The Mets used four pitchers last night and the one who likely did the worst job on the night – Ramon Ramirez – came away with the Win. That’s the way it goes sometimes because no statistic is perfect. The annoying thing is that many people will forgive the imperfections of certain stats, like Wins, yet are not willing to give other stats a chance because they focus only on the imperfection of the non Triple Crown stats (either pitcher or hitter version).
And this is not limited to just the alphabet soup metrics like DRS, FIP, UZR and WAR. Something as simple to understand and calculate like the Quality Start frequently gets dismissed because you can get one by allowing 3 ER in 6 IP. That’s a 4.50 ERA – the horror, the horror! But as we saw last night, Ramirez got a Win by allowing 1 ER in 1 IP. Yes, you can get a Win with a 9.00 ERA. If Ramirez had allowed two runs last night – an 18.00 ERA – he would have gotten the Win, too.
Ramirez’ 1 IP, 1 ER and a Win outing was the 16th time that has happened in the National League this season. There are at least 133 times this season that a pitcher picked up a Win in the NL despite having an ERA of 4.50 or above for his outing. There are probably a handful more but figuring out the exact number is not worth the extra time it would take me to do all the combinations.
On the flip side, there have been 1,144 Quality Starts in the National League this year, which is more than the 1,047 Wins accumulated by NL pitchers. Yet there have been exactly 112 times a pitcher earned a Quality Start with a 4.50 ERA. So just under 10% of Quality Starts fit the minimum requirements that get people so upset. And there are fewer QS with a 4.50 ERA than there are with Wins at or above the same threshold.
I like Wins because that’s one of the stats that I grew up with and I know both its strengths and weaknesses as a number. Wins have come under a great deal of scrutiny lately, due in no small part to Felix Hernandez winning the Cy Young Award in 2010 despite having a 13-12 record. Also there is the questionable wisdom of assigning a team outcome to an individual.
Yet for all of its pitfalls, I still use Wins to a limited extent to value a pitcher within a team context. Just as it’s a mistake to judge a pitcher primarily on Wins, I believe it’s a mistake to completely dismiss them, too. But if given a choice of only one number, I would rather know how many QS a pitcher had than his Wins.
I know there are some people reading this who do not like Quality Starts. But once you can look past the limitations of the statistic and see how it does in its entirety, I believe you will accept its usefulness. And once you do that for one stat, you can do the same for others. Perhaps QS can be the “Gateway Stat” to accepting and using some more advanced numbers that are out there.
Jeremy Hefner did not get a win last night but he pitched a strong game and deserves credit for it. He did not get the Win but he got a Quality Start with his 6 IP, 2 ER performance. Hefner has just 2 Wins this season, but he has 6 QS in his nine games as a starter. If you just looked at his Wins, you would not think much of his season, an impression that would probably be bolstered by looking at his 4.52 ERA.
But any pitcher who can give a QS in two-thirds of his outings is doing something right. For a few comparisons, R.A. Dickey has 17 QS in 27 starts or 63 percent of the time. Jonathon Niese has 10 QS in 26 starts or 38 percent of the time.
Obviously, Hefner has put up his numbers in a smaller sample size and this is in no way a plea to suggest that he bump Niese from the rotation next year. It’s just that QS gives a nice way to judge a pitcher and it indicates that Hefner has pitched better than either the traditional stats Wins or ERA would suggest.
I used to not care about QS’s, but recently this season i have grown a liking to this stat. The reason i didn’t like it before was exactly the reason writtein above. How could a 4.50 ERA outing(3 ER in 6 IP) be quality? I certainly don’t want a pitcher who has a 4.50 ERA on my team! But if you think about it more closely, if your pitcher doesn’t give up more than 3 runs in a game, you are always in the game, which in my view is the main job as a starter. Does a 3-0 loss sound that bad? I certainly don’t think so.
QS also isn’t as volatile as ERA. For example, lets say you had 2 outings of 6 IP and 1 ER’s. That’s a 1.50 ERA. Then let’s say you have a start where you go 6 IP and 8 ER’s. Your ERA jumps up to 5.00 ERA. But in terms of QS, you have 2 QS out of the 3 tries. In other words, in ERA, you are penalized for every run you give up, but in QS, every run after the 3rd run doesn’t affect your aggregate score.
However, there are flaws to it such as if you only give up 1 hit in a game, but that hit is a grand slam, you can’t get a QS even though you only made 1 mistake. But of course no metric is perfect.
Brian, great column.
No statistic is without its limitations. If stats were the be all and end all of baseball evaluations teams wouldn’t need scouts.
Yes, stats are valuable, some more than others. They must be taken in context (example: e.r.a and relief pitching— a limited value stat).
I like your take on QS. Takes as part of the whole picture of a hurler’s evaluation it is perfectly valid.
Middle relief pitchers are the weakest players on the pitchinf staff. Starting pitchers that have a good quality start percentage minimize the use of these pitchers. They are consistant and reduce the stress put on the bullpen.
To apply the concept a step furthur;the average NL starter goes 6 innings per start or 18 outs. A starter that can get the 19th, 20th or 21st out is even more valuable because the team is avoiding using one or two of the inferior middle relief pitchers to get those outs. The fewer inferior pitchers used to close out a game the greater the chance of being sucessful. I would prefer a starting pitcher that can consistantly go 6+ innings a start over an “electric” armed starter that barely gets the team to the 6th inning. To me,that is the value of the quality start percentage stat.
Very good point about minimizing middle relievers.