Let’s get rid of pitching winsby Glenn DuPaul
September 12, 2012
Everyone has dreams. I’m not referring to what happens when our mid-sleep imaginations run wild, but dreams in the sense that we all have aspirations, goals and hopes for our lives. I have a ton of dreams, big and small, but one of them might sound a little strange.
I dream that one morning I’ll wake up to a world in which the pitching win no longer exists as a baseball statistic. I can’t stand pitching wins. Pitching wins may not be the worst baseball statistic ever invented, but quite honestly it is the one I like the least. I listed some quotes below that always seem to make me cringe:
“He pitched well, but didn’t get the win”
“He never pitches great, but he always pitches just well enough to get the win”
When I read or hear quotes like that the first thought that always pops into my mind is an exasperated "Who cares?" But for some reason people still care and I’m slightly pessimistic that we may ever see the day when people don’t talk about pitching wins. The sabermetric community hates pitching wins, but not everyone listens to saber-heads. Pitching wins still alter the perceptions of many baseball fans, which is the main reason why I can't stand the statistic.
Here’s a quick blind test:
**Note: The FIP (fielding independent pitching) used is FanGraphs calculation
Which pitcher was more effective?
From just looking at this quickly it seems that Pitcher A was more adept at run prevention and gave up fewer home runs, while Pitcher B had better strikeout-to-walk numbers, and their ability to go deep into games is essentially a wash. But can we tell from just these metrics which pitcher was most effective?
Usually I think that information would be just about enough, but in this case I think the comparison is too close and we need some additional information. I listed some questions below that could be asked, to better our understanding of which pitcher was the most effective:
- 1. What is the difference between the hitting environment two pitchers’ home parks?
- 2. How good were the defenses behind the pitchers?
- 3. What was the quality of opponents that each pitcher faced?
- 4. What was the difference between the two pitchers’ BABIPs (batting average on balls in play)?
1. These are run and home run park factors from each pitcher's home park. For reference, a hitter-friendly environment would have a park factor over one:
Pitcher A: ESPN Run/HR Park Factor: 0.995/0.824
Pitcher B: ESPN Run/HR Park Factor: 0.967/1.085
-Maybe it wasn't Pitcher B's fault that he gave up so many home runs, as is indicated by his HR Park factor and his xFIP (expected fielding independent pitching.)
2. A positive defensive efficiency is good, while negative is bad:
Pitcher A’s team park-adjusted Defensive Efficiency: -0.83
Pitcher B’s team park-adjusted Defensive Efficiency: -0.63
-Both pitchers had below average defensive behind them.
3. A league average TAv (True Average) is .260, anything above is above average:
Pitcher A: had an oppTAv (opposing True Average) of .268
Pitcher B: had an oppTAv of .273
-Pitcher B faced slightly tougher opponents, but both pitchers faced above average opponents
4. A BABIP above league average is typically assumed to be unlucky, but if the team's defense isn't good (as is this case with both pitchers), then the team's staff BABIP usually ends up above league average:
Pitcher A’s BABIP/LgAvg BABIP/Team Staff BABIP: .301/.296/.301
Pitcher B’s BABIP/LgAvg BABIP/Team Staff BABIP: .316/.293/.299
-Pitcher B faced tougher opponents than Pitcher A, which could have led to his higher BABIP, but I don't think that accounts for all of it, especially because pitcher B's BABIP is so much higher than his team's staff BABIP. A case could be made for Pitcher B being unluckier than Pitcher A.
So I've dumped out a ton of information (probably too much) about these two pitchers, but we still haven't answered the question of which one is the more effective. When we consider the fact that run prevention is a pitcher's primary responsibility than the scale tilts slightly towards Pitcher A, but when the other factors are considered I'd say the comparison is pretty close to even.
Who are they?
- Pitcher A: 2008 AL Cy Young Award winner Cliff Lee
- Pitcher B: Lee's 2012 season
So, somehow Lee's Cy Young year versus this year, which is considered by many to be a “down-year” is almost a draw in terms of effectiveness. Where did the difference in Cy Young vs. less-than-stellar come from?
My opinion is a difference came from perception that comes from pitching wins. Lee won 22 games in 2008, but he has won only four games so far this season.
There is the argument that pitching wins mattered in the Cy Young voters minds a lot more in 2008 than they do today. The basis of this argument comes from Felix Hernandez winning the 2010 award, with just 13 wins.
That award was considered a win for the sabermetric community, but why is that Lee's four-win season is still considered a down-year, in 2012?
The difference in Lee's ERA and peripheral statistics like FIP, xFIP and SIERA (Skill-Interactive ERA) could be the first reason for this perception. Depending on the way the reader wants to interpret those three statistics, either Lee is going to better over the next three weeks than he was earlier this year, or he already has been better than his ERA would indicate.
The interesting thing is that Lee's ERA really isn't bad. HIs current ERA (3.50) is almost a full run worse than it was last season (2.54), but it still ranks in the top 15 among NL qualified starters.
I really think the perception of Lee's season has been distorted by his low win total; which has as much to do (if not more) with the Phillies' defense, bullpen and offense when Lee pitches than Lee's actual performance.
Here's another brief comparison. This time between Johnny Cueto and Lee, this season.
Cueto vs. Lee
Both the mainstream media and the sabermetric community consider Cueto to be one of the frontrunners for the Cy Young award.
Last week, Cliff Corcoran of Sports Illustrated ranked Cueto as the number two NL Cy Young candidate. Two weeks ago, Dave Cameron of FanGraphs wrote an article entitled "Johnny Cueto For Cy Young." Corcoran's article cited Cueto's run prevention as the main reason backing his Cy Young candidacy. Cameron also cited that ability, but proposed a theory for why Cueto is able to prevent runs lower than his peripheral statistics would indicate.
Cameron's theory is that Cueto's pickoff move and ability to control the running game is a major factor in his left-on-base percentage being so high and in turn his LOB-Wins (left-on-base wins) also being high:
Cueto’s pickoff move isn’t the stuff of legends yet, but it probably should be. The list of the top ten pickoffs by a pitcher this year includes nine left-handed pitchers and Johnny Cueto, and despite being right-handed...
Cueto’s been doing this kind of thing all year long, and it’s gotten to the point where there’s no real point even trying to get much of a lead off first base, much less think about taking second. Opposing baserunners have managed one steal off Cueto all season, matching the same number of stolen bases that he allowed in 2011...
Like we acknowledge that Dickey and Cain are likely influencing a decent amount of their hit prevention, we should also acknowledge that Cueto is influencing a large part of his runner stranding, and given that he also leads both of them in FIP, we should give Cueto enough credit for his FDP that he returns to the top of the heap in the Cy Young race once again.
Cameron makes a fair point about Cueto controlling the running game, but is the fact that he picked off seven runners and allowed only one steal enough to explain his entire LOB percentage? Here's a breakdown of the running game against Cueto in 2011 vs. 2012:
|Year||Pickoffs||SB||CS||(PO+CS)-SB||LOB%||All Scored||Runs Saved|
The last two columns are the most important. The second last (titled All Scored) shows what Cueto's left-on-base-percentage would be if all of the runners who had either been caught stealing or picked off had actually scored, which isn't really a fair assumption, but the result is interesting. The result for each season was exactly the same (73.9 percent), which is much closer to league average than Cueto's actual LOB percentage. The last column shows how many runs those runners would have scored had they stayed on base, based on Tom Tango's Run Expectancy Matrix for 1993-2010.
I think this actually backs Cameron's point pretty well. It seems that Cueto does save runs with his ability to control the running game, especially once we consider that Cincinnati Reds catchers throw out only about 21 percent of runners on average when Cueto isn't on the mound. It also seems that Cameron is probably right, with his theory that Cueto strands more runners than average, because of his ability to hold and pick runners off. This is pretty interesting stuff, but it really has nothing to do with the point I'm trying to make.
So, back to pitching wins. In neither article did the author discuss Cueto's win total, although the Sports Illustrated article did cite it in the post. I'm happy that the authors didn't discuss the total because I don't think any Cy Young discussion should involve wins. But at the same time, I think subconsciously both of these authors thought about Cueto's win total.
I really wonder whether Cueto is a Cy Young Candidate because he's been a great pitcher or if he is a Cy Young candidate because he's been a good pitcher who has a lot of wins.
Although, Hernandez won a Cy Young with just 13 wins, my theory is that authors of "Who is going to win the Cy Young"-type articles think that the voters still care about pitching win totals (and they're probably right). For instance here's Corcoran's top three for each league's win totals vs. their ranking in the league based on FRA (Fair-Run Average):
|Pitcher (League)||Wins (Rank)||FRA (Rank)|
|1. Justin Verlander (AL)||13 (t-8th)||3.31 (2nd)|
|2. Felix Hernandez (AL)||13 (t-8th)||3.24 (1st)|
|3. David Price (AL)||17 (1st)||3.78 (6th)|
|1. R.A. Dickey (NL)||18 (t-1st)||3.76 (t-7th)|
|2. Johnny Cueto (NL)||17 (3rd)||4.08 (16th)|
|3. Clayton Kershaw (NL)||12 (t-16th)||3.31 (t-2nd)|
FRA isn't a perfect statistic, but it's probably the best we have to use right now at describing what the pitcher actually deserved to give up on the field. It seems to me that wins and actual performance are still weighted equally. According to Rob Neyer's Cy Young Predictor it seems as though pitching wins are weighted even more heavily than actual performance.
If Price, Dickey and Cueto had 12-13 wins would they be ranked so high?
My guess is probably not, and I think the authors would admit that as well.
I'm just confused as to why pitching wins matter. Why can't Cueto win the Cy Young because he's been incredible run preventer and his pickoff move allows him to prevent runs even more effectively? Why are his 17 wins such an important factor in their decision, and more importantly in our perception of Cueto as a pitcher?
I'd also like to know whether a starting pitcher ever will win the Cy Young Award with a single-digit win total? Or if mainstream baseball fans will ever admit that a pitcher like Lee, with such a low win total, is actually having a really good season, despite his record?
The sabermetrically inclined Phillies bloggers at Crashburn Alley watch Lee every fifth day and understand how much better he's been than his win total, but most do not perceive him in this way.
Our perception of Lee is that he is having a down year, yet he leads Cueto in a bunch of pitching categories. But for some reason Cueto is the Cy Young candidate and Lee is not. The difference in their ERAs is obviously a large factor for Cueto being perceived as better, but I still think the wins are a major factor in distorting our perceptions of the two pitchers.
For instance, many people use Quality Starts (at least six innings and less than three earned runs for the starter) as a better substitute for wins. The statistic essentially tells us whether the pitcher gave his team the chance to win, but so many other factors affect who actually wins the game. Cueto has more Quality Starts (20) than Lee (16), but if we convert the statistic into Fair Quality Starts, which is the same statistic, but based on the runs the pitcher should have given up, then Lee ranks ahead of Cueto (15 to 14). If we use FAIR_QS instead of wins, Lee may in fact look like the better Cy Young candidate.
It may sound crazy to consider Lee a Cy Young candidate, but consider this fact: He ranks second in xFIP and SIERA, as well as is in the top five of FRA Fair_QS and also ranks in the top-10 in FIP.
Just because Felix Hernandez won a Cy Young with 13 wins does not mean the voters ignore wins now. It means they ignored it once. It still alters perception and is a major factor for Cueto being a Cy Young candidate, while Lee is nowhere near the conversation. I am in no way trying to say that Cueto shouldn't be a Cy Young candidate, because he should be. But it's just stupid that baseball fans still look at Lee, as not so good, but Cueto as great.
It's all a matter of perception, and pitching wins still distort our perceptions, no matter who we are. They're still published everywhere (including sabermetric sites) as part of each pitcher's main stat line, so they're almost impossible to ignore. They need to go. Fantasy leagues should stop using them, Cy Young and Hall of Fame voters should ignore them and baseball announcers, and fans should start ignoring them.
I know this argument has been beaten to death by sabermetric columnists for years, but I had to say something about it. The day probably will never come when pitching wins are eliminated from baseball's culture.
But you know what?
I'm going to keep dreaming for that day, and you should too.
References and Resources
Statistics for this post came from all over the place, including Baseball-Reference, Baseball Prospectus and FanGraphs.