WAR accounting

[JR Oriole]

I honestly don't know the answer to this, so I need the WAR guys to help me.

Last season, there was a game in Tampa where I think Nelson Cruz drove in all 7 runs. He came up with a massive clutch hit in the 9th and then again in extras. While we won the division at season's end by 12 games, at the time that game in Tampa was big because the Yankees were gaining momentum and still had a chance to catch us. Also, we were historically awful in The Trop against Tampa's pitching. So how does WAR account for the hugeness of that game at that time for Cruz?

Or what about that Paredes walk-off single against NY in extras, or the really big hits by Cruz and Pearce against NY on that Sunday night game which essentially clinched us the division? Does WAR assign extra points for big time hits?

What does WAR say about a guy like Pablo Sandoval, who may be just okay during the season but then in October, you literally can't get him out?

I guess like any stat, it is just one of many measures used to evaluate a player. But for me, I tend to remember those huge hits in big games and put a lot more value on that than stats like WAR....unless WAR captures that in some way.

[25 Nuggets]

WAR doesn't do a very good job of capturing isolated moments. It's a derived stat based on accumulated data. But a big game like Cruz had will show up in his score. You can really see the difference when a defender has a really good / really bad game, because most defensive plays are routine. I remember Adam Jones had a spectacular game defensively and his dWAR went from like -0.1 to 0.2 just like that. (that was the game he made the great catch and doubled the guy off second base)

I hear ol' Ken Harrelson came up with a "clutchiness" stat but I haven't seen the math behind it.

[JR Oriole]

I hear ol' Ken Harrelson came up with a "clutchiness" stat but I haven't seen the math behind it.

Speaking of Harrelson, Joe Crede (when he was with the White Sox) was super clutch. I lived in Chicago then and watched a ton of White Sox games...that guy came up huge in big moments. Not sure if there is a pure stat for it, but I knew that he was a guy you wanted to come up to the plate when it counted.

[weams]

I honestly don't know the answer to this, so I need the WAR guys to help me.

Last season, there was a game in Tampa where I think Nelson Cruz drove in all 7 runs. He came up with a massive clutch hit in the 9th and then again in extras. While we won the division at season's end by 12 games, at the time that game in Tampa was big because the Yankees were gaining momentum and still had a chance to catch us. Also, we were historically awful in The Trop against Tampa's pitching. So how does WAR account for the hugeness of that game at that time for Cruz?

Or what about that Paredes walk-off single against NY in extras, or the really big hits by Cruz and Pearce against NY on that Sunday night game which essentially clinched us the division? Does WAR assign extra points for big time hits?

What does WAR say about a guy like Pablo Sandoval, who may be just okay during the season but then in October, you literally can't get him out?

I guess like any stat, it is just one of many measures used to evaluate a player. But for me, I tend to remember those huge hits in big games and put a lot more value on that than stats like WAR....unless WAR captures that in some way.

Bad players have huge moments. Come up big in huge way. WAR just helps keep you from paying big money to those bad players. Just because the spotlight caught them being good.

[Frobby]

WAR does not take context into account, it is based solely on the average value of an event (e.g., a homer on average is worth X number of runs, and Y number of runs averages one win). There is a stat called Win Probability Added (WPA) that weights the outcome of every at bat according to how it affected the statistical probability of winning the game (so a game winning home run in the 9th inning has much more "value" under WPA than a solo shot in the 7th inning when you are already ahead by 7 runs). Notably, WPA only accounts for offense, unlike WAR, which takes both offense and defense into account. I think both are valuable tools, but I think you'd find that the offensive component of WAR is more stable from year to year than WPA is.

Offensive WPA leaders for the Orioles in 2014:

Pearce 3.65

Cruz 2.88

Davis 1.38

Jones 1.24

Hundley 0.95

What this tells you is that while Davis and Hundley had disappointing overall offensive stats, they came up with some very important hits at key times during the season.

Worst on the team were:

Schoop -2.69

Hardy -0.98

Flaherty -0.76

Lombardozzi -0.57

Young -0.53

Some of those are surprising. Hardy had good RISP numbers and it seemed to me he came up with his share of key late hits. Young was devastatingly good as a pinch hitter. I'm not surprised that Schoop is negative given how rarely he got on base, but he did seem to hit most of his homers in propitious situations (we won 14 of the 15 games in which he homered. It just goes to show you, all those hundreds of unproductive at bats do add up, even if they aren't the ones we remember.

[luismatos4prez]

I honestly don't know the answer to this, so I need the WAR guys to help me.

Last season, there was a game in Tampa where I think Nelson Cruz drove in all 7 runs. He came up with a massive clutch hit in the 9th and then again in extras. While we won the division at season's end by 12 games, at the time that game in Tampa was big because the Yankees were gaining momentum and still had a chance to catch us. Also, we were historically awful in The Trop against Tampa's pitching. So how does WAR account for the hugeness of that game at that time for Cruz?

Or what about that Paredes walk-off single against NY in extras, or the really big hits by Cruz and Pearce against NY on that Sunday night game which essentially clinched us the division? Does WAR assign extra points for big time hits?

What does WAR say about a guy like Pablo Sandoval, who may be just okay during the season but then in October, you literally can't get him out?

I guess like any stat, it is just one of many measures used to evaluate a player. But for me, I tend to remember those huge hits in big games and put a lot more value on that than stats like WAR....unless WAR captures that in some way.

When formulating WAR, a player who hits a bases loaded double with 2 outs in the 9th receives as much credit as a player who hits a double with bases empty down 10 runs in the 9th.

The theory is that all major league players are outstanding at handling immense pressure. If they weren't, they wouldn't have made it to the majors. Clutchness is mostly just a narrative given by fans, and isn't a good way to measure value. Derek Jeter is a good example. He hit .310 in the regular season over his career and was known as an amazing postseason player, even though he hit .308 in the postseason in his career. Nelson Cruz is another one. He's been an unbelievable hitter in the postseason in his career. But if he really took his performance to another level when he needed to, he would have caught that ball in RF to win the WS for Texas. It was a catchable ball, even for him.

Guys who perform in the clutch do so because they're good hitters mostly, not because of clutchness. There probably is some effect of the moment on the player, but it's just as likely to affect the pitcher and the fielders and it's probably much, much smaller than it's given credit for by fans.

[BohKnowsBmore]

WAR doesn't do a very good job of capturing isolated moments. It's a derived stat based on accumulated data. But a big game like Cruz had will show up in his score. You can really see the difference when a defender has a really good / really bad game, because most defensive plays are routine. I remember Adam Jones had a spectacular game defensively and his dWAR went from like -0.1 to 0.2 just like that. (that was the game he made the great catch and doubled the guy off second base)

I hear ol' Ken Harrelson came up with a "clutchiness" stat but I haven't seen the math behind it.

WPA does a good job capturing "clutch" by weighting successes and failures based on the impact on win expectancy.