Article written by

21 comments
Matty
Matty

Lots of possibilities for packing more info into the graph (which is already pretty good). Another interesting question is "What do I want to know about strikeout pitches?". In my case, it's mostly: 1. Pitch type 2. Location 3. Speed 4. Swinging, tipped, or called? In no particular order: 1. Indicate pitch location by placing the circle relative to a crosshair representing a down the middle pitch. This would give the reader an appreciation for whether it was the high fastball or a pitch in the dirt. Location matters as much as pitch type--are batters chasing bad pitches or is Timmy just that filthy? 2. Size the 'dots' by velocity? Particularly interesting to view as the game goes along--is Timmy uncorking 96 MPH fastballs in the late innings? This would obviously be tough if you used different shapes for different pitches (as idea I agree with BTW). Maybe use 88 MPH as an average and upsize/downsize accordingly? 3. I think it would be easier to consume the graph if the first inning were on top and subsequent innings were below. I also like the idea of placing the shapes in the 1st, 2nd, 3rd slots.

mhr
mhr

Looks good. I suggest you do more with the horizontal axis. Specifically, represent which outs in the innink were K's. E.g., the strikeout in the second inning was the third out, so the icon should be moved two 'outs' to the right.

cannonball
cannonball

I'd use "K"s and Backward's "K"s myself to denote swinging or called

BEn
BEn

You might want to use the squares/circles thing, with Ks and backward Ks

Jrad
Jrad

Great Chart! Thanks.

oldjacket
oldjacket

I like it! I can't tell Lincecum's change from his slider, though. I wonder how gameday does it.

BruxB
BruxB

The slider should have an extra 3mph or more and slightly more movement.

GoGiants
GoGiants

Is there anything to the fact that these hitters haven't really faced Lincecum? He's had two brilliant games vs AL opponents, and hasn't as faired as well against NL or the second time against the A's. The new slider obviously works well, but is there a trend for hitters catching on, figuring him out? And on the flipside, why no one can hit the new-and-improved Vogelsong?

Liem
Liem

That might be a plausible explanation for Vogelsong, but Timmy has been in the league for a while now, having won 2 Cy-Youngs and clutched up on baseball's biggest stage. Therefore I think yesterday's performance was more indicative of how he's pitched his whole career and that his recent string on bad starts are the anomalies that need explaining. It amazes me how people still want to look at Timmy's bad starts and use outings to build a narrative about Timmy, while at the same time dismissing his much larger body of work that indicates he's one of the best pitchers in our generation. It's as if people are still trying to validate the reasons why they didn't think Timmy would pan out in the first place ("Aha! He had a bad start! I'm right; he's too small, too frail, bound to breakdown!) If you want an explanation as to why Timmy fanned twelve over 7 shut-out innings, the answer is: he freakin' good.

Liem
Liem

I didn't get to see the game on TV so this chart was very helpful for me to see how dominant Timmy got later in the game. From the pitch f/x data, Timmy's change-up look particularly nasty that day. He threw it 12 times for 12 strikes, including 6 whiffs-- a swing-and-miss rate twice as high as his other pitches.

LeviDavis
LeviDavis

Chris Stewart was having a lot of trouble hanging on to those strike threes. Lincecum was filthy, but still, with Whiteside's recent passed ball problems, it's scary.

Lyle
Lyle

Love the chart - very clear and easy to understand. I really thought in the last few innings that it was Tim's "changeup" (splitter) that was getting the strike threes. Who compiles the data that says which pitch is what?

Chris Quick
Chris Quick

Howdy Lyle, the pitch classifications are from PitchF/X which, on occasion, are known to be inaccurate. I briefly went through the game footage and I think the pitch-types look right, but I might review it again later.

Roger
Roger

Love the graph as always, although Obo's addition certainly makes a great deal of sense. Even without the colorblind issue, as an old boss of mine used to say: "data redundancy is NOT a sin." But I really just wanted to ask a question: what was the total of K2-3 putouts yesterday? It seemed abnormally high to me. Possibly as many as 6?

Obo
Obo

I've published a lot of papers in circuits and physics journals and so have learned a lot about how to present data. Color should never be the sole indicator of a piece of information. Try using FB=red circle, Slider=green square, Changeup=yellow triangle, curve=blue X. Data depicted solely by color is a peeve of mine because I am colorblind (as are 10% of your male readers). Also, if someone prints this out in black and white all the information is lost.

Fwotyoz
Fwotyoz

Totally agree - it's too "hard" to hold 4 color meanings in my head. I had to keep referring to the key.

Paapfly
Paapfly

Thanks, Obo! Keep up the great comments.

Chris Quick
Chris Quick

Thanks, Obo. This is exactly the kind of stuff I'm looking for.