Dissecting a Dreadful May

On the last day of April, the Cardinals improved to 19-10 with a 3-2 victory at Nats Park. A day later, Miles Mikolas kicked off May by outdueling Max Scherzer in a 5-1 victory. The win gave the Cardinals the best record in baseball. All was right in Cardinal Nation. Then, all hell broke loose. They’ve won just eight games this month. That’s fewer than any full month (20+ games) since 1994. Much like an awkward middle school student, let’s dissect this dead frog to see what knowledge we can discover.

Through April, the worst facet of the team was starting pitching. They ended April with a 4.49 xFIP (19th in baseball), 5.52 FIP (third worst), and a K-BB% of 11.9% (25th). Complicating matters, they had even been a little bit lucky with an unsustainable 81.2% LOB%. It stood to reason that a course correction in the starting pitching would portend an impressive May for the team even if the LOB% regressed.

The xFIP for starters in May improved to 4.17 (9th in MLB), their FIP diminished by over a run down to 4.16 (13th), and their K-BB%... well, it got worse (11.2%), but they shaved nearly a full homerun per game off of their average in May. Here’s how it looks for starters this month:

That table doesn’t include Hudson’s Thursday start, but his FIP likely held serve or decreased. Wacha continued to be a disaster, which spilled into one of the ugliest outings I can remember on Wednesday night. Wainwright’s improvement was modest and inconsistent. However, Mikolas, Flaherty, and Hudson made significant strides. It didn’t matter.

Other components also fared well this month. Using FanGraphs’ BsR to gauge baserunning, they were third best in baseball in April and second best in May. FanGraphs doesn’t offer splits by month for Defensive Runs Saved, but we can use their DEF stat. They were sixth best in baseball in April and 10th in May. It’s a slip, but still above average. The bullpen’s FIP went from 4.81 in March and April down to 4.05 in May, ninth best in baseball during the month. Before Wacha’s blowout on Wednesday, it was 3.51- third best in MLB.

Both the bullpen and rotation collectively got better (much better if we exclude Wacha and Wainwright), the baserunning has been better, and the defense has been only slightly worse. In spite of all of that, they’ve had their worst month since that kid in the Blind Melon video put on her bee suit. What the hell happened?

Perplexing Offense

The offense slipped from scalding hot in April to feeble in May. This table tells a story:

They walked a little less in April, struck out less, hit the ball hard more frequently, and their expected wOBA slipped a little (though it still ranks 10th in baseball during May). The hard hit data is true whether you prefer FanGraphs’ hard hit percentage or Statcast’s exit velocity data. It’s worth noting that Thursday’s game surely helped the ISO, but the first two games in Philadelphia did major damage to the plate discipline stats. Before Tuesday, their BB% was actually higher in May, and their K% was even lower.

Somehow, that quality of contact and plate discipline translated into an astonishing 34-point drop in wRC+. Other parts of that table reveal what happened. Their BABIP collapsed, and about 1% fewer of their balls in play were optimal (between 22 and 38° launch angle, 95 mph or more). Even accounting for fewer optimal balls in play, they undershot what an average team would do with that quality of contact by .030 of wOBA. Add the BABIP struggles and it’s easy to see how production evaporated.

Collectively, the process was very similar to April, but the results were considerably worse. Do the same things again in June and they’re more likely to see better results.

Ol’ Pythagoras Has Done it Again

I’ve collected every game for every Cardinal team since 1969- the last 50 years. We can determine their monthly runs scored, runs allowed, pythagorean record, and compare that to their actual record. The more negative the gap, the more they underperformed their pythagorean record.

May 2019 has been the 18th worst month in the sample out of 318 months, with the team underperforming their pythagorean record by 2.56 wins. Their pythagorean record says they should have 10.56 wins this month, but they’re 8-18.

Of course, even if they performed to their pythagorean record this month, they’d have 10 or 11 wins. That’s an atrocious month. Bad luck or otherwise, there’s no sugarcoating that they’ve been bad. We’re simply parsing the degrees of bad.

One-Run Games

The Cardinals have lost seven consecutive one-run games in May. You have to go back to July 1970 to find that many one-run losses and zero victories in the same month.

One-run wins and losses are an efficient way to cause a gap with your pythagorean record. That’s all well and good, except one-run results border on random. They aren’t totally random- a few aspects of a team can help sway their record one way or the other in one-run games- but reliability of the effect is volatile. For instance, a good bullpen fWAR can help a team win one-run games. The Cardinals May bullpen fWAR through Wednesday is 0.5, which is 13th in baseball. The raw performance of the bullpen hasn’t been the culprit.

Simply playing close games at home instead of the road gives teams an edge in one-run games. Clutch performance is another factor (FanGraphs’ Clutch rating). From the FanGraphs page:

In the words of David Appelman, this calculation measures, “…how much better or worse a player does in high leverage situations than he would have done in a context neutral environment.”

Even taking all of that into account, you should expect a team’s record in one-run games to be close to their pythagorean record, and the Cardinals aren’t even close in May. You’d normally expect anywhere from 2 to 5 wins in those seven one-run games. Not zero.

The Worst Performances in the Biggest Moments

Let’s dig a little deeper on Clutch. Their Clutch score at the plate in May is -1.05. That’s 23rd in baseball, which isn’t great but could be much worse. The bigger problem is that the pitching staff’s Clutch score, -3.87 through Wednesday, is the worst in baseball in May. That doesn’t even do it justice. Here are the ten worst individual months for team pitching staffs since 1989.

Performing historically awful in the game’s highest leverage moments is an efficient way to lose close games in bunches, and undershoot your pythagorean record by proxy. Even more maddening, their May FIP-minus is better than league average. It’s the third best FIP-minus on that list. They’re pitching well enough... except at the worst possible moments.

Conclusions

The good news is that most of this voodoo isn’t sustainable. They’ll win a one-run game at some point. Maybe two! The clutchiness of the pitching staff should even out, and it’s increasingly likely that said pitching staff will not include Michael Wacha. If the lineup can hit more like it did for most of May and less like it did Tuesday and Wednesday, the production will return. That’s the good news.

The bad news, as if you need it in this hellish month of baseball, is that the ghastly month has already done the damage. Their playoff odds have collapsed to 28.6% after cresting on May 1st at 67.8%. Playing better over the summer isn’t a magic wand that converts May losses into wins. At this point, we’re all reduced to hope that whatever correction is coming will happen soon enough to rescue their playoff hopes.