Pitch type trends

[AaD88]

This post can also be viewed on my league's forums: https://statsplus.net/fightclubbaseb...php?f=13&t=347

We're also always looking for new members on our waitlist! https://forms.gle/FJTVTyLFqT5ERJrQ7

----------------------------------------------------------------------------------------------

I've always wanted to research individual pitches in OOTP and how they affect starters. This article is meant to help identify some trends and maybe give more insight to OOTP GMs.

METHOD
I created five fictional leagues in OOTP20 with 100% ratings and identical settings. I did not put a lot of time into detailing the settings; they are all fictional OOTP standard and may not mirror MLB stats perfectly.

I simmed a full 162 game season in each league and dumped the pitcher stats and ratings on the last day of the season.

Every pitcher with MLB innings and both R/L splits were counted. There were 2671 players in all. I also flipped splits for LHP to indicate same/opposite handed splits.

I calculated essential stats for all pitchers, and then sorted by pitch. For each pitch, pitchers with a 50-rated pitch or better were counted and stats were accumulated.

DISCLAIMERS
Obviously, there are numerous variables in play in OOTP. It is next to impossible to single out pitches and say with certainty how they affect performance. Consider these results with caution and view them as trends.

Despite having over 2600 players over five leagues, four pitches still had incredibly small sample sizes: the Circle Change (47 total players), Knuckle Curve (42), Knuckleball (11), and Screwball (5). Because of the limited sample, I did not include the results for these four pitches in the main body of the results, and instead give a highly speculative analysis at the end based on the results that I have.

As always, a correlation does not equal causation.

RESULTS

Link to data: https://docs.google.com/spreadsheets...it?usp=sharing

ANALYSIS

To analyse the results, I am going to rank the pitches by which I think are most useful to least useful. Everyone plays the game differently, so opinions may vary.

Cutter: The cutter stands out for being one of the most split-neutral pitches in OOTP. It has below average splits in K%, BB%, and BABIP allowed. It is slightly below average in HR%/FB and BABIP allowed, and has very few weaknesses.
Pros: Smallest splits of any pitch.
Cons: Doesn't otherwise stand out. Below average K%.

Changeup: The changeup stands out as being one of the best BB% pitches, a very low split pitch, and a low BABIP pitch. All around, it is one of the best secondary offerings, but seems to be one of the toughest to develop in the game.
Pros: Tied for lowest in BB%, very good splits, low BABIP allowed.
Cons: One of the lower K% pitches, overall above average HR%/FB rate

Fastball: Yes, the boring fastball. If you don't seek a cutter for a primary pitch, it better be a fastball in OOTP. The fastball, doesn't stand out in many areas, but it is the highest pitch in K% by a pretty large margin.
Pros: Best K% pitch
Cons: Highest HR%/FB

Curveball: The curveball is also a very low split pitch, but in almost every other way is very close to average.
Pros: Low split difference.
Cons: Doesn't really stand out otherwise in any way.

Forkball: The forkball is really, really similar to the changeup, but has slightly lower HR%/BB and not as good of splits.
Pros: Tied for lowest in BB%, below average HR%/FB
Cons: Below average K%, will give up long ball to opposite handed hitters

Splitter: Just as the changeup and forkball are very similar, the curveball and splitter also yield similar results overall. The splitter is different in that it is slightly better against same-handed batters, but does not hold it's own as well against opposite-handed batters.
Pros: all-around decent secondary offering
Cons: not very strong against opposite-handed batters

Slider: The slider has always been one of the premier MLB pitches, but from this data, it is not very exciting in OOTP. It has the highest K% rate of any secondary offering, but otherwise is average, if not below average, in almost every other way.
Pros: decent K%
Cons: one of the worst pitches against opposite handed batters

Sinker: Chicks love the long ball, and as a GM, I hate it when my pitchers allow it. The sinker is the pitch of choice for preventing home runs. However, it is one of the worst pitches in almost every other category, and one of the worst pitches for split differences.
Pros: prevents home runs, heavy groundball rate
Cons: Lowest in K%, lowest in BB%, highest in BABIP allowed, and worst for split differences in all of those areas.

SMALL SAMPLE ALERT!

Circle Change: In my study, the circle change simply didn't have enough of a sample size to be consider significant. However, the limited results were very promising. It had the highest K%, lowest BB%, and second lowest HR%/FB (after the sinker). Has OOTP provided us with an overpowered pitch? It's hard to say with such limited data, but I think the circle change may be one to target. If I had to rank it with the above pitches, I would cautiously put it first, above the cutter.
Pros: Incredible K% and BB%, low HR%/FB without the cons of the sinker.
Cons: Low sample size, not much else.

Knucleball: The knuckleball is a cult-like pitch, and people either seem to love it or hate it. In my EXTREMELY small sample, it had elevated HR%/FB, but elite BABIP allowed. It won't help your FIP, but it may still win you some games. If I had to rank it with the above pitches, I would put it ahead of the splitter but below the forkball.
Pros: Potentially a standout pitch in BABIP allowed.
Cons: Small sample size, high HR%/FB

Knuckle Curve: Craig Kimbrel is one of the game's elite relievers using an outstanding knuckle curve. However, the pitch didn't really stand out in any way in my data. Consider it similar to the slider, with lower split differences but also lower K%. If I had to rank it with the above pitches, I would put it just ahead of the slider but below the splitter and knuckleball.
Pros: lower BABIP allowed pitch.
Cons: doesn't stand out in many, if any, ways. Low sample size.

Screwball: Only 5 pitchers in 5 full fictional leagues had a screwball rated 50 or higher. It is hardly useful to make any judgements from such a small sample size, but from the data gathered, the screwball did not impress. Yes, it had the lowest split differences in K% and BB% (and is the only pitch with reverse K% splits), but underwhelmed overall across the board. If I had to rank it among the above pitches, I would put it second to last, just in front of the sinker and below the slider.
Pros: good pitch against opposite handed batters
Cons: very low sample size, below average in K%, HR%/FB, and BABIP allowed.

[jpeters1734]

Hey thanks for this research. I'm surprised noone responded but, that's probably because this is in the general discussions forum. I think what you've discovered is not that certain pitches by themselves affect starters, but that pitch types are not equal when they get calculated into the stuff rating.

[Craig Scarborough]

Great stuff! Well done.

Interesting as I had always heard that curveballs are largely ineffective from the opposite-side. Your data is showing that it's negligible.

[Zorian77]

Interesting data. I'm surprised sliders and sinkers are that bad. Although I guess it would explain why a lot of slider dependent relievers get rocked fairly consistently despite otherwise having good stats. It would also explain why pitchers with circle changes and forkballs seem to do better than their stats would otherwise suggest.

I'm especially surprised at how bad the sinker split difference is. This might help with pitch selection during a manual sim quite a bit.

[zjl630]

I'm curious if a starter can succeed with ANY three pitches, or if it needs to be one fastball-type, one breaking ball, and one offspeed pitch.

Is a fastball, curveball, knuckle curve just as good as a fastball, curveball, changeup mix?

[brotherblues]

Nothing beats a *truly* elite knuckleball... though in the years and thousands of seasons I've simmed, only one has been spit out by the sim. He pitched 20 years in the bigs and had to buy an extra house to fit his trophies. His velocity was at like 84 in the end and he didn't give a f.

[brotherblues]

Originally Posted by AaD88

This post can also be viewed on my league's forums: https://statsplus.net/fightclubbaseb...php?f=13&t=347

We're also always looking for new members on our waitlist! https://forms.gle/FJTVTyLFqT5ERJrQ7

----------------------------------------------------------------------------------------------

I've always wanted to research individual pitches in OOTP and how they affect starters. This article is meant to help identify some trends and maybe give more insight to OOTP GMs.

METHOD
I created five fictional leagues in OOTP20 with 100% ratings and identical settings. I did not put a lot of time into detailing the settings; they are all fictional OOTP standard and may not mirror MLB stats perfectly.

I simmed a full 162 game season in each league and dumped the pitcher stats and ratings on the last day of the season.

Every pitcher with MLB innings and both R/L splits were counted. There were 2671 players in all. I also flipped splits for LHP to indicate same/opposite handed splits.

I calculated essential stats for all pitchers, and then sorted by pitch. For each pitch, pitchers with a 50-rated pitch or better were counted and stats were accumulated.

DISCLAIMERS
Obviously, there are numerous variables in play in OOTP. It is next to impossible to single out pitches and say with certainty how they affect performance. Consider these results with caution and view them as trends.

Despite having over 2600 players over five leagues, four pitches still had incredibly small sample sizes: the Circle Change (47 total players), Knuckle Curve (42), Knuckleball (11), and Screwball (5). Because of the limited sample, I did not include the results for these four pitches in the main body of the results, and instead give a highly speculative analysis at the end based on the results that I have.

As always, a correlation does not equal causation.

RESULTS

Link to data: https://docs.google.com/spreadsheets...it?usp=sharing

ANALYSIS

To analyse the results, I am going to rank the pitches by which I think are most useful to least useful. Everyone plays the game differently, so opinions may vary.

Cutter: The cutter stands out for being one of the most split-neutral pitches in OOTP. It has below average splits in K%, BB%, and BABIP allowed. It is slightly below average in HR%/FB and BABIP allowed, and has very few weaknesses.
Pros: Smallest splits of any pitch.
Cons: Doesn't otherwise stand out. Below average K%.

Changeup: The changeup stands out as being one of the best BB% pitches, a very low split pitch, and a low BABIP pitch. All around, it is one of the best secondary offerings, but seems to be one of the toughest to develop in the game.
Pros: Tied for lowest in BB%, very good splits, low BABIP allowed.
Cons: One of the lower K% pitches, overall above average HR%/FB rate

Fastball: Yes, the boring fastball. If you don't seek a cutter for a primary pitch, it better be a fastball in OOTP. The fastball, doesn't stand out in many areas, but it is the highest pitch in K% by a pretty large margin.
Pros: Best K% pitch
Cons: Highest HR%/FB

Curveball: The curveball is also a very low split pitch, but in almost every other way is very close to average.
Pros: Low split difference.
Cons: Doesn't really stand out otherwise in any way.

Forkball: The forkball is really, really similar to the changeup, but has slightly lower HR%/BB and not as good of splits.
Pros: Tied for lowest in BB%, below average HR%/FB
Cons: Below average K%, will give up long ball to opposite handed hitters

Splitter: Just as the changeup and forkball are very similar, the curveball and splitter also yield similar results overall. The splitter is different in that it is slightly better against same-handed batters, but does not hold it's own as well against opposite-handed batters.
Pros: all-around decent secondary offering
Cons: not very strong against opposite-handed batters

Slider: The slider has always been one of the premier MLB pitches, but from this data, it is not very exciting in OOTP. It has the highest K% rate of any secondary offering, but otherwise is average, if not below average, in almost every other way.
Pros: decent K%
Cons: one of the worst pitches against opposite handed batters

Sinker: Chicks love the long ball, and as a GM, I hate it when my pitchers allow it. The sinker is the pitch of choice for preventing home runs. However, it is one of the worst pitches in almost every other category, and one of the worst pitches for split differences.
Pros: prevents home runs, heavy groundball rate
Cons: Lowest in K%, lowest in BB%, highest in BABIP allowed, and worst for split differences in all of those areas.

SMALL SAMPLE ALERT!

Circle Change: In my study, the circle change simply didn't have enough of a sample size to be consider significant. However, the limited results were very promising. It had the highest K%, lowest BB%, and second lowest HR%/FB (after the sinker). Has OOTP provided us with an overpowered pitch? It's hard to say with such limited data, but I think the circle change may be one to target. If I had to rank it with the above pitches, I would cautiously put it first, above the cutter.
Pros: Incredible K% and BB%, low HR%/FB without the cons of the sinker.
Cons: Low sample size, not much else.

Knucleball: The knuckleball is a cult-like pitch, and people either seem to love it or hate it. In my EXTREMELY small sample, it had elevated HR%/FB, but elite BABIP allowed. It won't help your FIP, but it may still win you some games. If I had to rank it with the above pitches, I would put it ahead of the splitter but below the forkball.
Pros: Potentially a standout pitch in BABIP allowed.
Cons: Small sample size, high HR%/FB

Knuckle Curve: Craig Kimbrel is one of the game's elite relievers using an outstanding knuckle curve. However, the pitch didn't really stand out in any way in my data. Consider it similar to the slider, with lower split differences but also lower K%. If I had to rank it with the above pitches, I would put it just ahead of the slider but below the splitter and knuckleball.
Pros: lower BABIP allowed pitch.
Cons: doesn't stand out in many, if any, ways. Low sample size.

Screwball: Only 5 pitchers in 5 full fictional leagues had a screwball rated 50 or higher. It is hardly useful to make any judgements from such a small sample size, but from the data gathered, the screwball did not impress. Yes, it had the lowest split differences in K% and BB% (and is the only pitch with reverse K% splits), but underwhelmed overall across the board. If I had to rank it among the above pitches, I would put it second to last, just in front of the sinker and below the slider.
Pros: good pitch against opposite handed batters
Cons: very low sample size, below average in K%, HR%/FB, and BABIP allowed.

Thank you for taking the time and most especially for sharing. Lots to dive into here.