018 MLB season has reached the 3/4 mark, it’s reasonable to start talking about which players deserve consideration for individual awards. Specifically, a sufficient number of games have been played to avoid a sample size fallacy. With respect to the AL MVP, the leading contenders seem to be Red Sox outfielders J.D. Martinez and Mookie Betts. Martinez leads the league in hits, HRs, RBIs, and total bases while Betts leads in runs, batting average, slugging, and OPS. The only player who might be able to ruin the Boston MVP party is Mike Trout, who reached the 1/4 mark with the “greatest season of all time.”

The following table highlights the extrapolated production of the three players in contention for the 2018 AL MVP

2018 AL MVP CANDIDATES
Player Year Games Hits HRs Runs RBIs Avg
Trout 2018P 150 158 41 113 82 .309
Martinez 2018P 153 196 50 117 140 .330
Betts 2018P 135 186 36 134 85 .350
               
Player Year OBP SLG OPS OPS+ OWAR WAR
Trout 2018P .459 .624 1.083 195 9.9 10.7
Martinez 2018P .401 .666 1.066 181 7.9 7.5
Betts 2018P .438 .669 1.106 193 9.2 10.8

Source: Baseball-reference.com

Notes:

  • Betts has a higher OPS than Trout, yet has a lower OPS+. Perhaps, OPS+ deserves a separate post about being a questionable statistic. 
  • The difference between Trout’s WAR and OWAR is now 0.8 versus 2.7 as of late May. Presumably, the Angels’ outfielder had some great defensive plays early in the year that were unsustainable. AGAIN, SAMPLE SIZE!!!

I’ll admit that I don’t know the specific calculation for offensive wins above replacement (OWAR), but it has to be flawed. Of note, Trout’s OWAR is projected to be 2.0 points higher than Martinez despite:

  • 38 fewer hits (196 vs. 158).
  • 9 fewer home runs (50 vs. 41).
  • 4 fewer runs (117 vs. 113).
  • 58 fewer RBIs (140 vs. 82).
  • Batting average 21 pts lower (.330 vs. .309).

To be fair, Trout is projected to have an OBP 58 pts higher than Martinez (.459 vs. .401). Still, that can’t be worth a 25% net increase in OWAR relative to the difference in all of the other stats.

WHAT’S THE POINT

I feel the need to repeat that Mike Trout is the best position player in the MLB today. At the same time, I don’t think we should evaluate players based on obscure statistics that few people can define (or defend). Despite my mantra that “numbers never lie,” I acknowledge that statistics can be used to prove whatever case someone wants to make.

In addition to being used to evaluate players, “advanced” statistics are being used to manage games (e.g. shifts, removing starting pitchers, changing lineups, etc.). Point being, an over-reliance on numbers may be ruining the game that used to be America’s pastime. As a result, I’d like to change my mantra to “your eyes never lie.”‘

JAYSON WERTH

Former MLB player Jayson Werth recently shared a similar view while appearing on the Howard Eskin Podcast.

I think [sabermatrics is] killing the game. It’s to the point where just put computers out there. Just put laptops and what have you, just put them out there and let them play. We don’t even need to go out there anymore. It’s a joke.

When they come down, these kids from MIT, Stanford, Harvard, wherever they’re from, they’ve never played baseball in their life…. We’re creating something that’s not fun to watch. It’s boring. You’re turning players into robots.

When they come down to talk about stuff like [shifts], should I just bunt it over there? They’re like, ‘No, don’t do that. We don’t want you to do that. We want you to hit a homer.’ It’s just not baseball to me. We’re creating something that’s not fun to watch. It’s boring. You’re turning players into robots. You’ve taken the human element out of the game.

As a graduate of one of the schools Werth mentioned, I resent resemble the assertion. Then again, I actually played baseball so I get his point. 

GOOSE GOSSAGE – THE CRITIC

Hall-of-Famer Goose Gossage expressed the same sentiment in February 2017 when he had the following exchange with Randy Miller of NJ Advance Media.

NJAM: I guess you hate the sabermetrics of baseball.

Gossage: “Oh, you know what it is: These guys don’t know baseball, so they’ve got one thing to tell them how to do it, and that’s numbers. And they won their rotisserie leagues at (bleeping) Harvard and all these (bleeping) Ivy League schools, and that’s all it is. And they think they’re (bleeping) general managers! And that’s the way the game has gone. They’re taking all the character out of the game and creating a bunch of soft guys.”

GOOSE GOSSAGE – THE PLAYER

I grew up as in the 1970s and 80s as a Yankees fan. The most indelible moments from those years include:

  • Reggie Jackson’s 3rd home run in Game Six of the 1977 World Series.
  • Graig Nettles catching a foul ball to end the 1978 AL East tie-breaker against the Boston Red Sox. Gossage got the save, but turned a 5-2 lead into a 5-4 nail-biter after giving up 2 runs in the 8th inning. 
    • Most baseball historians recount Bucky Dent’s home run in the 7th inning as the marquee moment of that game. I remember it differently given that the Red Sox still had 9 outs to recapture the lead. As a foreshadowing, I didn’t have confidence that Gossage would secure the win.
  • Walking into my first MLB game at Yankee Stadium for a night game during the summer of 1982.
  • Gossage giving up the home run to George Brett in the “Pine Tar Game” in July 1983. As I mentioned earlier, I never had the confidence in Gossage that other might have had.

Interestingly, I have a better appreciation for Gossage as an outspoken critic than I ever did for him as a player. 

LIKE GOOSE ON RICE

Gossage certainly knows how to stir up controversy when comparing the current game to the game he once played. I appreciate his willingness to speak openly even if others don’t. I started to gain that appreciation when he argued for Jim Rice’s inclusion in the National Baseball Hall of Fame. Specifically, Gossage said the following about his former competitor

I’d be staring him down, he’d be staring right back and I’m telling you there was no man I wanted to face less than Jim Rice. He was the strongest, toughest guy, most competitive man I ever faced.

As a Yankees fan, I hated Jim Rice. I acknowledged that he had a few good years with the Red Sox, but didn’t think he had the career stats to warrant inclusion with the best-of-the-best in Cooperstown. Specifically, Rice didn’t surpass any of the magical numbers of 3,000 hits (2,452), 400 home runs (382), 1,500 RBIs (1,423), or a .300 average (.298) so I rationalized that he didn’t deserve the honor.

As Rice approached his final year of eligibility, I changed my opinion based on what contemporaries like Gossage said about him. Goose faced all-time greats like Reggie Jackson, George Brett and Mike Schmidt during his HOF career, yet considered Rice the toughest of them all. At that point, I realized that the opinions of fellow competitors mattered more than any stats.. 

CHANGES TO HOF ELIGIBILITY

When the baseball HOF reduced the eligibility requirement from 15 to 10 years in 2014, I wondered which players might be excluded based on the change. In particular, I figured Tim Raines would miss out considering his voting progression mirrored Rice’s. The former Expos speedster had 3 years to go from 46% to 75%, which seemed insurmountable based on past progressions. Yet, somehow it happened.

JIM RICE VS. TIM RAINES – HALL OF FAME VOTING (BBWAA)
Year of Eligibility Jim Rice Tim Raines
1 1995 – 29.8% 2008 – 24.3%
2 1996 – 35.3% 2009 – 22.6%
3 1997 – 37.6% 2010 – 30.4%
4 1998 – 42.9% 2011 – 37.5%
5 1999 – 29.4% 2012 – 48.7%
6 2000 – 51.5% 2013 – 52.2%
7 2001 – 57.9% 2014 – 46.1%
8 2002 – 55.1% 2015 – 55.0%
9 2003 – 52.2% 2016 – 69.8%
10 2004 – 54.5% 2017 – 86.0%
11 2005 – 59.5%  
12 2006 – 64.8%  
13 2007 – 63.5%  
14 2008 – 72.2%  
15 2009 – 76.4%  

Source: Baseball-reference.com

After 8 years on the ballot, both Rice and Raines received 55% of the BBWAA (Baseball Writers’ Association of America) votes. With only two years left on the ballot based on the 2014 rule change, Raines got big bumps in 2016 and 2017 to easily meet the 75% threshold. So what happened in those years? In short, the writers evaluated Raines’ career based on his WAR. 

To further emphasize my point about the questionable value of WAR, I offer the peak and career production of both Jim Rice and Tim Raines.

JIM RICE VS. TIM RAINES – PEAK PRODUCTION
Player Year Games Hits HRs Runs RBIs BA
Rice 1978 163 213 46 121 139 .315
Raines 1985 150 184 11 115 41 .320
               
Player Year OBP SLG OPS OPS+ OWAR WAR
Rice 1978 .370 .600 0.970 157 6.7 7.6
Raines 1985 .405 .475 0.880 151 6.7 7.6

Source: Baseball-reference.com

Notes:

  • Rice won the AL MVP in 1978 while Raines finished 12th in voting for the NL MVP in 1985. 
  • Raines had a 5-pt advantage in batting average (BA) and 35-pt advantage in on-base % (OBP), but Rice performed better in every other offensive category. The former Red Sox outfielder also dominated in HRs, RBIs and slugging (SLG) yet only produced the same OWAR as Raines. Huh?  
JIM RICE VS. TIM RAINES – CAREER PRODUCTION
Player Years Games Hits HRs Runs RBIs BA
Rice 1974-1989 2,089 2,452 382 1,249 1,451 .298
Raines 1979-2002 2,502 2,605 170 1,571 980 .294
               
Player Years OBP SLG OPS OPS+ OWAR WAR
Rice 1974-1989 .352 .502 .854 128 45.7 47.7
Raines 1979-2002 .385 .425 .810 123 69.4 69.3

Source: Baseball-reference.com

Notes:

  • In 400+ more games than Rice, Raines had approximately 150 more hits and 320 more runs. 
  • In 400+ fewer games than Raines, Rice had 200+ more home runs and almost 500 more RBIs. 
  • According to OWAR, Raines somehow had a career 50% more productive than Rice’s. Huh?

During his career, Rice finished in the top 5 for MVP 6 times (1st, 3rd, 3rd, 4th, 4th, 5th) and won the award once. In comparison, Raines finished once in the top 5 (5th) and 3 times in the top 10 (5th, 6th, 7th). Presumably based on WAR, the writers retroactively viewed Raines’ career much more favorably than they did while he was playing.

Do you have any issues with that change? If not, I do. To be clear, I don’t have an issue with Tim Raines making the Hall of Fame. However, I have an issue when writers reevaluate someone’s career based on a new set of criteria that didn’t exist at the time. 

FULL CIRCLE

As part of his claim that current players are “soft,” Gossage emphasized the current emphasis on pitch count. Whereas pitchers like Nolan Ryan routinely threw over 200 pitches in a game during Gossage’s heyday during the 1970s, current pitchers rarely surpass a pitch count of 100 regardless of the circumstances.

This past Sunday night, the Nationals’ Max Scherzer and Cubs’ Cole Hamels engaged in an exciting pitcher’s duel. After 7 innings, Scherzer had given up 3 hits and 0 runs while Hamels had given up 1 hit and 1 run. However, each pitcher had thrown approximately 100 pitches so they both got pulled.

Cubs reliever Carl Edwards pitched a perfect 8th inning, but Brandon Kintzler and Justin Wilson gave up 2 runs in the 9th to go down 3-0. In comparison, Nationals reliever Koda Glover pitched a perfect 8th inning before Ryan Madson took over in the 9th. 

I mentioned earlier about seminal moments as a Yankees fan during my youth. Since then, I have indelible memories of watching:

  • Doug Flutie’s Hail Mary pass in November 1984;
  • Joe Carter’s World Series-clinching home run in 1993; and
  • David Bote’s game-winning grand slam in 2018.
ULTIMATE BASEBALL DREAM: DOWN 3-0 WITH 2 OUTS IN BOTTOM OF 9TH AND BASES LOADED

While my memories of Flutie and Carter’s heroics can be understood by most, the last one will only resonate if you happened to see the Cubs/Nationals game on August 12, 2018. After watching updates on SportsCenter for over an hour, I switched over to the MLB Network hoping to catch the end of the game. I changed the channel with the Cubs down 3-0 with 2 outs and bases loaded in the bottom of the ninth. Within 2 minutes, I saw Bote hit the type of game-winning home run that I dreamed about hitting as a kid.

I appreciated seeing something that has happened only 1 time in the last 150 years, but immediately looked at the box score and realized that it never should have happened in the first place. Specifically, here’s what I read.

NATIONALS PITCHING STATS – AUGUST 12, 2018
Pitcher IP H R ER BB SO HR Pitches
Max Scherzer 7.0 3 0 0 1 11 0 106
Koda Glover 1.0 1 0 0 0 0 0 12
Ryan Madson 0.2 2 4 4 0 0 1 21
Team Totals 8.2 6 4 4 1 11 1 139
CUBS PITCHING STATS – AUGUST 12, 2018
Pitcher IP H R ER BB SO HR Pitches
Cole Hamels 7.0 1 1 1 1 9 0 98
Carl Edwards 1.0 0 0 0 0 2 0 9
Brandon Kintzler 0.1 2 2 2 2 0 0 12
Justin Wilson 0.2 0 0 0 0 1 0 11
Team Totals 9.0 3 3 3 3 12 0 130

Source: Baseball-reference.com

Back in the day (i.e. 20-30 years ago), Scherzer and Hamels would have battled to the finish. Instead, numerous relievers, whom I never heard of, came in to ruin the game. 

CAN I INTEREST YOU IN SOME SHOWER CURTAIN RINGS?

As a quick aside, my wife mentioned that she liked this article best out of the 100+ articles I have written so far. For that reason, I feel compelled to update it by offering a quick anecdote that hopefully Neal Page would agree has a point.

I have known my wife since the late 1980s. From the beginning of our relationship, she didn’t understand sports etiquette when entering a room in which someone may be watching a live sporting event. I once joked that she would walk into a room in the bottom of the ninth of a World Series-clinching game and start talking about some trivial event.

Well, it happened to me when Joe Carter came to the plate with two men on and the Blue Jays down by 1 in Game 6 of the 1993 World Series. I remember thinking that Carter only had to hit a double to win the Series, when my then-girlfriend, now-wife walked into the room. As she started to talk, I shouted, “This is exactly what I feared you would do! Just watch!”

My wife saw less than 3 minutes of the 1993 World Series and witnessed the 2nd most exciting conclusion to an MLB season. In case you’re wondering, Bill Mazeroski’s Game 7 home run to beat the Yankees in the 1960 World Series ranks as #1. Thankfully, Carter’s home run happened early in our relationship so I simply say, “Joe Carter” when she tries to interrupt an important sporting event.

As I switched to the MLB Network to see David Bote’s at bat a couple weeks ago, I was waiting my wife to come into the room and ask me to watch a You Tube video starring Harley the Cockatoo.

CONCLUSION

I saw the quintessential ending to a baseball game (i.e. a game-winning grand slam when down by 3 runs), but didn’t feel fulfilled. Instead, I immediately thought about Werth’s quote.

We’re creating something that’s not fun to watch. It’s boring. You’re turning players into robots.

In retrospect, I realize the managers’ over-reliance on stats (i.e. pitch count) created the opportunity for the dramatic finish. Despite any belief that sabermetrics may hurt baseball, I acknowledge that the human element of sports will prevail. I have renewed faith because biology/evolution will always win in the end. If not, we wouldn’t be here.  

T10B BUSTED: WAR (WINS ABOVE REPLACEMENT)