THT Awards 2012: season finale, pitchers edition

Welcome to the awards. As a reminder, this covers the regular season as a whole. We will cover the hitters and put a bow on the season next week.

Please see the week one column for award definitions and explanations.

This week’s proof that assigning wins and losses to a pitcher is an archaic practice that must stop

Good luck division

Amongst qualified starting pitchers, Lance Lynn was 35th in the National League in ERA, 56th in the Major Leagues at 3.78. His xFIP was a little brighter at 3.60 but that still only placed him 19th in the NL. But thanks to run support, only Gio Gonzalez, R.A. Dickey, and Johnny Cueto exceeded his 18 wins. All three of those starters had sub-3.00 ERA’s. He was lucky several times. He won a game where he allowed six runs and another where he allowed four in five and a third. He salvaged no decisions in games where he allowed four in two innings, four in five innings, three in four and a third, four in six, and one where he was pummeled for six in five and a third. Meanwhile, the only quality start he took a loss in was one where he allowed three in six innings and he only carried no decisions in two games where he posted a quality start.

Felix Doubront went 11-10 for the Red Sox with a 4.86 ERA. To be fair, much of that ERA damage came in games he has reflected in his loss column. But it could have been so much worse, especially given how bad the Red Sox ended up being this season. He won a game where he allowed five runs and two more where he allowed four. He was saved from more losses when he received run support necessary to get no-decisions in a game where he allowed five runs in five frames and in two more where he allowed four in five innings. He went seven and two with three no decisions in his starts where he went at least six innings and allowed three or fewer runs.

Ivan Nova was 12-8 with an ERA over 5.02. He had bad luck on balls in play, but not in terms of decisions. Nova never lost any of the 15 games in which he allowed fewer than four runs. He also won games where he allowed five runs in six and two thirds and four in six. That goes with the no decisions he received where he allowed six runs once, five runs once, and four runs once.

Finally, Derek Holland went 12-7 for the Rangers with a 4.67 ERA. He had some of the best run support in the business. He never lost a start where he allowed three or fewer runs, going nine and oh in those games. Meanwhile, he won a game where he was shelled for seven runs by the Angels and won another where he allowed five runs to the Rays. He salvaged no-decisions in games where he allowed six runs in six innings and five in five and two thirds.

Bad luck division

On the other side of the scale was Josh Johnson, who took the loss in eight quality starts, including one where he allowed only one run in eight innings to the Phillies and another where he went seven and two thirds against Atlanta, allowing only two runs. He had five more games where he was stuck with a no-decision when he allowed two or fewer runs and went six or more innings. He had seven starts where he went at least six innings and allowed exactly three runs. He had only one win against six losses in those games.

Cliff Lee owned a 3.16 ERA and an unfathomable 6-9 record. The number of well-pitched no-decisions is staggering. He took a no-decision in a game where he threw 10 shutout innings. He took no-decisions in two games where he went eight innings, allowing one run, another where he threw seven innings with one run allowed, one where he went six innings with one run allowed. He lost a game where he went eight and allowed two runs and another where he went seven and two thirds and allowed two. He lost two where he went six innings and allowed three runs. And he took no-decisions in five games where he went at least six innings and allowed two or three runs. He never won a game where he allowed more than two runs.

Jeremy Hellickson had only five starts where he allowed four or more runs out of 31 total times on the mound. He finished the year with an 11-10 record despite a 3.10 ERA for a team that finished with 90 wins. That is especially rotten luck. He had a start where he went seven innings, allowing one run and still took the loss. He had another one meeting the same inning and run count where he received no decision. He lost three separate games where he went six innings twice and six and two thirds once, allowed two runs in each, and took the loss. He had two wins all year where he allowed three runs. By my count, he took the loss in five quality starts and a no-decision in five more.

In keeping with tradition, Felix Hernandez is here. His case is a little softer than usual but it still bears mentioning. King Felix really only had one true hard luck loss this year when he lost a game where he allowed two runs to Oakland in seven frames. On the other hand, he had a lot of no-decisions that were hard to take. He had one game where he held Cleveland scoreless for eight innings and didn’t get the win. He had another where he went eight, allowing one run to the Orioles, another with seven and two thirds and only one run allowed, a seven inning, one run, and two eight inning, one run starts, all with neither a win or a loss. He had two more, one where he went seven frames and allowed two runs and another where he went six innings and allowed two runs. Two more no-decisions.

Clayton Kershaw’s season was similar to Hellickson’s. He had three lucky no-decisions, but only six total starts where he allowed four or more runs. Meanwhile, he took a loss in five games where he went at least six innings and allowed two or fewer runs, including one where he held the Diamondbacks to one run in seven frames and two separate games where he went eight innings against the Giants, allowing two runs in each and took the loss. He had two separate no-decisions where he went eight innings and allowed one run and two more where he went seven and allowed one run. His 14-9 record was very unlucky given that he was one of the best pitchers in baseball with a 2.53 ERA, a 2.89 FIP, a 3.25 xFIP, and a 5.5 fWAR.

Vulture alert

Between July first and July 18th, Santiago Casilla pitched in six games. He entered all of them in a save situation and was only successful in holding the lead in two of them. In one, he took the loss. In another, he was removed and got a no decision in what would later be a Giants win. And in the remaining two, he walked away with the win, two of his seven on the year.

Mental Health and the CBA
A particular bit of language in the latest CBA could have negative consequences for some players.

Pedro Strop blew the lead in seven games this season and walked away with exactly zero losses and one win. He has taken two losses this season, but none of them were in games where he was given a blown save.

Wes Littleton Award

Every season there is a closer who puts up a large number of save despite not being a particularly good pitcher. This year everybody agrees that John Axford was terrible but he still accumulated 35 saves for the Brewers. Hitters batted .229/.330/.387 against him. He wasn’t quite as awful as Joe Borowski in 2007, but he was a horrible pitcher who played a large role in sinking his team’s playoff hopes by being a leading figure in one of the worst bullpens in the league.

Chris Perez saved 39 games for a terrible Cleveland club despite a 3.59 ERA, which was not a fluke as demonstrated by his 3.67 xFIP. Most of the other pitchers in the top 10 in saves were pretty dominant. Perez now has 107 career saves and a WAR of 1.9, just over half that of Craig Kimbrel just this year and less than the 2012 figures put up by Fernando Rodney and Aroldis Chapman. According to WPA, he added a quarter of a win all season. Nine of his saves came with a three or four run lead.

Please hold the applause

Axford’s teammate Francisco Rodriguez was credited with a hold in 32 games this season. He carried a negative win probability and a 4.38 ERA. He no longer has the velocity he once did and he has junked the slider that used to be his signature move in favor of a curve that doesn’t provide much value anymore.

Any sufficiently advanced defense is indistinguishable from pitching

For the first year ever, I have tried to isolate the pitchers that have some manner of demonstratable luck on batted balls in play from those who seem to have been genuinely lucky. This is why you will not see Jered Weaver, who had the best BABIP against of any pitcher with at least 110 innings this season. He has had an xFIP more than a run higher than his actual ERA in three of the last four years.

Jason Vargas has no such record in his performance history. His .254 BABIP was significantly lower than his career figure of .276. It is unclear to me how much this might have had to do with Safeco Field as his BABIP was .245 in his home starts and .264 away.

Ross Detwiler seems to be another good bet to be a fluke as his .263 BABIP is 19 points lower than his career norms and nine points lower than his 2011 figure. You don’t often put up a 3.40 ERA with 5.8 strikeouts per nine.

On the other end of the equation, Ivan Nova had what appears to have been an aberrant BABIP year. His .331 far exceeds what he had done before. His two pervious campaigns featured BABIP’s under .300. This led to his 5.02 ERA compared to his 4.16 xFIP. His strikeout rate actually climbed by nearly two points from 2011 but his ERA rose by a run and a third.

This year in trends

Just for this occasion, I flipped through some league data on Baseball-Reference, which is not my strong suit and not something that I have a history of doing. But I wanted to see where we were with certain statistical trends in baseball. I will leave it to other, more qualified individuals to make a more advanced statistical study and build graphs and solid conclusions. I just wanted to present some interesting numbers.

The first piece of news I have is that run scoring was up slightly from 2011 (4.32 runs per game up from 4.28) but is still down from 2010’s 4.38, way down from 2009’s 4.61, and not even in the same area code as the most recent peak of 2000 at 5.14. The three areas I was curious about leading into the raw run scoring figures were the percentage of batted balls turned into outs, the number of home runs hit, and the number of strikeouts.

In the first category, balls in play were slightly less likely to be turned into an out than they were last year. B-R lists this year’s defensive efficiency at .691, down from .694. This is less of a long-term trend than I had anticipated before I actually looked it up. The figure was actually the same or higher in every season from 2001 through 2005, when more runs were scored per game than there are now and the most recent peak was .695 in 2002, when 4.62 runs were scored per game.

In the second category, I took the number of home runs hit as a percentage of total plate appearances in baseball. This year’s figure of 2.68 percent was down from last year’s 2.75 but up from 2010’s 2.49, which represents the lowest figure since the peak 12 years ago when it was 2.99 percent of plate appearances.

The last of the three things I wanted to look up is the one with the clearest trend. Strikeouts as a percentage of total plate appearances have been going up consistently for some time now. This year’s 19.78 percent was as high a number as I could find and was more than one point higher than last year’s 18.62 percent. It has risen every year since 2005. There are now likely fewer balls in play than at any point in recent memory.

Best Pitcher

AL: Justin Verlander was the best pitcher in baseball this year, just like last year. He will not win this season because Jered Weaver and David Price have similar ERA’s and more wins. His 2012 was nearly a carbon copy of his 2011 except for one thing, wins. Nearly every other statistic is exactly the same except for the fact that his teammates scored fewer runs for him. B-R lists his 2011 run support as 4.73 and his 2012 as 3.98. He receiver three or fewer runs of support in 17 of his 33 starts.

Felix Hernandez was also great and had worse run support. He wasn’t quite as good as Verlander, but he was still, on the whole, a special pitcher.

Price had a better ERA and FIP than Felix, but 21 more innings have value.

NL: I’m going with R.A. Dickey here but any choice from among Kershaw, Cueto, Gio Gonzalez, and Matt Cain and I won’t have a significant problem with it. There were a lot of very good pitchers in the NL this year and none that matched Verlander in the AL.

Print This Post
Sort by:   newest | oldest | most voted

Verlander’s ERA was 0.24 higher, his K/BB 0.41 lower, xFIP 0.19 higher, and SIERA 0.27 higher than his 2011. So not quite as good, but still great.

Ironically, Price had a lower xFIP and SIERA than Verlander. And even more, both Fister and Scherzer were very comparable to Verlander in terms of xFIP, K/BB, and SIERA.

Bob Rittner
Bob Rittner

Hellickson was 10-11, not 11-10.

I think part of his problem was that he managed to pitch only about 5.7 innings per start. As you point out, he had a few longer outings, but generally was out of the game in the 6th inning, partly I think because of inefficiency.


A little harsh on Axford. No he wasn’t spectacular, but a 4.06 FIP, 3.29 xFIP, and 3.10 SIERA hardly warrants “everyone agrees he was terrible” and “a horrible pitcher”.

John M Barten
John M Barten
Rob: Thanks. That is worth mentioning and it certainly plays a role in it, especially the no-decisions. Pretty hard to make the case that the won/loss record accurately reflects his ability to prevent the other team from scoring runs. He was a well above average pitcher with bad run support. 3.23 runs per game is pretty bad. Matt: The Tigers had some good starters this year. It was nice to see Scherzer put up a good season. And Price had a great season. Verlander’s innings advantage on Price is something to consider and probably a factor in Verlander’s fWAR and… Read more »

“There were a lot of very good pitchers in the NL this year and none that matched Verlander in the AL. “

I disagree. Kershaw had a lower ERA, FIP, xFIP, tERA, WHIP, and SIERA than Verlander. Both had similar K% and B%. Very similar in ERA-, ERA+, and WPA.