clock menu more-arrow no yes mobile

Filed under:

Looking at Astros Pitchers’ “Luck” in the first month

Most advanced statistics are generally based around identifying the contribution of the player and removing other aspects that could be associated with “luck” (items outside of their control). These statistics are great because they can help predict positive or negative regression to what the player is truly doing in a more accurate fashion than the eye test or the traditional statistics. (Stats are from 5/1/2019 prior to the game) We’re still very early in the season, so there’s the standard Small Sample Size caveat as we dig into these numbers.

For pitchers, I will look at ERA vs SIERA and wOBA vs xwOBA

SIERA - Skill-Interactive ERA - Similar to FIP and xFIP looks to eliminate the “noise” and focus on a players true contributions. The difference is FIP looks at 3 variables (strikeout, walk, home run) but relatively ignores balls in play. xFIP does the same thing, but normalizes the HR/FB rate. SIERA follows a similar path but puts context to the other balls put in play to get a more accurate look. SIERA is a much more accurate measure of the pitchers performance in the past, not a predictive measure.

xwOBA - Expected Weighted On-Base Average - xwOBA is still somewhat new and experimental, but utilizes a bit more scientific of an approach. Each and every time the ball is hit into the field, it measures the exit velocity, launch angle, sprint speed etc - to determine what percent of time that hit would go for an out, single, double, triple, home run etc - and creates a value for it.

Justin Verlander - 2.45 ERA / 3.29 SIERA /// .259 wOBA / .303 xwOBA

Verlander has outperformed his expected statistics, although even the advanced stats have indicated he’s been pitching like an ace.

Gerrit Cole- 3.95 ERA / 2.86 SIERA /// .290 wOBA / .282 xwOBA

Gerrit Cole has been extremely “unlucky” in his starts so far this year, and has actually pitched better than anyone else in the rotation. His SIERA would be ranked 4th in baseball right now, so I’d look for him to have a large correction this month.

Collin McHugh - 4.78 ERA / 3.68 SIERA /// .305 wOBA / .283 xwOBA

Collin McHugh has also suffered from very poor luck to start the year. This wasn’t overly surprising to me as McHugh was a top performer minus his one start against the Rangers. I’m hopeful to see him continue to get extended out.

Wade Miley - 3.24 ERA / 4.66 SIERA /// .290 wOBA / .285 xwOBA

Miley’s performance so far in the year was very surprising to me. SIERA does not support it, but there has not been a very large differential in his woba vs xwOBA.

Brad Peacock - 3.67 ERA / 4.11 SIERA /// .261 wOBA / .250 xwOBA

Peacock is an interesting case where his wOBA and SIERA standards don’t seem to align well. I’ll do some more research on him and may do a whole article depending on my findings.

From a starters perspective, the strikeout / control numbers are oddly interesting to me as well. With Gerrit Cole ranking #1, Justin Verlander #12, and Collin McHugh ranking at #17 in K/9. Peacock doesn’t show in the starters rankings but would come in at #54 and Miley in at #86 (of 87).


Roberto Osuna - 0.73 ERA / 3.02 SIERA /// .068 wOBA / .138 xwOBA

I did an article on Osuna, asking if he was the best reliever in baseball, largely based around the wOBA/xwOBA statistic. He has been unbelievable this year. His K/9 has not returned to the levels we’d like yet, which hurts some of his advanced analytics, but he’s yet to walk a batter this year.

Ryan Pressly - 0.00 ERA / 1.42 SIERA /// .114 wOBA / .168 xWOBA

Bill of course countered my Osuna article, and rightfully so. Pressly has been outright dominant since his trade to the Astros and shows no signs of slowing down. He is currently chasing the Astros all-time scoreless innings streak record. Pressly also has had a dip from last year’s strikeout rates, but has also yet to give up a single walk.

Will Harris - 1.93 ERA / 3.09 SIERA /// .225 wOBA / .264 xWOBA

Will Harris always feels like the forgotten man who continues to put up above average numbers. Right now, the analytics show he has gotten some good luck to assist in driving his numbers down this year. I’d expect him to return closer to his career averages, but with a 3.05 ERA across 345 innings - that’s not really a knock on him.

Hector Rondon - 2.53 ERA / 3.74 SIERA /// .314 wOBA / .290 xwOBA

Rondon is an interesting case, he’s a mixed bag having received some poor luck when looking at the wOBA side, but an 86% strand rate and some luck in the HR/FB department has kept his ERA surpressed.

Josh James - 7.36 ERA / 4.17 SIERA /// .368 wOBA / .281 xWOBA

Josh James has been in the cross fire for quite a few fans. After him storming into the MLB last year flashing 100+ mph fastballs, he has failed to impress. The good news is that he has not been nearly as bad as it initially may seem. He’s had some terrible luck, but he still has not been replicating his performance from last year. I would not be surprised if he’s been trying to play through the injury that was found in Spring Training and if he makes a brief stint to the IL.

Overall, as you’d expect with such a small sample size, there’s a mixed bag of results from a regression standpoint. The Astros as a whole currently sport the 4th best team in regards to their pitching from a WAR perspective. Their 3.37 ERA would rank in 3rd, but the analytics show a 3.57 FIP, 3.67 xFIP, and 3.58 SIERA which indicates they’ve been slightly fortunate so far.

With that said, the Astros had some fierce competition in the first month, technically not playing a single team that was below .500 going into the series. They had the hardest schedule in the AL to begin the year. You would have to assume as they hopefully feast on the lesser teams in the next month, they would also get some benefits in regards to their stats.