Notice: Trying to get property 'display_name' of non-object in /var/www/html/wp-content/plugins/wordpress-seo/src/generators/schema/article.php on line 52
keyboard_arrow_uptop

Those of you who follow baseball via the computer—and unless your secretary printed this piece out for you I assume that description fits everyone reading—have probably encountered MLB.com’s Gameday feature. It’s the ingenious little program that tracks each pitch in real time and allows you to follow the action without actually seeing the action.

I’m not sure exactly when Gameday was launched, but I remember using it back in 2000. Gameday has made a lot of improvements since then. When it started, Gameday was a very simple application that wasn’t much more than a moving box score. It looked like this:

OK, that’s Pong, but it was a simple application with simple graphics and limited user-friendliness.

Since then, there have been a number of advancements. The player standing at the plate, the field view that uses a little red dot to show where the ball landed on the field, and even player names on the bases were added over time. Here’s what Gameday looked like back in 2003:

That Mark Prior could really pitch, huh? At this point Gameday had consistently updated stats so at any given point you could tell what the players had done during the game and how that tied into the year to date. From there things got more cool-looking, with advanced graphics, and a virtual field that floated in some nebulous universe free of buildings, stands, people, sky, or presumably, constitutional law. Here’s Gameday in 2008:


Joe Borowski struck out Jorge Posada so all is right with the world. Gameday has since added images of the field where each game is being played, and updated the graphics, but mostly the thing remains the same. It’s a quick and easy way to follow your team online without watching the game itself.

However there is a special addition this year that potentially points Gameday in a new direction: scout. scout, which is either all lowercase or all capitalized letters but in a lowercase font, is Gameday’s foray into color commentary and opinion. It pops up occasionally during the pitch-by-pitch section of Gameday’s play-by-play feature to provide extra information, quick bites to enhance our experience that only the trained eye of a scout could add. Sounds groovy, huh?

I first became aware of the scout early on when it popped up to note the following:

As you can see, scout was concerned about Josh Johnson’s trouble locating his four-seam fastball because, as scout noted, he had missed the strike zone with his last three pitches. This, technically, was true. Though it was a bit like noting during the Battle of Gettysburg that you’re concerned the firing on Fort Sumter could ignite a Civil War. The fact that scout chimed in only after a four-pitch walk was somewhere between gratuitous and weird.

After noting a few more of these oddities, I decided to investigate. So, the crack investigation team here at Out in Left Field set to work. First, like every truly scientific investigational team, we grabbed beer. Then we hit the series of tubes to find out and report the truth!

Here is the result of that beer (and the labor that surrounded it): a review of  scout’s performance Saturday, during the Orioles 8-2 beatdown of the Red Sox.

First inning
Play gets underway in Boston and scout remains quiet for two batters. He’s so quiet that if you didn’t know better you wouldn’t even think a scout was there! But, as you shall soon see, he is. After Red Sox starter Aaron Cook induces two groundouts, he gets ahead of Nick Markakis 0-1. He then throws three sinkers in a row out of the zone. scout jumps up helpfully to point out that, hey, Cook has thrown three sinkers in a row out of the zone.

We now have the first scout skill: ability to highlight the obvious.

In the bottom of the first with two outs and men on second and third, Orioles pitcher Jason Hammel gets Mike Aviles to ground out. scout sees his (her?) opportunity and jumps in:

 

Unlike the first attempt this at least adds information. But what kind of information it adds is open to interpretation. When scout says “Aviles is hitting .000 against [two-seam fastballs] in that location” some fans might think, “Wow, Hammel really knew what he was doing” while others might think “In how many plate appearances and in how many situations has he faced a truly similar pitch and where exactly is ‘that location*’ because without that information this is just drivel.

* When following the game live, Gameday shows you the pitch location, but since I’m going back to look at it after the fact, I can’t tell exactly where Gameday put the pitch in question.

There could be some utility there, but it’s hard to figure out how much. In any case, it’s close to the “He’s hitting .700 in day games followed by night games on days ending in ‘nesday’ during a lunar eclipse with a live bear on his head” style of entirely useless stats.

Second inning
After Cook gets Matt Wieters to ground out, scout jumps in to say that four of Cook’s five outs came on the ground. Well done, scout. That’s useful information and considering Cook’s groundball proclivities it tells us a lot about how his sinker is working that day and the way the Orioles are reacting to it. Sure, that information is available elsewhere on Gameday, but one might not know to look for it and so highlighting it is a good thing.

Then there are the less useful intrusions by scout, such as this next one:

Aaron Cook just spent 33 1/3 innings in Triple A Pawtucket. During that time he struck out 13 of the 133 batters he faced. In his 10-year career, Aaron Cook’s K/9 rate is 3.8. Aaron Cook doesn’t put batters away unless he’s playing a video game version of himself and has altered the settings. Also, if we’re paying even a modicum of attention to the game, we already know that.

In the bottom of the second, scout feels it necessary to add this:

The astute reader will note that Saltalamacchia struck out on the pitch in question, so whatever chance Hammel had taken paid off. Moreover, the pitch was blocked and the catcher had to throw to first to get the out. The catcher is Matt Wieters, at minimum a good defensive catcher, so it can safely be assumed that the pitch wasn’t right over the plate. In fact, it was a slider, so it was probably in the dirt and Saltalamacchia swung over it. All of which begs the question: how is Saltalamacchia hitting .571 on sliders in the dirt?

Third inning
After getting a one-pitch fly out, Aaron Cook faces Endy Chavez. scout gonna scout:

Putting aside the impossibility that Endy Chavez could be hitting .500 against anything ever (a kitten tossing a pom-pom?), maybe throwing a four-seamer to Chavez wasn’t the greatest of ideas, although that’s if you believe Gameday that it was a four-seamer (for the record, Brooks Baseball seems to agree). But it’s not like Cook threw a changeup to Dustin Pedroia and Pedroia crushed it over the Mass Pike.* It was a bunt!

* That would have been a feat considering they’re on the same team.

Maybe because of negative feedback, scout sat quietly for nine batters until one hitter into the bottom of the inning. Jason Hammel’s pitch count reached 45 and scout said this:

I suppose this falls under the category of moderately useful information as long as you don’t think too hard about it. If you do, you’d realize that just about every pitcher is less effective after 45 pitches than before. Pick just about any arbitrary point, say 63 pitches, and most pitchers will be more effective before that point than they will after it. This is for the simple fact that everyone, even major-league pitchers, get tired and the tired part tends to come at the end of their starts. And when you separate the information in that binary way, the second category is often going to include the worst inning a pitcher throws, the inning that got him removed from the game.

But maybe the worst part about this one was that Hammel was facing Nick Punto, who renders all statistics of this nature useless by the shear force of his ineffectiveness. Maybe what scout should have said is that even though Hammel had thrown 45 pitches, Nick Punto was still Nick Punto and thus there was no reason to fear.

Fourth inning
After walking David Ortiz and falling behind on Adrian Gonzalez 2-0, scout let fly with this:

Scout is correct, Gonzalez did double on a two-seam fastball back in the first inning. What’s more, Hammel eventually strikes Gonzalez out on a four-seamer, though that came after bringing the count to 3-1 with two more two-seamers. Still, I’m not sure a pitcher should abandon a pitch because a guy hit it once two innings ago. It probably depends on how good the pitch was and what the scouting reports say on the hitter. Clearly, Hammel thought he could get Gonzalez out with his two-seam fastball and, at least in this at-bat, he did.

Fifth inning
From the “Duh” file scout gives us this:

I’m just not sure that reiterating what just happened adds much value to the Gameday experience.

Now we get to my favorite. Four batters later in the bottom of the inning, Hammel gets Cody Ross to ground out. Scout says this:

So that looks bad. Four of 14 isn’t a particularly good ratio for a ground-ball pitcher, and it certainly doesn’t qualify as getting opposing hitters to hit the ball on the ground. In fact, it’s the opposite. So the comment by scout just looks weird. But, it turns out it isn’t. By this point, Hammel has faced 16 Red Sox hitters. He struck out seven of them, got four groundball outs, a line out, a fly out, a walk, and two hits, one of which was a ground-ball single. Of the eight hitters who have put the ball into play against him, five have hit it on the ground. That’s not bad, it’s good. What’s more, five out of eight is much more impressive than four out of 14. Scout could probably have phrased this one better.

Sixth inning
After going 2-2 to Marlon Byrd, scout lets us know that Hammel isn’t tiring, “with his fastball sitting at 94 over his past 10 pitches.” The wording here is a bit confusing, but overall this is a fine piece of information. That’s all for scout in the sixth.

Seventh through ninth innings
For the most part, the rest of the game consists of scout recycling lines. scout tells us Saltalamacchia is hitting 1.000 against sinkers in “that location” after a Salty single in the seventh, how Scott Atchison has four of his five outs on the ground in the eighth, and how Darren O’Day had Dustin Pedroia down 0-2 but couldn’t put him away after Darren O’Day had Dustin Pedroia down 0-2 but couldn’t put him away.

Maybe scout was off getting a hotdog or comparing notes with other scout, but scout kept quiet in the ninth inning. Or maybe it was that the Orioles were up 8-2 and the person whose job it is to do this silly stuff switched over to a more interesting game.

 * * *

Compared to other Gameday features, like batter images or the ability to call up a player’s statistics, scout is the least lifelike. It’s a departure from Gameday’s normal mission of opinion-less information. As we’ve seen, on some occasions it can add to the game but usually only if you aren’t really paying attention, which I’m sure is true for many Gameday users. But too often it fell back on repeating what had just happened, or its message got jumbled due to awkward wording. The idea, I think, isn’t a bad one and if an actual scout rather than the computer behind the curtain was there to type value-added messages 10 times a game, it could be interesting. In the end, Gameday is probably better served by just giving us the best information as quickly as possible and leaving scouting to the scouts. 

Thank you for reading

This is a free article. If you enjoyed it, consider subscribing to Baseball Prospectus. Subscriptions support ongoing public baseball research and analysis in an increasingly proprietary environment.

Subscribe now
You need to be logged in to comment. Login or Subscribe
rawagman
5/07
So I take it that you, too, are not a fan of "Nasty Factor"?
mattymatty2000
5/07
I think all these additions can make GameDay a little bit better in the minds of some, but I also don't think either adds much. If a feature's success rate (i.e. the number of times it adds useful information divided by the total number of times it attempts to provide useful information) is 50 percent or less, probably just scrap it.
zasxcdfv
5/07
I liked the article. I think that Scout is a waste as of this point, but over time it will improve through use, much like Pitch F/X and the overall Gameday application. At least I hope it does because, as of now, it's weak sauce.
rrvwmr
5/07
Was this comment computer generated by Scout?
MHaywood1025
5/07
Fun article!

Also, I'd like to know which ML batter has at least ten at-bats with a bear on his head.
mattymatty2000
5/07
Thanks.

Judging by his stats, I'll say Casey Kotchman and Marlon Byrd both have bears on their heads. Also probably on their faces preventing them from seeing the baseball.
harrypav
5/07
It will get better. It's a very simple implementation, AFAIK, at the moment, but clearly MLBAM is not afraid to put new things out there and work on them.
It reminds me of the color commentary feature in Diamond Mind, so that's a plus.
mattymatty2000
5/07
It has some potential (I hope I pointed that out in the article), but definitely needs some more work.
JWeinberg
5/07
I think scout is just Joe Buck typing his thoughts
jmanig
5/08
Oh come on! I don't remember the last time I laughed so hard at a column. The present version of SCOUT if it's not useful is at least entertaining. Tell me you didn't have fun researching that piece.

I personally applaud MLBAM's approach here. The key part of this exercise is figuring out what the viewers will consider useful or useless information, so you're betting off releasing early and getting feedback. And as someone pointed out, Pitch F/X was a gigantic mess when it first came out as well.

For the Saltalamacchia AB, I think what SCOUT was trying to say was "it's a good thing Jason Hammel threw that slider in the dirt because if he misses and it ends up catching more plate than he intended that's right in Salty's wheelbase, so he's toast." Sometimes it's just the wording that needs work.
Behemoth
5/08
I kind of assumed it was all computer-generated, as a human would surely come up with something more sensible. Hopefully, if it gets better, you can let us know, so we can switch scout back on.