It’s mostly forgotten now, on account of the drubbing, but after the Giants scored early against James Shields, the Royals tried real hard to also score early against Madison Bumgarner. By that I mean the Royals made two outs and then had a batter get hit by a pitch, but with two down and Lorenzo Cain on first, Eric Hosmer clobbered a centered fastball toward the left side of Gregor Blanco. Off the bat it was a screaming line drive, and it looked like a ball that might get the Royals back into the game, but Blanco managed to make some kind of sensational play and the inning was over.

As Hosmer batted, the Royals’ win expectancy was about 22%. When the catch was made, that dropped to about 20%. Had the ball found ground for a run-scoring double, the win expectancy would’ve jumped to about 32%. So the catch was worth about 12 percentage points, between made and not made.

What I ask of you, the audience: how difficult of a play do you think it was? Pretend you’re an Inside Edge scout. Categorize the play! I understand this isn’t something you do, so maybe as an audience you’ll suck at this, but I’m interested in seeing how the opinions come in. I guess I should show you the play and give you some additional information.

The catch, from one angle:

The catch, from another angle:

Here’s where Blanco started, and roughly where he wound up:

Here’s a different way of considering that, that doesn’t indicate much about depth but does feature a standing ghost(!):

Off the bat, the ball was in the air for about 3.87 seconds. It was hit something like 390-400 feet. Buster Posey celebrated, understatedly.

So then, what do you think? Difficult play for an average center fielder? Ordinary play for an average center fielder? Extremely amazing play for an average center fielder? Mets fans might end up skewing this, but to balance that out perhaps they’ve mostly stopped paying attention to baseball for the time being.

Last year during October and November, the author published a series of statistical reports on the Arizona Fall League — less because such reports are of great utility to prospect analysis, and more because, for those of us currently not residing in the Greater Phoenix area, one of the few ways to participate in that very compelling league is by dwelling for too long on very small statistical samples.

What follows is an example of such a statistical report for this year’s iteration of the AFL, currently about 12 or 13 games into its season.

Read the rest of this entry »

Before this pitch, you might have heard a lot about James Shields‘ changeup being the best righty change in the business.

But early in the game, both Gregor Blanco and Brandon Belt managed to touch the pitch for singles. They aren’t alone this year.

In terms of whiff rate, James Shields‘ changeup ranked 55th of 105 pitchers that threw 200 changeups this year. Even if you limit the sample to righty starting pitchers, his changeup ranked 31st in that department. Even if you open the sample to the last three years, Shields’ change isn’t top of the line — his changeup has had a 19% whiff rate, which is above average but not elite. Over the same time period, Stephen Strasburg’s right-handed changeup has had a 26% whiff rate, for example.

In terms of shape and movement, maybe it’s not such a surprise that the change is only okay. Harry Pavlidis did work that suggested that, for whiffs at least, the ten mile per hour velocity gap between the fastball and change was important. Shields has a 7.6 mph difference this year. Tilt and fade are also important. Shields’ changeup fades less than his sinker and drops just short of three inches more than his sinker. Sounds good but not great.

Here’s a comparison of Shields’ movement and velocity on the changeup to league average right-handed changeups:

Less vertical movement, more horizontal, and a little bit faster. But that faster quality can cut both ways. Here’s a table that shows the relative difference between the movement and velocity on Shields’ changeup and sinker versus the league average:

The velocity gap between Shield’s changeup and sinker is less than league average. So is the difference in his horizontal movement. That might be surprising to some.

If you look at his pitches this year, his knuckle curve was the most impressive. Among the 114 pitchers that threw 200 curveballs, his was 27th in whiff rate. He had the 14th-best whiff rate among right-handed starting pitchers. Of course, Shields has thrown eight knuckle curves so far, with five balls, one swinging strike, one called strike and a Pablo Sandoval double.

Sometimes it’s just not your night.

The numbers suggest the Royals are slight favorites. Vegas suggests that the Royals are slight favorites. What do FanGraphs readers think, though? Let’s find out.

We’ll run this poll again tomorrow and see how much the outcome of the first game affects our view of what we think is going to happen over the next week.

The great thing about plate discipline statistics is that they’re relatively easy to understand, but the bad news is that there are so many variations and it can be hard to keep them perfectly straight.  Fear not, we’ve expanded out Library entry on plate discipline stats to combat this problem.

The entry includes more specifics about all of the numerators and denominators and provides some more detail about how to interpret the data and the differences between our two sets of plate discipline data.

As always, feel free to ask questions in the comments section here, find me on Twitter @NeilWeinberg44, or stop by our weekly chat designed for this type of inquiry, Wednesdays at 3pm eastern.

From the Catcher’s POV

As both Dave Cameron and Mike Petriello have noted earlier today in different ways, the Royals bullpen has contributed substantially to the club’s successful — and largely unimpeded — run through the 2014 postseason. In their victory over Baltimore on Wednesday, for example, the triumvirate of Wade Davis, Kelvin Herrera, and Greg Holland combined for 3.2 scoreless innings. As noted by Cameron, that same threesome accounted for over 40% of the innings thrown by Royals pitchers during the Championship Series.

Herrera accounted for 1.2 of those innings on Wednesday, and he finished his outing in a notable way, first by means of this 1-1 pitch to Caleb Joseph:

Read the rest of this entry »

Back in March, MLB Advanced Media announced a new project aimed to create the tracking data of our dreams. Instead of just measuring the velocity of a pitch or the exit speed of the ball on contact, this new system — eventually named Statcast — would track the full flight of the ball, the distance covered by fielders, the acceleration and top speed of the baserunners, and just about everything else we could want to know.

And then, for months, there was basically no news. MLB had originally said they were going to treat 2014 as a beta-testing year, giving the 30 teams access to the information to help iron out the wrinkles before the system was fully rolled out in 2015. Only the teams weren’t given any data to beta test. The system was teased in things like this video of Jason Heyward’s catch, but beyond proof-of-concept highlights, Statcast wasn’t that much more real than the mythical FIELDF/x system that had been talked about for years.

Well, that appears to be changing. Last week, MLB opened a @StatCast Twitter account, and after installing the system in all four parks hosting LCS games, they’ve been flooding the market with new examples of the system’s capabilities. We got five videos from just last night’s Orioles-Royals game, for instance, from measuring Lorenzo Cain‘s range in the outfield to Jarrod Dyson’s speed on the bases. They added four more videos from the Giants-Cardinals game, giving us examples of Statcast data from just yesterday.

Additionally, I’ve been told that teams have now received a small sample of the regular season data, and so while MLB might not have been able to beta-test nightly feeds during the season, the system does appear to be closer than ever to going live. And for those who prefer spreadsheets to videos, Daniel Brim has even collected the data from the videos released by MLB, and is updating a Google Doc with the information, which you can find here.

Likely due to the licensing agreements in place with Sportvision and the fact that MLB teams have paid for the rights to access HITF/x data, we haven’t seen the full capability of what Statcast could do on any one play, as MLB has focused mostly on just running and fielding. We don’t know how the league is going to go forward with the F/x data systems, or what is going to be made public and what is not. But given the amount of information that MLB has made available over the last week, it seems more likely than ever that Statcast could be a legitimate public utility, and it’s potential might be realized in the not too distant future.

During yesterday’s edition of FanGraphs Audio, the present author and Dave Cameron began a meandering and ultimately conclusion-less inquiry into the amount of actual, physical baseballs used during the course of a typical nine-inning game, with estimates ranging from around 50 to, according to Cameron, “hundreds.”

Below are three answers, aggregated with little effort from the internet and according to credible sources.

“[A]n average of seven dozen baseballs per game.”

Source: Steve Vucinich, Oakland Athletics Equipment Manager

“[T]he average nine-inning game requires nine dozen baseballs.”

Source: Jake Efta, Progressive Field Umpires Attendant

“[A]n average of eight to 10 dozen baseballs are used each game.”

Source: Tim Burke, Minnesota Twins Assistant Equipment Manager

The conclusion: somewhere between 90 and 120 balls per game, probably, making the average lifespan of a baseball about 2.5 to 3.0 pitches.

Credit to reader Ben Hall, who endeavored to supply actual facts to this conversation.

In Sunday’s NLCS Game 2, the Cardinals had to replace an injured Yadier Molina with a bad Tony Cruz, and while the Cardinals went on to still win the game in dramatic fashion, oblique injuries are a real problem, which means Molina will at least be compromised if he’s even able to play, and that’s awful news for St. Louis. Now, about Molina vs. Cruz — below, please find a mostly pointless comparative exercise, focusing on Trevor Rosenthal’s ninth inning.

Before Kolten Wong was the hero, Rosenthal looked like the goat. Tasked with preserving a one-run lead, Rosenthal gave up a single, then a single, and then he threw a wild pitch on a walk that allowed a runner to score all the way from second base. This is that full-count wild pitch:

There exists a line of thought that, had Yadier Molina still been catching, that pitch wouldn’t have gotten away. By the numbers we’ve got, relative to Cruz, Molina is the far superior pitch-blocker, which shouldn’t come as a surprise since Molina is terrific across the board. I agree that Molina is better than Cruz at this, but to be honest I think this might’ve gotten away from anybody. The pitch was 99 miles per hour, and it bounced, and Rosenthal had never before in the major leagues thrown a fastball so low. This pitch was almost destined for the backstop. Maybe with Molina, the runner doesn’t actually score, but that’s a tough block for anyone.

Let’s rewind to mere moments earlier. You want to compare Molina to Cruz? Here is Joe Panik’s plate appearance:

The closest ball was the third ball, and here’s what that looked like:

Cruz stabbed, and dropped his head, and flinched his right arm. Catching a borderline two-strike fastball, Cruz did the opposite of limit his movement, and you figure that contributed to the pitch being called a ball. Now, when the count is 0-and-2, we know that the called strike zone shrinks. Umpires don’t like to call three-pitch strikeouts. But, Molina is a better receiver than Cruz. We know that generally, and we know that specifically.

This year, on fastballs over the plate and between 1.5 – 2 feet off the ground, Molina got 80% called strikes, and Cruz got 66%. Over the past three years, the difference is 81% vs. 70%. And Cruz, yesterday, showed unusually poor technique, and Molina probably would’ve done a better job.

Baseball Prospectus also allows us to evaluate framing data by battery. Since Rosenthal debuted, Cruz has caught him with 252 framing opportunities, and he’s come in about nine strikes below expected. Molina, meanwhile, has caught him with 1,099 framing opportunities, and he’s come in about 16 strikes above expected. So, Molina receives better than Cruz, and Molina receives Trevor Rosenthal better than Cruz.

Which is all to say, had it been Yadier Molina catching there instead of Tony Cruz, maybe there’s no wild pitch, but even before that, maybe Panik gets called out and Rosenthal gets the save and the game ends. Rosenthal threw a good 0-and-2 pitch. Cruz just didn’t make a good 0-and-2 catch. You see that the next two pitches, also, were close, so they also provided framing opportunities and Cruz couldn’t sell well enough.

Of course, had Molina not been removed for Cruz, everything could’ve happened very differently. The Giants, at least, would’ve had different plate appearances, and Panik wouldn’t have been in the exact same situation. So that’s why this is a mostly pointless exercise.

But I guess the general point is valid: Yadier Molina sure is a hell of a lot better than Tony Cruz in every possible area. You knew that before yesterday. You knew that before you even knew who Tony Cruz was.

The independent Atlantic League’s season ended about a week ago now, the Lancaster (PA) Barnstormers defeating the Sugar Land (TX) Skeeters 3-0 in a best-of-five series. The Atlantic League is notable probably for a number of reasons, not the least of which is that it serves as a perpetual reminder that most ballplayers — who are human beings with self-doubt and expenses — do not ascend seamlessly through affiliated baseball and then onto a major-league roster. Indeed, many of the names in the Atlantic League are familiar ones — and belong to players who’re talented enough to have toiled in something slightly less than obscurity.

With a view to acquainting all of our selves more intimately with this year’s edition of the Atlantic League, what I’ve done is to produce below two leaderboards featuring this year’s top hitters and pitchers — accompanied by notes concerning same. (All data courtesy Pointstreak.)

The clubs in the APBL (and their relevant abbreviations) are as follows: Bridgeport Bluefish (BPT), Camden Riversharks (CMD), Lancaster Barnstormers (LAN), Long Island Ducks (LI), Somerset Patriots (SOM), Southern Maryland Blue Crabs (SMD), Sugar Land Skeeters (SL), and York Revolution (YRK).

Read the rest of this entry »