Stats-Based CPU Sliders [MLB 14 Version]

[ralphieboy11]

Re: Stats-Based CPU Sliders [MLB 14 Version]

The first real patch is out.

http://forums.theshownation.com/showthread.php?p=209733#post209733

Most notable for this thread would be: - Average runners have better steal success than before. This is a small change, but there are a lot of close plays getting them tagged out. The effect is pronounced when playing against certain pitchers, and for these pitchers the runners were being capped from getting their best jump (compared to last year).

I am interested to see if the steal percentage with pitch speed 10 can match real life and if not if pitch speed 5 does.

Really awesome that they looked into this. My sliders were looking pretty good except for the stolen bases. They continue to prove they are the best development team in the business.

[My993C2]

Re: Stats-Based CPU Sliders [MLB 14 Version]

My sliders were looking pretty good except for the stolen bases.

I have been using a minor variation of your sliders since last week (I don't have fielding errors set as high as you do and I have also lowered the foul ball frequency to see if I can get more starters throwing into the 7th inning as opposed to being pulled in the 5th or 6th once they hit 100 pitches). Anyway I have played about half a dozen MoM games, a number of CPU vs CPU using Fast Play as well as some CPU vs CPU QC + Fast Play games using your slider settings with my minor tweaks and yes your sliders do seem to work pretty good.

[ralphieboy11]

Re: Stats-Based CPU Sliders [MLB 14 Version]

I have been using a minor variation of your sliders since last week (I don't have fielding errors set as high as you do and I have also lowered the foul ball frequency to see if I can get more starters throwing into the 7th inning as opposed to being pulled in the 5th or 6th once they hit 100 pitches). Anyway I have played about half a dozen MoM games, a number of CPU vs CPU using Fast Play as well as some CPU vs CPU QC + Fast Play games using your slider settings with my minor tweaks and yes your sliders do seem to work pretty good.

Thanks. The credit goes to the fine folks at SCEA. The game plays pretty great right out the box. Plus the work nomo does here is terrific. He tracks many more elements than I do, and I'm always using it to reconfirm anything I've seen myself.

One thing I do like to do right out of the box is lower the arm strength of the fielders. A little too much station to station baserunning for my liking. Plus you see a lot of would be doubles turn into singles as the outfielder fires the ball back into 2nd base. I usually start with that and go from there.

[BrianU]

Re: Stats-Based CPU Sliders [MLB 14 Version]

Really awesome that they looked into this. My sliders were looking pretty good except for the stolen bases. They continue to prove they are the best development team in the business.

I saw your sliders and the stats you provided. Could we get an update, have you changed anything? I am taking the same approach as you using nomos base then tuning it for User vs CPU it would be helpful to see where you are at with it.

Also what pitch speed did you use/are you using. I was going to use pitch speed 10 but after seeing how much it affects steal success I am going to stick with 5. It's a little bit of a bummer but what are you gonna do.

[ralphieboy11]

Re: Stats-Based CPU Sliders [MLB 14 Version]

I changed pitch speed to 5 way back when the devs mentioned it as an issue. I never saw evidence that it made much difference. I think you can safely leave it at 10. The stolen base issue seems like it's mostly what SCEA addressed in the patch. I have not downloaded it yet to test it out(at work now).

I'm just testing still but here are 80 games in season mode...

Everything default except...

Solid hit 6
Starter Stamina 6
Reliever Stamina 7
Pitcher Control 3
Pitcher Consistency 3
Manager Hook 4
Foul Frequency 2 (Probably should have stuck with 3 here)
Fielder Run Speed 4
Fielder Reaction 4
Fielder Arm Strength 3
Baserunner Speed 4
Fielding Errors 8
Throwing Errors 4

Mine(MLB Avg.)

Batting Average .254(.256)
Runs 3.85(4.33)
BB 2.79(3.12)
SO 7.08(7.22)
HR 0.84(0.97)
2B 1.61(1.73)
3B 0.22(0.18)
GIDP 0.73(0.75)
Errors 0.58(0.64)
HBP 0.18(0.32)
WP 0.35(0.34)

[BrianU]

Re: Stats-Based CPU Sliders [MLB 14 Version]

I changed pitch speed to 5 way back when the devs mentioned it as an issue. I never saw evidence that it made much difference. I think you can safely leave it at 10. The stolen base issue seems like it's mostly what SCEA addressed in the patch. I have not downloaded it yet to test it out(at work now).

I'm just testing still but here are 80 games in season mode...

Everything default except...

Solid hit 6
Starter Stamina 6
Reliever Stamina 7
Pitcher Control 3
Pitcher Consistency 3
Manager Hook 4
Foul Frequency 2 (Probably should have stuck with 3 here)
Fielder Run Speed 4
Fielder Reaction 4
Fielder Arm Strength 3
Baserunner Speed 4
Fielding Errors 8
Throwing Errors 4

Mine(MLB Avg.)

Batting Average .254(.256)
Runs 3.85(4.33)
BB 2.79(3.12)
SO 7.08(7.22)
HR 0.84(0.97)
2B 1.61(1.73)
3B 0.22(0.18)
GIDP 0.73(0.75)
Errors 0.58(0.64)
HBP 0.18(0.32)
WP 0.35(0.34)

Thanks! The biggest area I have found needing to be changed from nomos CPU v CPU sliders is to raise the human control slider. I am going to track my batting numbers they might need adjusting but it will be awhile before I can tell.

What difficulties for hitting/pitching and what control methods are you using?

[ralphieboy11]

Re: Stats-Based CPU Sliders [MLB 14 Version]

Thanks! The biggest area I have found needing to be changed from nomos CPU v CPU sliders is to raise the human control slider. I am going to track my batting numbers they might need adjusting but it will be awhile before I can tell.

What difficulties for hitting/pitching and what control methods are you using?

I should have mentioned they are all cpu vs. cpu games, but I use the stats to gain knowledge about the engine. I like to play Manage only Mode games and run cpu vs. cpu games a lot although I also enjoy playing myself.

When I play myself I'm usually on All-Star level, Classic pitching and Timing hitting if I'm looking for a true offline simulation of the game. I do try almost all the controls at some point though.

[BrianU]

Re: Stats-Based CPU Sliders [MLB 14 Version]

I should have mentioned they are all cpu vs. cpu games, but I use the stats to gain knowledge about the engine. I like to play Manage only Mode games and run cpu vs. cpu games a lot although I also enjoy playing myself.

When I play myself I'm usually on All-Star level, Classic pitching and Timing hitting if I'm looking for a true offline simulation of the game. I do try almost all the controls at some point though.

Ah I thought your stats were user vs cpu. Still good to see more data.

I hope you are right about pitch speed not affecting steals as much as I think it is. I just am looking at nomos data from last year when he used pitch speed 5 compared to this year when he has been using pitch speed 10 and it just seems such a wide gap. I know there are other factors, it just seems how Brian described the steals fix targetting slower speed guys stealing bases it won't big a huge difference to steal success, but of course we will see in the data soon. With the steal percentage at 40% currently it needs to add another 30% and almost double just to be around MLB average range, that seems quite a bit.

[nonoirie]

Re: Stats-Based CPU Sliders [MLB 14 Version]

I changed pitch speed to 5 way back when the devs mentioned it as an issue. I never saw evidence that it made much difference. I think you can safely leave it at 10. The stolen base issue seems like it's mostly what SCEA addressed in the patch. I have not downloaded it yet to test it out(at work now).

I'm just testing still but here are 80 games in season mode...

Everything default except...

Solid hit 6
Starter Stamina 6
Reliever Stamina 7
Pitcher Control 3
Pitcher Consistency 3
Manager Hook 4
Foul Frequency 2 (Probably should have stuck with 3 here)
Fielder Run Speed 4
Fielder Reaction 4
Fielder Arm Strength 3
Baserunner Speed 4
Fielding Errors 8
Throwing Errors 4

Mine(MLB Avg.)

Batting Average .254(.256)
Runs 3.85(4.33)
BB 2.79(3.12)
SO 7.08(7.22)
HR 0.84(0.97)
2B 1.61(1.73)
3B 0.22(0.18)
GIDP 0.73(0.75)
Errors 0.58(0.64)
HBP 0.18(0.32)
WP 0.35(0.34)

Ralphie,

How's reliever stamina 7 working out for you?

[Heroesandvillains]

Re: Stats-Based CPU Sliders [MLB 14 Version]

I changed pitch speed to 5 way back when the devs mentioned it as an issue. I never saw evidence that it made much difference. I think you can safely leave it at 10. The stolen base issue seems like it's mostly what SCEA addressed in the patch. I have not downloaded it yet to test it out(at work now).

I'm just testing still but here are 80 games in season mode...

Everything default except...

Solid hit 6
Starter Stamina 6
Reliever Stamina 7
Pitcher Control 3
Pitcher Consistency 3
Manager Hook 4
Foul Frequency 2 (Probably should have stuck with 3 here)
Fielder Run Speed 4
Fielder Reaction 4
Fielder Arm Strength 3
Baserunner Speed 4
Fielding Errors 8
Throwing Errors 4

Mine(MLB Avg.)

Batting Average .254(.256)
Runs 3.85(4.33)
BB 2.79(3.12)
SO 7.08(7.22)
HR 0.84(0.97)
2B 1.61(1.73)
3B 0.22(0.18)
GIDP 0.73(0.75)
Errors 0.58(0.64)
HBP 0.18(0.32)
WP 0.35(0.34)

Very impressive! Are you going through all games on the schedule like Nomo?

If so...bump power a click and call it a day???

Other than SB's of course.

[nomo17k]

Re: Stats-Based CPU Sliders [MLB 14 Version]

Ah I thought your stats were user vs cpu. Still good to see more data.

I hope you are right about pitch speed not affecting steals as much as I think it is. I just am looking at nomos data from last year when he used pitch speed 5 compared to this year when he has been using pitch speed 10 and it just seems such a wide gap. I know there are other factors, it just seems how Brian described the steals fix targetting slower speed guys stealing bases it won't big a huge difference to steal success, but of course we will see in the data soon. With the steal percentage at 40% currently it needs to add another 30% and almost double just to be around MLB average range, that seems quite a bit.

If the last year gives any indication, max pitch speed won't be an issue for steals because the effect caused by that change is within a range correctable with sliders. You just need to make them easier to steal by adjusting by a click or two.

[BrianU]

Re: Stats-Based CPU Sliders [MLB 14 Version]

If the last year gives any indication, max pitch speed won't be an issue for steals because the effect caused by that change is within a range correctable with sliders. You just need to make them easier to steal by adjusting by a click or two.

I hope so but looking at your data from 4/11 Steal Ability 10 lead to a 41.5% steal rate and then the data from 4/15 Steal Ability 0 lead to 40% steal rate. All the other sims are around 40% as well except for 4/9 which was at 50%. Is the sample size too small to say that the slider is broken? From the sounds of the patch notes it wouldn't have changed the slider behavior.

It just is so noticeably out of line compared to the rest of the averages. And last year it was spot on at that 70-75% range in all your data sets.

[ralphieboy11]

Re: Stats-Based CPU Sliders [MLB 14 Version]

Ralphie,

How's reliever stamina 7 working out for you?

It's been fine. Just a personal preference here. It's more about the cpu's usage of the bullpen in the long haul than one particular game. In past years when running cpu vs. cpu seasons I've noticed some bullpens get extremely fatigued during some stretches. Maybe it's a few long extra innings games or an extended period without an off day.

In past years once all of the bullpen pitchers were used the cpu would turn to the next day's starter. I'm trying to avoid situations like that. I haven't seen anything like that this year but I have seen all available bullpen pitchers used in some extra inning games.

Very impressive! Are you going through all games on the schedule like Nomo?

Yes. Same as nomo. I had one season going on PS3 and one on Vita. I'm probably going to be PS3 only now to sort out the stolen base thing until a patch comes for the Vita.

[BrianU]

Re: Stats-Based CPU Sliders [MLB 14 Version]

Ralphie you did your sims with pitch speed 5 and steal success 5 right? Did you track SB rate you could share, was that around 40% too?

[ralphieboy11]

Re: Stats-Based CPU Sliders [MLB 14 Version]

While I was running those cpu games I was moving stolen base sliders every which way I could, so I don't really have a good base of numbers for that.

I could never get stolen base success over 50% no matter what I tried before the patch.

I'll start up a new batch of tests with the game patched now. The explanation given by Brian here and what I read in the patch notes seem to be exactly what I was seeing. A lot of close plays at 2nd were outs. If they just tweaked a few of those in favor of the runner it could make a big difference in the success rate.