I give you pitch edits:
https://docs.google.com/spreadsheets...it?usp=sharing
From what I gather, pitch edits has historically been seanjeezy's specialty and I was fully aware that he could pop up anytime with his own version as I was doing this work, but that didn't happen. I am also aware of a few others who have tried to go with the pitch edits from last year basically and tweak here and there especially with super slow changeups. Well, my pitch edits are different than you're used to. Let me elaborate in the FAQ below...
Q: Why pitch edits? Are the differences even really that big compared with the SCEA roster?
A: I fully believe that the pitch edits result in a more authentic experience and it is clear that there are some major differences in repertoires/velocities/and especially control|movement ratings. Whether it's a full 1.5 mph velocity difference on a star pitcher's primary fastball or a breaking ball reclassified as something else, pitch edits make the roster more authentic.
Q: Where did these ratings come from? Where did you pull the data from?
A: Brooks Baseball. Every velocity, movement and command rating is 100% based on publicly available data on Brooks Baseball.
Q: So did you just tweak the 2014 pitch edits or are these new?
A: These are 100% brand spanking new. These data are based on the 2014 and 2015 seasons during the time frame 4/1/2014 - right before Memorial Day 2015. All the data is new.
Q: How is this different than what we're used to seeing from seanjeezy?
A: There is no difference in the velocity numbers, which come straight from Brooks. However, I attempted to not only re-rate movement according to Brooks Baseball data as seanjeezy has done previously, but this is the first time (to my knowledge) the control ratings have been re-rated as well.
Q: How did you do the control ratings? How can we know where the pitches were supposed to go?
A: We actually can't know where the pitches were supposed to go. Even watching the catcher's glove on every pitch doesn't necessarily tell us. But, what I've done is used a stat called "Ball %" which is available for each pitch type of each pitcher's repertoire on Brooks Baseball. After tabulating the Ball % for each of the 771 pitchers' repertoires in the spreadsheet, I sorted by pitch type and made comparisons among like pitches (so 4SM control is only compared to other 4-seamers and not to curveballs). Ultimately, the pitchers with the lower Ball % are getting more called strikes and more swinging strikes on well-placed pitches in and around the zone and those pitchers are thus 'controlling the zone' and got the higher control ratings.
Q: Why Ball %? Why not some other method?
A: I decided on Ball % as a measure of a pitcher's ability to control the zone. This way a pitcher gets credit for a called strike and a swinging strike on a well-placed pitch even if it's out of the zone. It penalizes pitchers who either can't locate the pitch in the strike zone to get a called strike (or get contact) or who can't get the pitch close enough to entice a swing. Normalizing by each pitch type also makes sure control of a curveball isn't being directly compared to control of a fastball (which the league throws much more frequently for strikes).
Q: How did you classify pitches?
A: It depends on the pitch. Anyone who is used to Brooks Baseball knows the peculiarities of that site as far as nearly all "2 seamers" are classified as "sinkers" and the vast majority of changeups are listed as just straight changes. I used that to start from but then re-classified some pitches to be a bit more familiar. Classification of each pitch type is spoilered below.
Q: How did you do movement?
A: Again, it depends on the pitch. Much like for control, the movement on a pitcher's pitch was compared to other pitches of the same type and rated accordingly. Pitches that have no discernible lateral break in the game (i.e. a 12-6 curve) were rated solely according to vertical break. Pitches that have substantial break in 2 directions (i.e. a normal curveball) were rated using both Hmov and Vmov components of their movement.
Q: My favorite pitcher is missing a pitch type...what gives?
A: This is more true for relievers, but SCEA generally has too many pitch types per pitcher. There are lots of 4- and 5-pitch relievers, which just isn't realistic, so I made an effort to reduce repertoires for some guys based on usage %. Generally, if a reliever uses a pitch well below 5% of the time, that pitch was removed to give a more realistic selection of his "go to" pitches. For starting pitchers, it really depended on the pitcher and the pitch but even pitches used as little as 2% (in some cases 1% with decent pitch volume) were retained for SPs. The repertoires are ordered from Pitch 1-5 according to usage % with a few exceptions here and there: if a pitcher throws a slider 43% of the time and a 4SM 41% of the time, I tended to put the 4SM as the primary pitch.
Q: Some of the ratings don't make any sense...why did some star pitchers get downgraded?
A: All I can say here is what we are taught to believe from watching broadcasts/a single highlight pitch and what the pitches actually do over hundreds/thousands of deliveries are not always the same thing.
Another issue is that pitches with higher velocity tend to not have as much time to move, so guys with premium velocity in some cases took an understandable hit on movement.
I have run sims and pitched manually with star pitchers like Clayton Kershaw and although the movement on his pitches was adjusted drastically down, he still feels and accumulates stats as you would expect. Even being classified as simply having a 12-6 curveball, for example, already puts that pitch amongst elite curveballs in the game.
The pitch edits make the pitch trajectory and visual aspect more accurate to real life whereas how pitchers perform is also controlled by the /9 ratings and by user skill/sliders. Agree or disagree with certain ratings, but these pitch edits are all based 100% on real, live Pitchf/x data. Even if you only take velocity and repertoires and some of the control/move ratings from the pitch edits, that's a big step up from the SCEA roster.
Q: You mentioned that some star pitchers got hit reasonably hard on their pitch ratings (movement especially)--so do stars become scrubs after entering these?
A: Nope. For example, Kershaw is still a 99 OVR even after inputting the pitch edits. His sim stats are still scary similar to projected stats, and I felt that I could strike plenty of hitters out in played games as well. While some OVRs go down 1-3 points (and a few more like 4 points), not all pitchers even decrease. David Price gets a nice ratings boost for his yeoman's work at throwing strikes.
Q: So what do you mean when adding a star * next to a player's name?
A: The * either signifies that the data could be a bit older (from 2012 or 2013) due to the pitcher not pitching at all in 2014/2015 or that the data is from Fall/Winter/Spring training instead of a regular season game (this is true for a handful of prospects).
Q: Anything else I should know?
A: The FB-CH velocity discrepancy has been naturally fixed in doing this process.
Any questions, comments, and even criticisms are more than welcome in this thread, but if you are posting about something already answered in the FAQ without reading the FAQ, I will probably ignore the post.
https://docs.google.com/spreadsheets...it?usp=sharing
From what I gather, pitch edits has historically been seanjeezy's specialty and I was fully aware that he could pop up anytime with his own version as I was doing this work, but that didn't happen. I am also aware of a few others who have tried to go with the pitch edits from last year basically and tweak here and there especially with super slow changeups. Well, my pitch edits are different than you're used to. Let me elaborate in the FAQ below...
FAQ
Q: Why pitch edits? Are the differences even really that big compared with the SCEA roster?
A: I fully believe that the pitch edits result in a more authentic experience and it is clear that there are some major differences in repertoires/velocities/and especially control|movement ratings. Whether it's a full 1.5 mph velocity difference on a star pitcher's primary fastball or a breaking ball reclassified as something else, pitch edits make the roster more authentic.
Q: Where did these ratings come from? Where did you pull the data from?
A: Brooks Baseball. Every velocity, movement and command rating is 100% based on publicly available data on Brooks Baseball.
Q: So did you just tweak the 2014 pitch edits or are these new?
A: These are 100% brand spanking new. These data are based on the 2014 and 2015 seasons during the time frame 4/1/2014 - right before Memorial Day 2015. All the data is new.
Q: How is this different than what we're used to seeing from seanjeezy?
A: There is no difference in the velocity numbers, which come straight from Brooks. However, I attempted to not only re-rate movement according to Brooks Baseball data as seanjeezy has done previously, but this is the first time (to my knowledge) the control ratings have been re-rated as well.
Q: How did you do the control ratings? How can we know where the pitches were supposed to go?
A: We actually can't know where the pitches were supposed to go. Even watching the catcher's glove on every pitch doesn't necessarily tell us. But, what I've done is used a stat called "Ball %" which is available for each pitch type of each pitcher's repertoire on Brooks Baseball. After tabulating the Ball % for each of the 771 pitchers' repertoires in the spreadsheet, I sorted by pitch type and made comparisons among like pitches (so 4SM control is only compared to other 4-seamers and not to curveballs). Ultimately, the pitchers with the lower Ball % are getting more called strikes and more swinging strikes on well-placed pitches in and around the zone and those pitchers are thus 'controlling the zone' and got the higher control ratings.
Q: Why Ball %? Why not some other method?
A: I decided on Ball % as a measure of a pitcher's ability to control the zone. This way a pitcher gets credit for a called strike and a swinging strike on a well-placed pitch even if it's out of the zone. It penalizes pitchers who either can't locate the pitch in the strike zone to get a called strike (or get contact) or who can't get the pitch close enough to entice a swing. Normalizing by each pitch type also makes sure control of a curveball isn't being directly compared to control of a fastball (which the league throws much more frequently for strikes).
Q: How did you classify pitches?
A: It depends on the pitch. Anyone who is used to Brooks Baseball knows the peculiarities of that site as far as nearly all "2 seamers" are classified as "sinkers" and the vast majority of changeups are listed as just straight changes. I used that to start from but then re-classified some pitches to be a bit more familiar. Classification of each pitch type is spoilered below.
Spoiler
Q: How did you do movement?
A: Again, it depends on the pitch. Much like for control, the movement on a pitcher's pitch was compared to other pitches of the same type and rated accordingly. Pitches that have no discernible lateral break in the game (i.e. a 12-6 curve) were rated solely according to vertical break. Pitches that have substantial break in 2 directions (i.e. a normal curveball) were rated using both Hmov and Vmov components of their movement.
Q: My favorite pitcher is missing a pitch type...what gives?
A: This is more true for relievers, but SCEA generally has too many pitch types per pitcher. There are lots of 4- and 5-pitch relievers, which just isn't realistic, so I made an effort to reduce repertoires for some guys based on usage %. Generally, if a reliever uses a pitch well below 5% of the time, that pitch was removed to give a more realistic selection of his "go to" pitches. For starting pitchers, it really depended on the pitcher and the pitch but even pitches used as little as 2% (in some cases 1% with decent pitch volume) were retained for SPs. The repertoires are ordered from Pitch 1-5 according to usage % with a few exceptions here and there: if a pitcher throws a slider 43% of the time and a 4SM 41% of the time, I tended to put the 4SM as the primary pitch.
Q: Some of the ratings don't make any sense...why did some star pitchers get downgraded?
A: All I can say here is what we are taught to believe from watching broadcasts/a single highlight pitch and what the pitches actually do over hundreds/thousands of deliveries are not always the same thing.
Another issue is that pitches with higher velocity tend to not have as much time to move, so guys with premium velocity in some cases took an understandable hit on movement.
I have run sims and pitched manually with star pitchers like Clayton Kershaw and although the movement on his pitches was adjusted drastically down, he still feels and accumulates stats as you would expect. Even being classified as simply having a 12-6 curveball, for example, already puts that pitch amongst elite curveballs in the game.
The pitch edits make the pitch trajectory and visual aspect more accurate to real life whereas how pitchers perform is also controlled by the /9 ratings and by user skill/sliders. Agree or disagree with certain ratings, but these pitch edits are all based 100% on real, live Pitchf/x data. Even if you only take velocity and repertoires and some of the control/move ratings from the pitch edits, that's a big step up from the SCEA roster.
Q: You mentioned that some star pitchers got hit reasonably hard on their pitch ratings (movement especially)--so do stars become scrubs after entering these?
A: Nope. For example, Kershaw is still a 99 OVR even after inputting the pitch edits. His sim stats are still scary similar to projected stats, and I felt that I could strike plenty of hitters out in played games as well. While some OVRs go down 1-3 points (and a few more like 4 points), not all pitchers even decrease. David Price gets a nice ratings boost for his yeoman's work at throwing strikes.
Q: So what do you mean when adding a star * next to a player's name?
A: The * either signifies that the data could be a bit older (from 2012 or 2013) due to the pitcher not pitching at all in 2014/2015 or that the data is from Fall/Winter/Spring training instead of a regular season game (this is true for a handful of prospects).
Q: Anything else I should know?
A: The FB-CH velocity discrepancy has been naturally fixed in doing this process.
Any questions, comments, and even criticisms are more than welcome in this thread, but if you are posting about something already answered in the FAQ without reading the FAQ, I will probably ignore the post.
Comment