MLB 12 The Show Pitch Repertoires Corrections Thread

[seanjeezy]

Re: MLB 12 The Show Pitch Repertoires Corrections Thread

I think -7.0 is near the cutoff used in-game between 4-seamer and running fastball but doesn't the running fastball not get up that high in velocity in The Show anyway?

It is possible to hit 100mph with the RFB, but it is very rare and must be up in the zone. For the sake of preserving movement, I would give him a maxed out RFB with 65 movement, he wont hit 101, but he'll sit ~97-99

[seanjeezy]

Re: MLB 12 The Show Pitch Repertoires Corrections Thread

For all those who want to have Chapman as a starter:

check this article out

[Cavicchi]

Re: MLB 12 The Show Pitch Repertoires Corrections Thread

It also says in 2011 the horizontal movement on his fastball jumped up while the vertical movement dropped:

- In 2009 and 2010 his horizontal movement was -3.6 and -3.4 respectively. In 2011 it was -6.9.
- In 2009 and 2010 his vertical movement was 9.1 and 8.3 respectively. In 2011 it was 6.9.

Admitted I have not seen much of Brian Wilson pitching but the numbers suggest that something changed. Either he's throwing his four-seamer differently and maybe from a different arm slot or pitchfx is just failing to differentiate between two different fastballs that he's throwing.

I think it's the latter given the charts in this post: http://www.fangraphs.com/blogs/index...ns-two-seamer/

Just looking at the scatter plot it's clear that he's throwing a four-seamer AND a two-seamer (I might even throw it in running fastball territory given that it's more horizontal than vertical so far as these things go) but pitchfx is merging the two pitches together.

This is exactly why this game needs a "pitch czar" and why blinding looking at pitchfx data can be misleading. The system accurately records movement and velocity but it does not necessarily have a perfect algorithm for distinguishing similar pitches with related movements. People in this thread are missing that, assuming that the pitchfx classifications are accurate, and the devs have failed to acknowledge that blindly going by the numbers can produce woefully inaccurate pitch repertoires.

There's a lot of arguing that can go on over what pitches each pitcher should have and the precise ratings they should receive but I'll say this: Brian Wilson threw two different fastball last year and pitchfx reported it incorrectly.

http://www.fangraphs.com/pitchfx.asp...485&position=P (2 years)

http://pitchfx.texasleaguers.com/pit...o=3%2F9%2F2012 (2years)

And this last one showing same as fangraphs that he didn't throw a 2-seam fastball last year, just 4-seam fastball and slider.

[Cavicchi]

Re: MLB 12 The Show Pitch Repertoires Corrections Thread

Just pitched with Anthony Swarzak of the Twins and he has a 94-97 two seam and an 89-91 four seam in the game. Two seam velocity needs to be turned down. Averaged 92 last year.

http://www.fangraphs.com/statss.aspx...xpitchvelocity

The 2-year average is just under 91 mph, last year it was 91 for 2-seamer and 92 for 4-seamer. I don't think they use just last year when more years are available.

http://www.fangraphs.com/pitchfx.asp...466&position=P

[ComfortablyLomb]

Re: MLB 12 The Show Pitch Repertoires Corrections Thread

http://www.fangraphs.com/pitchfx.asp...485&position=P (2 years)

http://pitchfx.texasleaguers.com/pit...o=3%2F9%2F2012 (2years)

And this last one showing same as fangraphs that he didn't throw a 2-seam fastball last year, just 4-seam fastball and slider.

I'm going to be clear here: you're dead wrong and clearly unfamiliar with pitchfx and these charts. It's all the same data (do you think texasleaguers and fangraphs are starting with different numbers?) but texasleaguers is just regurgitating it, combining seasons, and not analyzing it. You need to look at his velocity charts anyway to understand what I'm talking about. There are two distinct groupings of fastballs in 2011 and that distinction was not present in 2010. Don't look at the one that combines 2010 and 2011 on the texasleaguers site, look at the ones for 2011 and 2010 that fangraphs separates:

2011 V & M: http://www.fangraphs.com/pitchfxg.as...11&date=0&dh=0
2010 V & M: http://www.fangraphs.com/pitchfxg.as...10&date=0&dh=0

If you want to argue this stuff you can't just cite the numbers. You need to know how pitchfx works including its limitations and flaws (specifically, that it just uses an algorithm to distinguish between similar pitch types and that algorithm can make really dumb mistakes or fail to make distinctions when distinctions are present - this happens often between fastballs or between cutters and sliders). Without even watching a single Wilson fastball I can tell you that starting last season he threw two different ones based off those game charts comparing 2011 versus 2010. It's obvious if you have any sort of familiarity with this stuff.

One was a four-seamer with 8-9 vertical movement and -3-4 horizontal movement (in line with his career numbers for the pitch) and the other was a two-seamer with 5-6 vertical movement and -8-9 horizontal movement. It fits with people saying he added a two seamer and that his numbers look like a mess compared to the ones posted in 2009 and 2010... specifically like he was throwing two different pitches and pitchfx failed to distinguish between the two when categorizing.

[seanjeezy]

Re: MLB 12 The Show Pitch Repertoires Corrections Thread

http://www.fangraphs.com/pitchfx.asp...485&position=P (2 years)

http://pitchfx.texasleaguers.com/pit...o=3%2F9%2F2012 (2years)

And this last one showing same as fangraphs that he didn't throw a 2-seam fastball last year, just 4-seam fastball and slider.

I think what he is trying to say is that f/x is incorrectly classifying Wilson's pitches

BWilsonmovementEdit.png

Notice how the part circled is not lumped in with the black splotch; it is also very concentrated, implying that Wilson is throwing two separate fastballs

Not convinced? Check out this chart from Brooks Baseball:

BWilsonfx.png

These are manually assigned classifications by the good people at Brooks Baseball using a variety of techniques, notice the difference in spin deflection and movement

The cutter is subjective, I personally believe its a slider because of the low RPM, but the speed and movement are typical of a cutter, so it must be a hybrid of the two

Edit:
It seems like Brooks assigns all two-seamers as sinkers, the way I distinguish the two is that a sinker has twice as much vertical movement in relation to horizontal movement - in this case the ratio is 1.45:1 so to me the pitch is a two-seamer

[gator3guy]

Re: MLB 12 The Show Pitch Repertoires Corrections Thread

The 2-year average is just under 91 mph, last year it was 91 for 2-seamer and 92 for 4-seamer. I don't think they use just last year when more years are available.

http://www.fangraphs.com/pitchfx.asp...466&position=P

Regardless, his two-seam is too fast. It's 97!!!!!!!!! Although this year he will most likely be in the bullpen.

[dkrause1971]

Re: MLB 12 The Show Pitch Repertoires Corrections Thread

Some of these were mentioned earlier but i included a link.
SP Jake Arreita
Curve from 80 to 77 MPH
Changeup from 78 to 85-86 MPH
http://www.fangraphs.com/statss.aspx...153&position=P

The Change-up one is a big one.

Padres- SP Cory Luebke. Changeup from 79 to 85.
http://www.fangraphs.com/statss.aspx...984&position=P

Jays- Henderon Alvarez FB 90 to 93. Change 79 to 85
http://www.fangraphs.com/statss.aspx...669&position=P

Jason Frasor- remove changeup
http://www.fangraphs.com/statss.aspx...906&position=P

[KHarmo88]

Re: MLB 12 The Show Pitch Repertoires Corrections Thread

Just out of curiosity, are these changes going to be made in the next SCEA update or are these "corrections suggestions" just for us to sift through and pick out ones we like and use them if we like?

[Cavicchi]

Re: MLB 12 The Show Pitch Repertoires Corrections Thread

I'm going to be clear here: you're dead wrong and clearly unfamiliar with pitchfx and these charts. It's all the same data (do you think texasleaguers and fangraphs are starting with different numbers?) but texasleaguers is just regurgitating it, combining seasons, and not analyzing it. You need to look at his velocity charts anyway to understand what I'm talking about. There are two distinct groupings of fastballs in 2011 and that distinction was not present in 2010. Don't look at the one that combines 2010 and 2011 on the texasleaguers site, look at the ones for 2011 and 2010 that fangraphs separates:

2011 V & M: http://www.fangraphs.com/pitchfxg.as...11&date=0&dh=0
2010 V & M: http://www.fangraphs.com/pitchfxg.as...10&date=0&dh=0

If you want to argue this stuff you can't just cite the numbers. You need to know how pitchfx works including its limitations and flaws (specifically, that it just uses an algorithm to distinguish between similar pitch types and that algorithm can make really dumb mistakes or fail to make distinctions when distinctions are present - this happens often between fastballs or between cutters and sliders). Without even watching a single Wilson fastball I can tell you that starting last season he threw two different ones based off those game charts comparing 2011 versus 2010. It's obvious if you have any sort of familiarity with this stuff.

One was a four-seamer with 8-9 vertical movement and -3-4 horizontal movement (in line with his career numbers for the pitch) and the other was a two-seamer with 5-6 vertical movement and -8-9 horizontal movement. It fits with people saying he added a two seamer and that his numbers look like a mess compared to the ones posted in 2009 and 2010... specifically like he was throwing two different pitches and pitchfx failed to distinguish between the two when categorizing.

I don't expect the people responsible for game data to look at charts for all pitchers trying to figure out movement. The reasonable approach to me is looking at what is reported by those 2 links I provided. They are two different sites reporting the same thing.

[ComfortablyLomb]

Re: MLB 12 The Show Pitch Repertoires Corrections Thread

I don't expect the people responsible for game data to look at charts for all pitchers trying to figure out movement. The reasonable approach to me is looking at what is reported by those 2 links I provided. They are two different sites reporting the same thing.

How is it reasonable to ask for help assigning pitches and then just provide a completely incorrect answer due to laziness? Why even come into this thread in the first place? So because a couple internet sites are showing what is essentially inaccurate data due do to failure to properly split his pitches you don't want Brian Wilson in-game to actually reflect Brian Wilson in real life? That's what you're suggesting: the in-game Brian Wilson have no resemblance to real life Brian Wilson. Real life Brian Wilson, a very notable player, threw two distinctly different fastball last year. Your Brian Wilson has just one because you're just not going deep enough into the available data.

The data is out there to do it right, one of the devs created this thread asking us to provide proper information, and what you've provided is misinformation. Now multiple posters have corrected you and provided not just evidence that supercedes yours but also have tried to educate you in multiple posts and with charts on how to look at this data. Instead of admitting you're wrong due to a lack of understanding of pitchfx you're just barging ahead and insisting the wrong data be used because it's easier to comprehend and utilize.

I hope this post complies with the terms of service here because I've tried to be as civil as possible in my last couple posts. Misinformation and lazy and inept statistical analysis are pet peeves so hopefully I've kept things in check. People can come to different conclusions based on the same data at times but sometimes some people just don't know what they're doing.

[Cavicchi]

Re: MLB 12 The Show Pitch Repertoires Corrections Thread

When they give pitchers ratings, are they looking at those charts? How are ratings given? Have you noticed many people here using the site(s) that I posted to give feedback?

[ComfortablyLomb]

Re: MLB 12 The Show Pitch Repertoires Corrections Thread

When they give pitchers ratings, are they looking at those charts? How are ratings given?

My understanding is The Show simply uses a three year weighted average of the raw velocity, horizontal, and vertical numbers. They do not consider the scatter-plots and they do not consider whether pitchers are actually throwing different pitches than what they have thrown in the past. The is no human element which is why every year Russell comes here and asks us to give him a rundown of pitch edits. The devs simply do not check these things.

You get particularly problematic results when pitchfx merges two different pitches and may not have in the past. Not only does a pitcher not get a pitch that they may throw a fair percentage of the time but the numbers for the pitch that previously existed can be skewed. With some pitchers fangraphs goes in and makes an adjustment to distinguish between the two pitches and applies it retroactively. At some point I bet this happens with Brian Wilson.

Have you noticed many people here using the site(s) that I posted to give feedback?

Yes, in a previous post I noted the the misunderstand of pitchfx data that we're seeing in this thread and it doesn't just apply to you. The raw data on the internet is good but it's not necessarily complete. Just because an internet site published it doesn't mean it's necessarily correct either! With some pitchers you can just look at the velocity, horizontal, and vertical numbers and be good to go. With some others you need to check the scatter-plots and make sure that pitchfx isn't failing to distinguish between some distinctly different pitches. Said another way: the raw data is all there but a lot of work is necessary to parse through it and determine what it's actually telling us given what we want to know. Just going by another site's take can be a mistake.

[Phillies_Aces35]

Re: MLB 12 The Show Pitch Repertoires Corrections Thread

Just out of curiosity, are these changes going to be made in the next SCEA update or are these "corrections suggestions" just for us to sift through and pick out ones we like and use them if we like?

a little of both. I remember in 2010, at our suggestion they changed:

(off the top of my head)

Lincecum's 4sfb to a RFB
added a sinker to CC's arsenal
added a cutter to Cole Hamels's arsenal.

[ksig24]

Re: MLB 12 The Show Pitch Repertoires Corrections Thread

I don't think the pitchfx numbers are wrong in terms of separation. An average 12 mph is massive separation off a fastball.

mph separation between fastball and change last year from a bunch of the ERA leaders:
Beckett 4.5
Buehrle 6
Buchholz 11.7
Cain 5.6
Carpenter 6.8
Garza 8.3
Halladay 8.1
Hudson 8.4
Lackey 7
Lee 7
Lester 7.9
Lincecum 8.6
Lohse 9.3
Sabathia 7.3
Santana 9.7
Shields 6.9
Verlander 8.2
Weaver 10
Wolf 9.8

Notice the guys known for having great changes are mostly shy of double digit separation on average? Remember, they're averages, they could hit as high as 93 and as low as 81 but average 91 and 84 quite easily.

I think this is an excellent response, well versed and even better supplied with valid information. I however, am still oposed to certain speeds and its because I am very familiar with facing changeups from professional pitching. I was never much of a pitcher, but know that a pronating motion from throwing the pitch take a substantial amount of speed off the pitch. Where I originally stated that most changeups are between 8-10 mph slower than their average FB, I should have stated that the good changeups are between 8-10 mph. Futhermore, I know that it varies amongst the pitcher and therefore my statement was a bit ambiguous.