Quantifying the Impact of COVID on the Cardinals

COVID impacted the 2020 Cardinals season. This is known. Everyone associated with the Cardinals felt the impact of it – from the owner and his wallet to the players and their bats and the beer vendors and their calls.

COVID is not done impacting the Cardinals. The Cardinals not only lost significant revenue from not having fans in the stands but they can’t yet project full stands for 2021. Team President John Mozeliak has made it clear that the club is entering its offseason planning with significant payroll uncertainty.

(Addendum: The club not picking up Wong’s extension is the first brutal proof of that. I’ll have analytics on this move Saturday morning.)

COVID is impacting the club’s ability to plan accurately. The Cardinals, like most teams, rely heavily on analytics to determine team production and roster performance. COVID took all those spreadsheets and projections and burned them to ashes.

The Cardinals are entering the offseason with significant and somewhat unanswerable questions.

How meaningful are the performances from 2020? Sample size alone – just 58 games – is enough to introduce significant variance in production. How much does the training environment – an aborted Spring Training followed by an accelerated Summer Camp – matter in evaluations? Then there is the playing environment. The in-season travel schedule was ridiculous. The club played multiple double-headers per week. Home games were moved out of Busch. Off days were limited. The team languished in quarantine in a haunted hotel instead of working out for two weeks.

That’s not the worst of it. The club does not only have to evaluate players who dealt with the virus’s peripheral disturbances. They had ten players who contracted COVID during the season. Carlos Martinez Instagramed from a hospital bed. At least one other staff member and maybe a few other players (the history here is unclear because of medical privacy) made trips to the ER.

While there were reports that some players experienced only “minor symptoms” or were “asymptomatic”, the impact of the virus on play can not be written off. Persistent fatigue, mental cloudiness, lack of energy, and even ongoing heart damage are just a few of the symptoms that can persist after a once COVID+ person tests negative, regardless of the severity of their symptoms upon contraction.

How can the team evaluate the performance of their roster considering the variety of extreme and unique circumstances presented by the season?

This article will not be able to sufficiently answer that question. It will try, though, to quantify the general impact of COVID on the roster and provide loose guidelines that the Cardinals (and by extension, their fans) should consider in making short- and long-term personnel decisions.

The following charts represent the difference between actual performance and projected performance (or 2019 performance in the case of barrel %) for players who fit into four categories: Cardinal offensive regulars (top 9), Cardinal pitching regulars (top 9), COVID positive hitters, and COVID positive pitchers.

I limited the selection of statistical categories considered to those that I believed would be the most impacted by the persistent fatigue caused by both COVID and the wearyingly extreme playing environment for the healthy players. For hitters, I selected traditional and Statcast power stats – SLUG% and Barrel% – and coupled them with traditional & Statcast general production stats – OPS and wOBA. Pitchers were a bit more straight forward – K/9, BB/9, ERA, and FIP.

Peruse the numbers if you choose, but the charts are designed as a visual. The “difference” category between projected and actual performance are color-coded. Shades of red demonstrate that the player’s actual performance was below projected performance. The darker the shade of red, the more the player underperformed. Shades of blue demonstrate that a player’s performance was above projected. The darker the blue, the more the player overperformed.

Let me repeat that for those who are used to Baseball Savant’s system. In this case, red is bad. Blue is good.

COVID+ POSITION PLAYERS

(Note: Austin Dean’s 7 PA’s didn’t warrant his inclusion). That’s a lot of red ink. Dark red ink. Simply put, the players who tested positive for COVID during the season returned to action as a shell of their former selves. Every player experienced decline in the categories chosen. Their under-performance was not equally distributed. Both Ravelo and Thomas experienced >30% declines across the board. Declines for DeJong and Molina were more moderate.

This is significant when talent is considered. Thomas and Ravelo entered the season as role players and were projected to be just slightly better than replacement level. After contracting COVID, both players became essentially unplayable and neither lasted long on the major league roster.

DeJong and Molina, on the other hand, were well-established bats with room to fall. Like Thomas and Ravelo – both power hitters – DeJong saw his slugging percentage drop significantly. Molina didn’t have much power to begin with, but his ability to generate enough exit velocity to barrel balls essentially disappeared.

From an evaluation standpoint, the players who contracted COVID-19 witnessed a dramatic decline in power that impacted their overall production. Talented players were able to partially overcome this, but fringe players were not.

NON-COVID POSITION PLAYERS

Here, the production is much closer to expected. Those who contracted the virus have one variable that asserted control over the others – COVID sapped their production and power. For those who did not contract the virus, the impact of weariness and the grind of an extreme schedule was more subtle.

Still, it is measurable. There is significantly more blue mixed into this chart, but a predominance of red ink remains. Because of the extreme environment created by COVID, offensive players were simply more likely to experience a general decline compared to projections.

This should throw a little cold water on the “blame Jeff Albert” narrative that’s become common since the club’s hitting coach was retained for 2021. It’s unlikely that over such a small sample size, with such an accelerated training environment, that the hitting coach was the primary influencer on the team’s overall offensive decline. That doesn’t mean that Albert did his job well. It just means that he isn’t the obvious source of blame.

Instead, it is logical to conclude that the extreme playing environment contributed to a general but small offensive decline from players, with significant variance in those numbers due to small sample sizes. Simply playing a normal schedule with a normal training environment could lead to offensive production that is closer to projections in 2021.

COVID+ PITCHERS

As with the COVID+ hitters, sample size and timing are a real issue here. The only blue ink to be found is from Kodi Whitley and he earned almost all of that before COVID struck. The rest of the COVID+ arms are just as ugly as their hitting counterparts. Like Ravelo and Thomas, Junior Fernandez – a projected near-replacement level reliever – simply could not provide enough production to justify a consistent presence on the roster.

Martinez and Helsley – players with significantly higher talent levels – also experienced real trouble. Martinez, in particular, is a noteworthy case. He has faced criticism from a loud minority of fans over his terrible performance in 2020. There are rumors from valid sources that the Cardinals are listening to trades for the enigmatic righty. With the data above, how much of Martinez’s struggles this season can we attribute to a COVID diagnosis that sent him to the hospital? The answer has to be almost all of it! He had one aborted started just days before testing positive – he might have been in the early stages of COVID at the time. The club tried to bring him back, but he could just never get going. The team shut him down to avoid any further injury. Helsley’s season followed a similar vein, though his exposure was lessened. Almost across the board for COVID+ pitchers, K’s were way down, BB’s were way up. That causes a massive jump in FIP. Viral fatigue is as good an explanation for this as any!

NON-COVID PITCHERS

What a difference! Blue ink everywhere! Part of this is a product of method. I limited the sample size to the top 9 producing pitchers (the same as the offense). We’re seeing the best of the arms that the Cardinals regularly used.

Why did these pitchers succeed when nearly everyone else on the roster struggled? This is partly because the playing environment for the top pitchers did not change as much as it did for the offensive players. Regardless of the number of games and their frequency, Shildt and Mike Maddux kept their pitchers on tightly controlled pitch and inning counts. The availability of a significant number of AAAA-MLB caliber arms on or near the 40-man roster allowed Mozeliak to dig deep into his stable of pitchers. With so much 40-man space devoted to a revolving door of pitchers, the MLB offense just had to keep playing without a deep well of support.

The Cardinals also benefited from their starters-turned-relievers, who were asked to throw multiple innings with a few days of rest in between appearances. Shildt and Maddux did an excellent job of treating his pitching staff like it was Spring Training – using essentially “scheduled outings”. Certain pitchers were set to go each day, almost regardless of the circumstances of the game, and their workload was carefully managed. This frustrated game-to-game armchair managers, but the strategy seemed to work.

(Maybe this does say something about Jeff Albert? The Cardinals pitching coaches were able to develop a strategy to take advantage of the strengths of their staff while minimizing the exposure of their arms. In a crisis, that’s great coaching. Could the offense have done something similar?)

CONCLUSIONS

COVID dramatically impacted the Cardinals season. Those who contracted the disease became little more than shells of their former self. Assuming health and normal circumstances, it is reasonable to expect that all of these players will bounce back to their expected levels of production in 2021 if given the chance. 2020 stats for several players – including Carlos Martinez, Paul DeJong, Ryan Helsley, and Lane Thomas – should pretty much be thrown out as non-predictive.

For the rest of the roster, especially the hitters, grace is recommended. That’s not normally applicable in the cold-hearted world of statistical analytics. In this case, though, the stats demonstrate that Cardinals hitters almost universally suffered because of the unique circumstances presented by COVID. When those circumstances are eliminated, normal production should return.

When it comes to projecting stats and making roster decisions for 2021, my advice is to allow 2020 to have only minor influence. Other variables – like past production, aging curves, minor league performance, and scoutable skills – should far outweigh the influence of 58 extremely unusual games.