Stats-Based CPU Sliders [MLB 14 Version]

[nomo17k]

Quote: Originally Posted by Trojan Man

First-timer to this thread. Are these meant to be used in HUM v. CPU games as well? And are the slider values meant to be adjusted from All Star default?

I don't think I *mean* anything, but rather I am just sharing information that could be used in any ways people find useful.

My intention is that I first want CPUs to play realistically against each other, partly because I enjoy watching them play against each other, and also because it forms the base of whatever changes that I want to make to make my own playing realistic.

For example, if my playing tendency is exactly the same as CPU, then this CPU slider set is probably a good set for my own playing as well. If I deviate a bit from CPU's tendency, then I can adjust HUM sliders. Do I need a bigger PCI? I change HUM Contact. Am I too accurate in locating pitches? I vary HUM Pitcher Consistency/Control sliders. So in view of HUM usage, I'm just trying to come up with a "reference" set from which I vary the HUM side to account for my own tendencies.

[RogueHominid]

Quote: Originally Posted by nomo17k

I don't think I *mean* anything, but rather I am just sharing information that could be used in any ways people find useful. My intention is that I first want CPUs to play realistically against each other, partly because I enjoy watching them play against each other, and also because it forms the base of whatever changes that I want to make to make my own playing realistic. For example, if my playing tendency is exactly the same as CPU, then this CPU slider set is probably a good set for my own playing as well. If I deviate a bit from CPU's tendency, then I can adjust HUM sliders. Do I need a bigger PCI? I change HUM Contact. Am I too accurate in locating pitches? I vary HUM Pitcher Consistency/Control sliders. So in view of HUM usage, I'm just trying to come up with a "reference" set from which I vary the HUM side to account for my own tendencies.

Fair enough, and thanks!

[bcruise]

Quote: Originally Posted by nomo17k

I know... with what has been going on at Wrigley for over 100 years, you need something to celebrate, even if it's just a base on balls...

Low blow. Laser-guided 100% accuracy, but low.

BTW, I went back to default for a game and played the same matchup. Cole's line this time: 8 IP, 0 ER, 4 H, 0 BB, 11 K. You could argue that that's a much more realistic line for facing the Cubs.

[nonoirie]

When Pitcher Control is decreased, does pitchers throw more "meatballs" along with pitches outside the zone?

[nomo17k]

Quote: Originally Posted by nonoirie

When Pitcher Control is decreased, does pitchers throw more "meatballs" along with pitches outside the zone?

In short, I think it does increase both meatballs and pitches outside the zone.

From what I have gathered from things like slider descriptions and Brian's old posts and such, the way I currently guess and understand the Pitcher Control/Consistency sliders goes like this.

First of all, you should look at the attached pitch location chart when Lester was throwing all his pitches aiming down the middle. You are looking from the view of catcher. You can see the distribution of pitches roughly forms a slanted oval. Since Lester is a lefty with a three-quarter delivery, when he releases his pitch a bit early, it tends to go upper right corner, and when he releases a bit late, it tends to go toward lower left corner.

Now, it's clearer to draw this than explain in words, so in the bottom figure I reproduced a kind of pitch location chart but for the case the pitcher was aiming outside low on the black, i.e., the center of the ovals.

There are three ovals for three different Pitcher Control settings: blue for high, green for default, red for low slider value for Pitcher Control. The size of oval roughly indicates how much pitch location can vary, if a pitcher actually tries to locate a pitch at the center of oval. We notice that the about half of the oval is in the strike zone in each case, despite the widely different sizes.

Looking at the figure, we see that by lowering Pitcher Control the oval gets so large that a significant portion of it lies close to the deepest part of the strike zone. This is probably why lowering Pitcher Control increases the number of meatballs. (The slider description recommends lowering Contact when Pitcher Control is lowered, likely because there will be too many meatballs caused this way.)

We also see that changing the Pitcher Control oval does not change the natural strike % of pitchers all that much (for pitches on the black anyways), because about the same proportion of area within an oval stays in and outside the strike zone, regardless of the oval size. This is probably why Pitcher Control does not have as large an effect on the amount of walks induced by its adjustment.

I also drew a bunch of yellow heart marks. Those are examples of pitch locations when the pitcher totally misses his spot, far from the intended spot (outside low). While the vast majority of pitches go within the oval, some miss wide like this and result in things like wild pitches and hit-by-pitches, etc. How often a pitcher misses wide this way is controlled by the Pitcher Consistency slider. Since errant pitches like this are more likely to miss far off the strike zone, this is likely why Pitcher Consistency is the most effective slider for increasing the number of walks.

[BrianU]

Quote: Originally Posted by Trojan Man

First-timer to this thread. Are these meant to be used in HUM v. CPU games as well? And are the slider values meant to be adjusted from All Star default?

Like Nomo said this work makes a good base to make your own set. Some of the sliders would work better for Human vs CPU than others, like the error sliders, CPU pitching, stamina/hook..

A lot of it depends on your own playing style and control methods. How 'realistically' do you pitch and the decisions you make in hitting. I personally track the same metrics of my own play as I do of the CPU's play using these spreadsheets I created: http://www.operationsports.com/forum...ful-links.html

Comparing your actions to the MLB averages lets you know whether you throw 'too many' balls/strikes, swing too much, chase too much, take pitches too much etc.. I try to play as true to life as possible. I also use classic pitching/timing hitting/auto fielding&baserunning and I feel this setup plays best with nomo's findings.

[ralphieboy11]

Quote: Originally Posted by nomo17k

In short, I think it does increase both meatballs and pitches outside the zone. From what I have gathered from things like slider descriptions and Brian's old posts and such, the way I currently guess and understand the Pitcher Control/Consistency sliders goes like this. First of all, you should look at the attached pitch location chart when Lester was throwing all his pitches aiming down the middle. You are looking from the view of catcher. You can see the distribution of pitches roughly forms a slanted oval. Since Lester is a lefty with a three-quarter delivery, when he releases his pitch a bit early, it tends to go upper right corner, and when he releases a bit late, it tends to go toward lower left corner. Now, it's clearer to draw this than explain in words, so in the bottom figure I reproduced a kind of pitch location chart but for the case the pitcher was aiming outside low on the black, i.e., the center of the ovals. There are three ovals for three different Pitcher Control settings: blue for high, green for default, red for low slider value for Pitcher Control. The size of oval roughly indicates how much pitch location can vary, if a pitcher actually tries to locate a pitch at the center of oval. We notice that the about half of the oval is in the strike zone in each case, despite the widely different sizes. Looking at the figure, we see that by lowering Pitcher Control the oval gets so large that a significant portion of it lies close to the deepest part of the strike zone. This is probably why lowering Pitcher Control increases the number of meatballs. (The slider description recommends lowering Contact when Pitcher Control is lowered, likely because there will be too many meatballs caused this way.) We also see that changing the Pitcher Control oval does not change the natural strike % of pitchers all that much (for pitches on the black anyways), because about the same proportion of area within an oval stays in and outside the strike zone, regardless of the oval size. This is probably why Pitcher Control does not have as large an effect on the amount of walks induced by its adjustment. I also drew a bunch of yellow heart marks. Those are examples of pitch locations when the pitcher totally misses his spot, far from the intended spot (outside low). While the vast majority of pitches go within the oval, some miss wide like this and result in things like wild pitches and hit-by-pitches, etc. How often a pitcher misses wide this way is controlled by the Pitcher Consistency slider. Since errant pitches like this are more likely to miss far off the strike zone, this is likely why Pitcher Consistency is the most effective slider for increasing the number of walks.

This was very informative.

It was interesting to me that you mentioned the three-quarter delivery from Lester. Arm angle definitely has a way in determining where a pitcher misses his pitches.

I have noticed through the years sidearm cpu pitchers perform rather poorly compared to their more normal counterparts. I haven't documented it, but I'm almost certain a pitcher like Brad Ziegler or Pat Neshek is a little wilder than their ratings would indicate.

This may have to do with their pitch repertoire combined with arm angle. Mostly I've witnessed a lot of balls ending up in the dirt.

These pitchers can be effective from time to time, but usually if I'm just playing MoM and trying to build a team, I avoid them.

[nomo17k]

Quote: Originally Posted by ralphieboy11

This was very informative. It was interesting to me that you mentioned the three-quarter delivery from Lester. Arm angle definitely has a way in determining where a pitcher misses his pitches. I have noticed through the years sidearm cpu pitchers perform rather poorly compared to their more normal counterparts. I haven't documented it, but I'm almost certain a pitcher like Brad Ziegler or Pat Neshek is a little wilder than their ratings would indicate. This may have to do with their pitch repertoire combined with arm angle. Mostly I've witnessed a lot of balls ending up in the dirt. These pitchers can be effective from time to time, but usually if I'm just playing MoM and trying to build a team, I avoid them.

I actually don't believe that sidearm guys are naturally handicapped in the command department in the game, but their pitch trajectory tends to be different from those with normal deliveries (i.e., pitch tends to break more laterally than vertically).

I think it is true that sidearm guys tend to be on the wilder side in real life, partly because of the unnatural ways that they deliver, but also because they can get away with it. The biggest advantage that these guys have is the deception due to their delivery and the unfamiliarity of hitters to the way their pitches break, so I think they can remain successful even when they cannot locate pitches that well. If you look at their stats, they tend to walk a lot (high BB/9) but less hittable (low H/9), so the overall WHIP remains respectable.