The win is a horrible measure for a pitcher, the quality start is garbage: Introducing QS+

Spanning back several years at this point, Ive challenged both the Win and the Quality Start. I believe there are not only flaws in our interpretation of the best pitcher starts on an individual basis (Quality Start), but also our reward system for them (Win).

Spanning back several years at this point, I’ve challenged both the Win and the Quality Start. I believe there are not only flaws in our interpretation of the best pitcher starts on an individual basis (Quality Start), but also our reward system for them (Win).

The argument against Wins is much simpler in nature and harder to refute, so let’s knock it out quickly and move on. The only parameter for a Win in the pitcher’s control is the completion of five innings. So technically, a starter can give up a three-run homer each inning, surrender 15 runs through 5 IP and still be in line for a Win (with a 27.00 ERA, mind you) providing his team can score 28; something absolutely out of his control (more so with the introduction of the Universal DH). Wins reflect neither the analysis nor player skill. It’s an antiquated stat, that much is certain. However, without a viable replacement, it would be impossible to make forward progress. Enter the Quality Start a few decades ago, defined as a pitcher start of ≥6 IP and ≤3 ER, which at least is performance-based.

Advertisement

Though a marked improvement, it doesn’t take much scrutiny to expose the weakness of the Quality Start for modern fantasy baseball. A baseline Quality Start of 6 IP, 3 ER boasts a 4.50 ERA, a ~50th percentile metric in earned run allowance. It doesn’t properly serve as a benchmark for excellence, or as a 5×5 roto category. Again, not one to solely complain, I sought after the solution. By adjusting parameters to make the distance and earned run values more representative of per start excellence, I ended up creating the Money Start (aka “QS$”); which I defined as an outing of 7 or more innings pitched and two or fewer runs allowed to highlight the very best single games (these can immediately be rewarded proportionately in a point league).

The Money Start proved extremely helpful immediately throughout my analysis (it’s specifically impactful for best-ball leagues that reward the good and discard the bad). For all intents and purposes, the Quality Start was replaced, in my eyes. However, there was a problem with Money Starts if we were indeed going to Lose the Win; a distinct lack of frequency. For reference, in 2021, there were 1,584 Quality Starts, 1,346 Wins, and only 537 Money Starts. There are too many Quality Starts and not enough Money Starts.  So, it was back to the chalkboard.

I began to doubt if my idea of replacing the Win with a single stat best reflecting both analysis and skill could work. If it was ever to succeed and be widely accepted, I’d have to increase the number of events without compromising its integrity. I decided to table the Money Start for the time being despite being very pleased with its ability to describe the best pitcher starts. I had to go back to retest distance with another full year of data at hand to see if I missed anything. I initially set 7 IP as the benchmark for a Money Start qualification with reason at the time. In 2017, a starting pitcher getting at least 18 outs was only a 50th percentile distance metric. However, with the inception of openers and the furthered specialization of bullpens, I feel anecdotally like we have witnessed the trend toward the quick hook continue, even accelerate. Let’s pick this work back up with our newly added data from 2021 to reassess MLB starter distance. I went back five seasons for two perspectives; the number of games started of +6 IP, but also the percentage of the total to account for the shortened 2020 season:

Regardless of the image you prefer, there is a distinct, undeniable, and ongoing downward trend in starts of at least six innings over the last several years. We’ve had consecutive seven and eight percent year-over-year drops from 2017 to 2019, followed by another 15 percent plunge spanning 2020-2021. The overall 27% drop is all the evidence I need to reconsider eighteen outs as a sufficient benchmark for inclusion. Refusing to relinquish my position on an ERA north of four, I went back to Quality Starts as originally constituted to dissect ERA by out and run combination to look for some type of fat to trim.

Advertisement

Whenever I run into obstacles, I try and take a step back to get a better look at the whole picture. Below is the difference in earned run average by out in the sixth, seventh and eighth innings.

One of these things is not like the other. One of these things just doesn’t belong. Can you guess which one of these things, by the time I finish my song? Look at the final six-inning bucket of stats.  The entire bottom row sticks out like a sore thumb; it has to go and take that ugly +4.00 ERA with it. Considering the rest of that final +6 IP board is fantasy viable in terms of ERA, plus the weight of today’s discovery regarding the increased value of going six innings, I feel strongly we have data-backed reasoning to include ≥6 IP, ≤2 ER with our desired outcomes.

Could it be done this easily? Could the answer to all of our questions lie with the simple removal of a third earned run in the sixth inning only? It would take some finesse to properly mine but I had to calculate the new frequency of this outcome if we removed only +6 IP, =3 ER from the original Quality Start. This new metric, the Plus Start (“QS+”) would add outcomes to the total, yet still remain within all of our parameters and represent a wholly skilled-based and data-backed metric. Let’s test its frequency against the win in terms of total and percentage, shall we?

My first instinct was shock seeing how closely Plus Starts (QS+) tracked Wins (W) over the last half-decade. In regards to their total occurrence, the two metrics were within 38 events (less than two percentage points) of each other over a 6,654 sample spanning two seasons. Let’s drive the final nail in this coffin home as I’m sure certain detractors will equate the movement in Plus Starts with Wins, but this is absolutely not the case.

In 2021, there were 1,584 Quality Starts — only 864, or 54.5% of those, resulted in Wins, while there were 1,345 pitcher wins awarded to starters who didn’t complete six innings. From 2020 to 2021, there were 118 starter wins of four earned runs or more, 24 starter wins of five runs or more and four starter wins with six runs or more. What does this mean to us? It means this exercise was a wild and unabashed success. Here we stand, having provided detailed validation for a wholly skilled-based and data-backed metric that is viable from a volume standpoint, and an easy replacement for an irrefutably flawed category responsible for ten percent of our scoring.

Advertisement

Having removed subjectivity and random luck from the argument is the best we can do to back detractors into the proverbial corner. Some will gnash their teeth, others hem, and haw, and others still will scream about the ways things have always been. Might I remind the latter that rotisserie baseball was once 4×4, not 5×5, and it wasn’t until the collective spoke out for progress that we saw a meaningful, and long-lasting change.

That’s my Plus Start Redux incorporating 2021 data. It was a long and arduous road, but well worth it in the end. On one hand, we have Money Starts to isolate pitchers with the most (or highest frequency of) spike weeks and on the other, we have Plus Starts which could and should be used in place of Wins as a category. Hopefully, this piece will somehow sound the klaxon for change. When you think about the amount of work and money we spend on fantasy baseball, it’s hard not to argue for fairness. Be on the right side of history, people!

With that out of my system, let’s apply some of these cool, new custom statistics to MLB 2021 (Full and sortable leaderboard provided at the end)

14 Pitchers Had At Least 8 Money Starts (≤7 IP,≤2 ER) In 2021:

11 Pitchers Had A +30% Money Start Percentage (QS$%) In 2021:

  • Adam Wainwright 40.63%
  • Walker Buehler 39.39%
  • Sandy Alcantara 39.39%
  • Zack Wheeler 37.50%
  • Chris Bassitt 37.04%
  • Framber Valdez 36.36%
  • Tyler Glasnow 35.71%
  • Max Scherzer 33.33%
  • Wade Miley 32.14%
  • Shohei Ohtani 30.43%
  • Gerrit Cole 30.00%

12 Pitchers Had +17 Plus Starts In 2021:

10 Pitchers Had A +59% Plus Start Percentage In 2021:

  • Jacob deGrom 73.33%
  • Walker Buehler 72.73%
  • Adam Wainwright 68.75%
  • Trevor Bauer 64.71%
  • Brandon Woodruff 63.33%
  • Corbin Burnes 60.71%
  • Sandy Alcantara 60.61%
  • Max Scherzer 60.00%
  • Zack Wheeler 59.38%
  • Chris Bassitt 59.26%

Top 10 Unluckiest Pitchers: 6 or More Plus Starts Than Wins (Deemed ‘Cheated’)

Top 5 Luckiest Pitchers: 5 or More Wins Than Plus Starts

That’s entirely enough out of me for the time being; now get out there and enjoy the fruits of our labor! Feel free to use the Money Start leaderboard below in your analysis and make sure to get the idea of the Plus Start in front of your league Commissioner immediately! Even if you don’t enact it right away, spend the 2022 season tracking it next to the Win and I promise you the upgrade will be incontestible.

FULL 2021 LEADERBOARD: https://docs.google.com/spreadsheets/d/18lC8_3KSYdwnUOSTDZDEYBKs4kk3D5aLhF27kpoApA4/edit?usp=sharing

If there’s anything you have questions on, disagree with, or even would like to expand on, please leave a comment below or follow my Twitter handle @MLBMovingAvg. Don’t forget to check out the Corked Stats audio podcast, and also the Corked Stats video series on YouTube at the Mayo Media Network.

(Top photo: Mark Brown/Getty Images)

ncG1vNJzZmismJqutbTLnquim16YvK57kmpscm1hbHxzfJFrZmlrX2eAcMPIp6pmmZ6ZerLBwKWgrbFdqMGivtOsZKKmXZuur8DArLBmpJWWsaa%2BwaiYq5xf

 Share!