Yesterday, I examined the Cardinals baserunning talent using Statcast’s new sprint speed metric. We found that the Cardinals actually have six of ten qualified players with an above average sprint speed. However, we also found that every Cardinals player save Yadier Molina is slower this year than last. For a team whose manager told Derrick Goold that “aggressive is always better,” slower players is probably a bad thing.
Yet, if you go by BsR (the baserunning component of WAR), the Cardinals have actually been exactly average on the bases this season at 0.0 runs. Last year, though, they were the worst in the National League at -19.8 runs. Put the two together, and the St. Louis Cardinals have been the sixth worst baserunning team in the MLB since the start of 2016. They are still the worst baserunning team in the National League by seven runs.
Now that we have sprint speed, we can use this metric to evaluate how the Cardinals have performed on the bases relative to their speed talent. Using 2015 and 2016 data and a 200 PA minimum, I found fairly strong relationships between sprint speed and BsR per 600 PA (r=.6522), UBR/600 PA (r=.5896) and wSB/600 PA (r=.3419). For those who don’t know, UBR (Ultimate Base Running) measures a player’s baserunning value while the ball is in play. wSB (Weighted Stolen Base Runs) measures baserunning value gained or lost through success steal attempts and caught stealing. I use 600 plate appearances to put each player’s value on a full-season rate basis.
Obviously, there’s a lot more going into the equation than the baserunner’s speed, but using these relationships will get us to an expected baserunning value for a player given his speed and assuming league average baserunning opportunities. The difference between the expected value and actual value, under these assumptions, can be attributed to the player’s and coach’s decision making.
First, let’s look at the 2015 and 2016 seasons.
Above, I charted every season with at least 200 plate appearances and sprint speed data. The trend line shows expected performance at a given sprint speed. Since these baserunning metrics rate players as above or below average, the trend line is at 0.0 at the league average sprint speed (between 27.0 and 27.1 ft/s).
The qualified 2015 Cardinals players are highlighted in black, and the 2016 Cardinals players are in blue. In 2015, St. Louis had two players overperform by at least two runs while four underperformed by at least two runs. In 2016, there were again two players who overperformed by at least two runs, but five who underperformed by at least two runs. Not a huge difference, but a tick in the wrong direction.
Notice first that my expected stolen base line flatlines at zero for players with below average speed. Players with below average speed, on average, should not be stealing. If they don’t steal, they don’t add or lose value.
The St. Louis Cardinals have been terrible when it comes to wSB for some time now. Their last above average season was 2004. In 2015 they were bad, with six underperformers and one overperformer. The team’s -3.9 wSB was sixth worst in the MLB and second worst in the NL. It got worse in 2016, when the team ranked last in the MLB with -6.9 wSB. They had nine underperformers. Their stolen base success rate of 57.4% ranked last in the MLB.
UBR measures baserunning value while the ball is in play. It rewards players for taking extra bases successfully and penalizes them for making outs on the basepaths. It might be the best way to evaluate third base coaches, since they are the ones in charge of holding runners at second or third and sending them to third or home while the ball is in play.
In 2015, with Jose Oquendo as their third base coach, the Cardinals +4.2 UBR ranked 10th in the MLB. They had three overperformers and three underperformers. In 2016, with Chris Maloney handling third base coaching duties, the Cardinals ranked last with negative 14.3 UBR. If you believe UBR is the best public metric to evaluate third base coaches, this is a significant indictment against Maloney. More on that shortly.
I want to take a second here to recognize how great Jose Oquendo was as a third base coach. FanGraphs has UBR data beginning in 2002. From 2002 to 2015, the Cardinals ranked first in the MLB with +92.1 UBR. They were 35.2 runs better than the second place Rangers. Jose Oquendo was a great third base coach.
Now, let’s look at the 2017 season.
For 2017, I looked only at players with 100+ plate appearances and who qualified for the Statcast Sprint Speed Leaderboard.
The St. Louis Cardinals have had a weird baserunning season. According to Baseball Reference, they have made the second-most outs on the bases in the MLB and most in the NL. If you include pickoffs, they’ve made the most outs in the entire MLB. Yet, according to FanGraphs, they own an exactly average 0.0 BsR.
According to BsR/600, the Cardinals have four players overperforming their expected level this season and only two underperforming. The black dot you see about nine runs below his expected line is Stephen Piscotty. Today, Piscotty said that talking about the team’s bad baserunning is detrimental to the team. Maybe, but Piscotty’s own terrible baserunning is detrimental to the team. Without Piscotty’s -3.9 BsR to date, the Cardinals would be a top ten baserunning team right now.
The Cardinals have, unsurprisingly, been a below average base stealing team by wSB. Their overall numbers, however, are pulled down by two underperformers: Dexter Fowler and Stephen Piscotty. Fowler has only attempted five steals and has been caught twice. According to Derrick Goold, Fowler is also dealing with a chronic heel injury, so at least he’s realized his limitation and stopped running.
Piscotty, on the other hand, is not injured. Nor has he stopped running. Last night, he stole third successfully. He now has three steals in eight attempts. He’s been caught five times. His wSB is fifth worst in the MLB. Maybe Piscotty shouldn’t be trying to steal bases.
Despite all the outs on the bases, the Cardinals actually have an above average team UBR of 1.5. They have two underperformers and two overperformers. Tommy Pham is the black dot near the top of the graph. Fowler is the one lower and to the right of him. The two underperformers are Kolten Wong and, again, Stephen Piscotty. Overall, though, the picture isn’t too bad if you believe in this metric’s utility.
I mentioned previously that Chris Maloney may have been at fault for the Cardinals horrible UBR last year. The St. Louis front office appeared to agree that he was to blame for the poor baserunning performance when they fired him on June 9th. However, by UBR, the Cardinals have made dramatic improvements since last year. So, if you trust UBR, maybe Maloney didn’t deserve to be fired. If you’d rather just look at the team’s NL-leading outs on the bases, then maybe he did.
Lastly, to tie it all together, I calculated each player’s expected baserunning metrics (xBsR, xwSB, and xUBR) and compared them to their actual statistics to get the difference in value (dBsR, dwSB, and dUBR). The differences show us which players are under- or overperforming their speed talent the most to date. The important numbers are highlighted: blue is good, red is bad.
If you’ve been following along, it shouldn’t surprise you that Stephen Piscotty is a noted underperformer. Fowler has struggled with stolen bases, but has made up for it with his dUBR. Pham has a great dUBR, so his overall line reflects what an aggressive baserunner looks like when nearly all his decisions pay off. Wong has, surprisingly, underperformed on the bases by moderate margin, including a full run underperformance in UBR. Strangely, in the year he puts together a strong offensive season, he’s been a below average fielder and baserunner.
Overall, taking the sum of the qualified dBsR, the Cardinals have actually overperformed their speed talent by 4.0 runs. That doesn’t pass the eye test, so anyone questioning these results has a fair case. I prefer to rely on the metrics, so I’ll opt to give credit where it’s due. The Cardinals baserunning looks ugly, but it’s working.