Author Archive

Adam Wainwright Might Have Turned a Corner

For the last year and a half, Adam Wainwright has been singing the same tune after bad starts.

“My arm feels great. My body feels great. I know what adjustments I need to make. I’ll be back.” Cardinals fans have heard those lines from Uncle Charlie since his struggles began. For all of 2016, and most of this season, the idea of Wainwright returning to pre-Achilles tear form seemed preposterous.

There have been games in which Wainwright looked like he should hang it up, like June 6 against Cincinnati (otherwise known as the Scooter Gennett game). At other times, he looked a lot like the Adam Wainwright of 2012-2014, like May 27 at Colorado. That day, he went seven shutout innings at Coors Field, and only gave up three hits.

Wainwright’s ERA is 5.20 going into Monday’s start in New York. But, if you take out the 24 runs allowed in 6 1/3 innings against Miami, Cincinnati, and Baltimore, his ERA would be 3.14. That would be top-10 in the NL, as Jose de Jesus Ortiz noted in the Post-Dispatch.

Why the wild discrepancy? I looked at each start Wainwright has made this season, and divided them into two groups: quality starts and non-quality starts.

Usage Rates

The first thing I looked at was how often he throws each pitch, broken up by quality starts and non-quality starts.

There’s not much to see here. The only significant change is that Wainwright throws more four-seam fastballs in quality starts, but that’s offset by an increase in sinkers in non-quality starts. Either way, the variance isn’t enough to account for such a massive discrepancy in outcome.

Velocity

If Wainwright isn’t mixing his pitches differently, maybe he just throws them harder (or slower) on certain days. Thanks to Brooks Baseball, took the average velocity of each of his pitches in every start. Then, I calculated the quality start average velocity and the non-quality start average velocity.

Again, not what I expected. Since Wainwright is a pitcher presumably in decline due to age, I didn’t expect to see him throwing harder in his bad outings. Wainwright has only thrown his four-seamer harder in quality starts than non-quality starts, and the difference was only 0.5 miles per hour. He’s thrown every other pitch harder in non-quality starts.

At this point, after many calculations, I was beginning to get discouraged.

Changing Speeds

On Brooks Baseball, if you click on a pitchers game log, it will show usage rates, strike percentages, average velocity, and max velocity. I didn’t intend to track max velocity, but I noticed something as I went along: it seemed like the difference between Wainwright’s average velocity and max velocity was greater in quality starts.

I know that’s a lot of numbers, but bear with me. The key columns are the two right-most. In quality starts, Wainwright has more velocity variance in every pitch except the four-seamer (I excluded the change from this analysis because he doesn’t throw it often enough).

I especially want to focus on the cutter and the curve, since up to June 22 opponents were hitting .286 against the curve and slugging .512 against the cutter.

In Wainwright’s last start against the Mets, his average cutter was 82.8 miles per hour. He also ran it up to 88.5 miles per hour. On that afternoon, hitters had to deal a pitch that moves a fair amount, but could also come at them at any speed within an eight to ten mile per hour range (if the average is 82.8, there had to have been some slower than that).¬†In that same start, he threw his curve between 71.9 miles per hour and 76.5.

Doubling Down

In his last four starts, it appears Wainwright has doubled down on changing speeds within the same pitch.

I looks like Wainwright has made an adjustment. It’s not a surprising one, as Wainwright is the type of pitcher that would alter the tempo of his delivery in order to disrupt the timing of the hitters. The league might adjust to him. However, if this is sustainable, Adam Wainwright might have found his way to continue pitching at a high level for several more years.

This article first appeared in The Redbird Daily.


The Cardinals Might Have Lost Three Wins on the Bases

The Cardinals’ have struggled to run the bases for the better part of two years now. So far, the only tangible effect has been third-base coach Chris Maloney’s “reassignment” to the minor leagues. Nevertheless,¬†Cardinals manager Mike Matheny has continued to preach aggressiveness on the basepaths.

I intend to show the effect the Cardinals’ outs on the bases have had on their ability to score runs. A run-expectancy matrix can help. A run-expectancy matrix shows you the number of runs, on average, a team can expect to score from a given on-base state to the end of the inning. For example, with the bases loaded and no outs, a team can expect to score about 2.2 runs by the end of the inning. On the other hand, with nobody on and two outs, the offensive team’s run expectancy is about 0.098 runs. Here’s the basic run-expectancy matrix:


To estimate the number of runs the Cardinals have left on the bases, I charted every out on the bases thus far in 2017 (53). In each of those 53 instances, I charted the actual outcome and the outcome had the mistake not been made. Then, I subtracted the run-expectancy of the actual outcome from the mistake-free one.

In total, the Cardinal’s actual run expectancy is about 22 runs lower than it would be without baserunning mistakes. If you add those 22 runs to the Pythagorean record formula, the Cardinals should be 38-37, or 1.5 games behind the Brewers.

Not all outs on the bases are created equal, though.

All those formulas are useful, but they make a few key assumptions. First, they assume average speed on the bases. Second, they assume an average hitter at the plate. The creators of run-expectancy arrived at the above numbers by studying the results of MLB games over a six-year period. That’s thousands of innings and at-bats for the numbers to even out. But, when you look at just 53 instances, it’s possible for there to be some small-sample-size error. So let’s look at a couple of specific plays from this season.

April 18

With the Cardinals leading the Pirates 1-0 in the 5th, Greg Garcia came to bat with Jose Martinez on first. With nobody out the run expectancy was 0.8.

Garcia lined a double into center. Martinez rounded third and scored easily, but Garcia was thrown out trying for third. Now, it’s possible a throw from the outfield was cut off by the first baseman and redirected to third to nab Garcia. However, quick review of the video shows that not to be the case.

With one run in, the Cardinals could have expected about 1.1 more runs had Garcia stayed put at second. Instead, with nobody on and one out, their run expectancy dropped to .59. There’s about 1/2 of one of those 22 runs.

Luckily the Cardinals hung on for a one-run win.

May 13

Leading the Cubs 3-1, Magneuris Sierra was on first with one out and the pitcher, Carlos Martinez, at bat. Sierra tried to steal second (Lester was on the mound) but was thrown out for the second out.

Run expectancy says the Cardinals went from scoring about .5 a run on average to .2. But the pitcher was hitting. Assuming Carlos would have bunted him over, the run expectancy would have risen to .319. Lower than it was, but higher than if Carlos would have, say, struck out.

This is an example of a time where run-expectancy breaks down. In the National League, pitchers hitting has a tendency to ruin even the best laid plans. And because most formulas make the basic assumptions mentioned above, it’s hard to criticize Sierra’s mistake.

May 18

I bet you’re surprised I got this far without mentioning Matt Carpenter.

Well, on May 18 Carpenter committed one of the stupidest, irresponsible, boneheaded, bordering-gross-criminal-negligence baserunning mistakes I’ve ever seen.

Carlos had pitched an utter gem, and the game was 0-0 in the 9th. Carpenter lashed a sure double into left. It appeared the Cardinals were well on their way to a win, as their run expectancy rose from about half a run to 1.1.

Then Carpenter rounded second, and headed for third.

He was nailed at third easily. The Cardinals run expectancy dropped all the way down to 0.25. They didn’t score in the inning, and went on to lose the game.

As you can see, run expectancy isn’t the perfect tool for evaluating baserunning. Sometimes calculated risks have to be taken based on the speed of the runner or quality of the hitter, two things run expectancy ignores.

Taking the extra base is always a calculated risk. By ignoring the times the Cardinals have been successful, I was setting them up for failure in this scenario. But when you are among the league leaders in outs on the bases, the particulars of those outs require some serious consideration.

The conclusion is this: the Cardinals reckless baserunning has cost them as many as three wins thus far this season.

This article first appeared in The Redbird Daily.