Josh Collmenter, Mr. Deception | Disciples of Uecker

Disciples of Uecker

We'd like to go to the Playoffs, that would be cool.

The Brewers will face Josh Collmenter on Tuesday in game three of the National League Division series. In two starts, Collmenter has stymied the Brewers, striking out 10 against just one walk in 14 shutout innings. It’s easy to see why Kirk Gibson is willing to throw his rookie to the wolves in game three — he’s handled these wolves before.

Collmenter’s success is so confusing because he doesn’t look like a real pitcher. He has this weird, pitching-machine style deliver where he turns his hand over in a style most hitters have never seen before. In pretty much every feature on the 25-year-old righty, his deceptive delivery is mentioned. A google search for “Josh Collmenter Deceptive” returns a whopping 37,600, over 20% of the number of results for just “Josh Collmenter.”

Snakes on Jefferson’s Tyler Emerick sees this as cause for concern, as he described in my Q&A with him last week:

As crazy as it sounds, and as unpopular as it might be, I’d go with Joe Saunders in game three, Ian Kennedy in game four and avoid Josh Collmenter at all costs. I know Collmenter didn’t allow a run to the Brewers in 14 innings but I just can’t seem to trust him. His delivery, in my book, is still the only thing he has going for him. And I believe Milwaukee will spend enough time the next few days figuring it out. I can’t imagine the Brewers not hitting him a third time around.

Tyler goes on to present some numbers for Collmenter’s second times (and beyond) against teams. They’re not good. Allow me to expand.

In total, Collmenter has appeared in 31 games, starting 24 of them. He has pitched 154.1 innings, giving up 58 runs for an ERA of 3.38. His peripherals have been very good as well: 100 strikeouts to 28 walks and a moderate 17 home runs allowed. His 3.80 FIP is significantly worse than his raw ERA, but still good.

Now, it’s rarely helpful to cut out a significant portion of your sample size when analyzing pitcher performance. In order for the split to be particularly useful, not only does it have to be more predictive than the innings you’re cutting out, but you also have to be reasonably sure the innings that are cut out are either detracting from the analysis or, at the least, shouldn’t be considered very meaningful at all. Also, the effect of the split you’re looking at has to be big enough to make it worth looking at a smaller sample.

However, at the very least, we do have records of what actually happened in these second starts and beyond for Collmenter. And if deception — and particularly, a lack of familiarity with said deception — is the reason for Collmenter’s early success against teams, then we might have something meaningful here.

Key word: might. Now let’s continue.

With Collmenter’s first appearance against each team removed, we lose a significant portion of our sample — down to 15 starts and 86.2 innings pitched. Collmenter has been significantly worse in these starts, allowing a 4.15 ERA. His 54-17 strikeout-to-walk ratio is great, albeit low on punchouts, but 12 of his 28 home runs have come in these more familiar starts, a 1.24 HR/9 ratio which essentially explains the struggles. His FIP in these starts sits at 4.34.

Slicing and dicing even further, removing even the second time a team has faced Collmenter, the trend continues. We’re down to only seven starts, predictably only against NL West opponents (SF once, SD twice, LA once, and COL thrice). In these seven starts, Collmenter has lasted 39.2 innings and allowed 23 runs (all earned) for a 5.22 ERA. Again, the strikeout-to-walk numbers are good, at 20-to-5. It’s the home runs. The rate continues to rise as the familiarity increases, up to 1.59 per nine innings. As a result, his FIP sits at 4.86 — not as bad as the ERA, but dangerously close to replacement level.

Given how widespread the thoughts of Collmenter’s deception are, I wouldn’t entirely dismiss these small samples. The Brewers will have a good idea of what’s coming, and there is no denying that other teams have done better in their second or third times against the young righty. Whether or not that is predictive, or even if it is predictive, whether it can manifest itself in runs when Milwaukee faces Arizona on Tuesday night remains to be seen. Just don’t be shocked if the Brewers knock a couple out of the park against Collmenter now that they know what they’re up against.

Share Our Posts

Share this post through social bookmarks.

  • Delicious
  • Digg
  • Newsvine
  • RSS
  • StumbleUpon
  • Technorati