Appreciating the 2015 Cardinal Bullpen

Recently I’ve been sifting through a lot of numbers related to bullpens, and last night I discovered something that really caught my attention. The Cardinals bullpen had an absolutely remarkable season in spite of their almost insane manager.

If you take a quick look at the numbers like many of the major television networks – @ESPN, MLB Network etc. – the Cardinals didn’t field the best bullpen in baseball in 2015. The Pirates and Royals both sported lower ERA’s at 2.67 and 2.73 respectively. St. Louis tied for 10th best in baseball with 20 bullpen losses and ranked 15th in both BAA and strikeouts at .242 and 466.

Maness + Cruz.jpg
Photo by Jeff Curry for USA Today Sports

They don’t have the lowest ERA, they don’t have the stockpile of power arms that can strike out everyone in sight, and to the naked eye they look like a middling bullpen that was placed in favorable situations by a historically good starting staff. But oh my lord would you be wrong to assume that last statement was correct.

Was the Cardinals starting staff historically good in 2015? Absolutely, and not just when you consider what they were working with. But that doesn’t mean the Cardinal bullpen was constantly inheriting monster leads and being tasked with simply not screwing up too bad. In fact, the case was almost the exact opposite.

Recently I’ve started studying LI and WPA. Leverage Index and Win Probability Added both study the idea of how every play impacts the overall game as a whole. Leverage Index applies one single number to every given situation to determine how important it is in the overall context of a game. 1 is the average LI of every situation; anything below 1 is less important, and anything over 1 is more important. ‘High leverage’ situations are any LI of 1.5 or over, while ‘low leverage’ situations are any LI of 0.7 or under. ‘Medium leverage’ situations range in between those two numbers, obviously with a situation of 1.49 LI being more important than a situation of 0.71 LI.

This is where the brilliance of the Cardinal bullpen starts to shine through. As I said, 1 is the average Leverage Index in any given situation and the Cardinal bullpen’s LI when entering a game was 1.101, which lead the league over second place Pittsburgh, which finished at 1.069 average LI.

Siegrist 1
Photo by Dilip Vishwanet for Getty Images

Basically, any time a Cardinal reliever entered the game the situation was of above average importance. It’s impossible to quantify with a number, but I can guarantee you that having to pitch knowing any one mistake is going to be magnified and could potentially lose a game is much harder than having to pitch knowing your mistakes probably won’t matter in the overall scheme of the game.

And the Cardinal relievers still thrived.

Bear with me for a second here, I need to explain more statistics. ‘Clutch’ is a stat that comes up with one number to attempt to encapsulate how clutch a player is; pretty simple, right? It takes a player’s WPA – explained above – and divides it by that same player’s pLI – the average leverage index when a player is batting or pitching.

In the words of Fangraphs’ David Appleman, ‘clutch’ defines “…how much better or worse a player does in high leverage situations than he would have done in a context neutral environment.” It also compares a player against himself, so a player who hits .300 in high leverage situations when he’s an overall .300 hitter is not considered clutch.”

Also, shutdowns and meltdowns are a stat designed to replace saves and blown saves. It utilizes WPA to measure how well a pitcher performed. Any outing of +0.6 WPA or more is considered a shutdown, and any outing of -0.6 WPA or less is considered a meltdown. Again, pretty simple. Moving along…

Among their top 5 relievers – which I identified via IP as Kevin Siegrist, Trever Rosenthal, Miguel Socolovich, Carlos Villanueva and almighty Seth Maness – the Cardinals thrived in pressure situations. Between the 5 of them, the average ‘clutch’ factor was 0.776. A clutch factor of 0.6 is considered great, and the Cardinals had 5 guys averaging more than that. Leading the charge in the clutch department is Mr. Double Play himself, Seth Maness, with a mind blowing 2.54 clutch factor.

Villanueva 1
Photo courtesy of Getty Images

And if you’re a person who relies upon the eye test to tell you how good relievers are, answer me this. How many times did you see Seth Maness enter a tight ball game with men on base and wiggle his way out of that jam with no harm done?

These same five Cardinal relievers combined for 117 shutdowns and only 29 meltdowns. Comparatively, the top 5 relievers determined by IP from the immortal Royals’ bullpen – Greg Holland, Wade Davis, Kelvin Herrera, Franklin Morales, and Ryan Madson – combined for 115 shutdowns and 38 meltdowns.

The Royals’ average LI was 1.003 and they only faced 172 high leverage situations (LI of over 1.5) to the Cardinals 189 high leverage situations. Overall, the Cardinals’ bullpen led all of baseball with an average of 3.24 runs allowed per game. The Royals sit all the way down in 9th on that list, allowing 3.96 runs per game.

Huh…so according to these numbers the Royals’ bullpen allows more runs per game in lower leverage situations than the Cardinals’ bullpen does in higher leverage situations. I don’t really want to get into mindsets or anything like that, but all I’m trying to do here is just show how freaking good the Cardinals’ bullpen was in 2015, and the credit that is not given them.

MLB: Spring Training-Miami Marlins at St. Louis Cardinals
Photo courtesy of Getty Images

Lastly, in my opening paragraph I mentioned that the St. Louis bullpen had a brilliant 2015 in spite of a near madman manager and you’ve waited long enough; it’s time to touch on that teaser. The number that got my attention most was something that all Cardinal fans inherently knew, but can now point to as numeric proof that Mike Matheny overworks his bullpen.

The Cardinals had 146 relief appearances where the pitcher was entering on zero days’ rest, a number that no other team came within 10 of. Kevin Siegrist led the league in relief appearances with 81, and all the wear and tear showed through down the stretch as he gave up monster home runs to Anthony Rizzo – on more than one occasion – and Kyle Schwarber to essentially lose St. Louis the NLDS against Chicago.

And there is absolutely no way that both Kansas City and the New York Mets being below the league average of 111 zero days’ rest appearances is a coincidence. Both Ned Yost and Terry Collins kept very good bullpens fresh throughout the regular season whenever they could, and it paid off in October as their fire departments were stronger than all the others and carried both teams to the Fall Classic.

You can look at that last number one of two ways. You can either hang your hat on the fact that Mike Matheny needs to lay off the gas pedal with his top relievers way more often or you can further appreciate the feats that the Cardinal bullpen was able to accomplish whilst being under both enormous stress and somewhat extreme fatigue.

The Cardinal bullpen was a truly remarkable and vastly underappreciated asset in 2015 that only got stronger this offseason.

Missing from that list of top 5 relievers was super set-up man Jordan Walden, who went down in May of 2015 with a biceps injury and never returned. He is set to return at the beginning of spring training and will provide a massive boost should he stay healthy. Combined with Walden is the acquisition of Korean import Seung-hwan Oh, a dominant closer in his native country that will presumably provide set-up help to spell Kevin Siegrist and ‘Matheny-proof’ the bullpen.

A dominant bullpen only got better this offseason, which it actually needed to do, contrary to the primary message delivered through this article. Oh, but the Cardinals are only worth 84 wins as a team in 2016 according to Fangraphs.

We’ll see about that.

Thanks for reading…

-Ryan

All numbers are credit of Fangraphs and Baseball-Reference

 

 

Advertisements
Appreciating the 2015 Cardinal Bullpen