Kill The Win, Then What?

There is one statistic in the land of baseball that is getting a lot of criticism lately and that is the pitcher “win”. Most argue that the win is one of the most misleading statistics in the game. On a single evening, one pitcher in Los Angeles can go 5 innings, give up 7 runs and get a “W” because his offense supported him. In Seattle, some guy can throw a complete game 5 hitter with 2 earned runs and lose 2-1 because the Seattle offense can be horrendous at times. Let me ask you this, would you rather have King Felix or Joe Blanton? Examples like this are why I have jumped aboard the #KillTheWin bandwagon, which is currently being championed by Brian Kenny of the MLB Network.
Brian claims, as most of us do, that the win is no longer useful, because it tells us nothing except a pitcher went a minimum of 5 innings, his team was ahead when he left the game and they went on to hold the lead for him. It tells us nothing else. Nothing. It doesn’t tell us how effective he was or how deep in a game he went. It gives us no other criteria in which to judge the performance of any one pitcher on any one night.

One thing I would like to see, if we killed the win stat, which by the way, probably will not happen, but if we did….replace it with a stat that told us how a pitcher pitched that night.

My suggestion is that we change the “Quality Start” to better reflect quality and not mediocre performances. Currently, to qualify for a quality start a pitcher must complete 6 innings and give up no more than 3 earned runs. Those qualifications seem to be a bit pedestrian considering the aforementioned Joe Blanton who has a 5.39 ERA and a WHIP of 1.525 currently has 9 quality starts in 2013. If we simply increased the expectation from the QS to a minimum of 7 or more innings pitched with no more that three earned runs allowed, that would make Blanton’s QS stat drop to 5. If you look up his games, Joe was terrific in those 5, where his record is 1-3 with a no decision. In those 5 games, the Angels have scored exactly 4 runs in support of Blanton, but it tells us he did his job those five times.

The basis behind moving the number of innings pitched to 7 instead of 6 is quite elementary. If someone throws 6 innings and gives up 3 earned runs, then the ERA for that performance is 4.50. An ERA of 4.50 over a full season would be far from “quality.” However if you increase that to 7 innings pitched, then the ERA for that performance is a much better 3.84. Now, that’s not setting the world on fire, but at least that approaches the baseball definition of quality over a full season. An ERA of 3.84 and the pitcher lasting more than 3/4 of the game versus 2/3 of the game and an ERA of 4.50 is a much better. We could get more in depth in this with WHIP and other criteria, but for the time being let’s make this as simple as possible.

If we take away the W-L record, and changed the QS to 7 innings minimum, how do we keep track? The win-loss record has been around forever, so how do we replace it with quality starts and have it make sense? Simple. Quality Starts (QS), No Decisions (ND) and Lost Starts (LS). Lost starts would be calculated for when a pitcher really takes one on the chin. Something in the neighborhood of 5 innings or under and 4 or more earned runs allowed. Anything in between the QS and the LS would simply be a no decision.

For example, Derek Holland in 2013 has a record of 9-6, and anyone that has seen him pitch this year can certainly agree that he is better that his won-loss record. Through the eyes of a quality start, you’ll see that it more accurately shows how Derek has pitched in 2013:

                               Starts     QS     LS     ND     ERA     WHIP
Holland (2013)       23         12       1       10      3.02     1.252

Max Scherzer for example, would no longer be 16-1, but 12-0. This isn’t quite as sexy as 16-1, but Max getting 2 wins in April with giving up a total of 9 runs in those two games is not very sexy either, he’s just lucky the Tigers put up a bunch of offense, and we don’t see that in wins and losses.

Holland and Scherzer are not the best example for killing the win. For that we’ll use the obvious, Felix Hernandez’s 2010 season. If you remember King Felix won the American League Cy Young Award with a record of 13-12 and an ERA of 2.27. Some old school baseball scribes either didn’t vote for him or outright complained that his won-loss record did not warrant the Cy Young Award, or both. So, lets break it down so that the Phil Rogers’ and George Kings’ of the world can understand. By the QS criteria we have established, Hernandez’s stats would have looked like this:

                                    Starts     QS     LS     ND     ERA     WHIP
Hernandez (2010)       34        25       2       7        2.27      1.099

25 quality starts out of 34 starts looks infinitely better than a 13-12 record for a team that couldn’t score. Felix took the hill 25 times and went a minimum of 7 innings and gave up 3 runs or less. That is why he should have and did win the 2010 AL Cy Young.

Now, this is a total change in thinking for baseball but it’s all about perspective. Obviously, 20 QS would be a feat since maintaining that level is nearly a Herculean task, but as we saw in the above table, it can be done. And how many 20 game winners do we see these days? Not a whole lot. However a bar of 18 QS doesn’t seem to far out of reach for the elite pitcher does it? Under this format, Clayton Kershaw has 18 QS as of today.

This is far from a foolproof plan. But at least it tells us more than simple wins and losses. For example Yu Darvish would have 9 QS and 0 LS. He leads the league in strikeouts and is near unhittable at times. However Yu is averaging 3.1 walks per game, which is something ace’s don’t do. Darvish is a fantastic pitcher, but ace’s go deep into games and don’t get their pitch count elevated like he does.

As I said, it’s not foolproof, but it’s a start. I think using quality starts gives us a much better idea of a performance. Knowing that out of 34 outings in 2010, Felix Hernandez had 25 quality starts by the criteria we established is not only astounding, but a truer reflection of his year. Obviously, unless somebody is absolutely terrible over a period of time, their most likely won’t be bunches of Lost Starts, but Major League pitchers aren’t terrible for long stretches. If they are, they aren’t Major League pitchers for long.

The won-loss records would go out the window in this scenario, but at least the QS stat is close to the pitcher win totals of the day. I for one will begin using this instead of pitcher wins and losses on this website until someone comes up with something better.

That’s my idea, what’s yours?

Patrick Despain is the CEO and Co-Founder of ShutDown Inning. He can be reached at or on Twitter @ShutDownInning
Patrick Despain
Patrick is a member of the IBWAA and creator of Shutdown Inning. He was raised him Arlington, Texas and grew up watching games on HSE and listening to Eric Nadel and Mark Holtz on the radio. He is a long time Rangers fan and never achieved his dream of being a bat boy. He know lives in Georgia with dreams of a Texas return.

Leave a Reply