Stats-Based CPU Sliders [MLB 12 Version]

[DarkSith777]

Re: Stats-Based CPU Sliders [MLB 12 Version]

Can you make a slider set for madden?!

[nomo17k]

Re: Stats-Based CPU Sliders [MLB 12 Version]

Can you make a slider set for madden?!

Haha I'm almost exclusively a baseball person, so I wouldn't be good at tuning Madden.

I think baseball is easier to do this sort of "stats-base" tuning in a sense because we get a lot of meaningful stats from everywhere, almost taken for granted...

[ralphieboy11]

Re: Stats-Based CPU Sliders [MLB 12 Version]

I'm really enjoying these sliders for cpu vs. cpu games. Thanks for your research. I was able to take in quite a few more over the weekend. I had a run of a lot of lower scoring games, but I think that was because the games were with a lot of top of the rotation guys.

I guess the GIDPs is still a little a bit of an issue, but not too bad. The infielders really seem to get the ball in and out of their glove so quick sometimes.

The one animation that doesn't seem to pass the look test is you still see a lot of hits down the line past 1st and 3rd baseman that end up as singles instead of doubles. I'm not sure if that due to the outfielder getting to the ball a little quickly or the ball not rolling out all the way to the corner.

I would be scared to play around with that, because the numbers on singles, doubles, and triples are so good right now.

[nomo17k]

Re: Stats-Based CPU Sliders [MLB 12 Version]

I'm really enjoying these sliders for cpu vs. cpu games. Thanks for your research. I was able to take in quite a few more over the weekend. I had a run of a lot of lower scoring games, but I think that was because the games were with a lot of top of the rotation guys.

I guess the GIDPs is still a little a bit of an issue, but not too bad. The infielders really seem to get the ball in and out of their glove so quick sometimes.

The one animation that doesn't seem to pass the look test is you still see a lot of hits down the line past 1st and 3rd baseman that end up as singles instead of doubles. I'm not sure if that due to the outfielder getting to the ball a little quickly or the ball not rolling out all the way to the corner.

I would be scared to play around with that, because the numbers on singles, doubles, and triples are so good right now.

GDIPs... the best we can do up to this point is about 1 per game. I think something needs to be drastically changed (with other consequences of course) to do better.

That down the line thing... I think it's been one of the most hotly discussed plays in 12. I think it's a combination of outfielders hugging the line a tiny bit too much, and the grounder slowing down a bit too much. I guess we could try reducing Fielder Run Speed and Reaction one more click, if that won't have any significant effects on stats... but it's not that we are short of doubles at this point so as far as stats are concerned, it's not a huge issue.

Before doing anything, I want to see if the upcoming patch changes anything that affects things like this.

Triples are a bit much, I can see 50% reduction there, but a lot of extra triples are given because of missed dive attempts... we cannot do much there.

[ralphieboy11]

Re: Stats-Based CPU Sliders [MLB 12 Version]

I will be interested to see if they patch anything that makes a drastic change as well. I almost hope they don't really mess with anything major regarding gameplay. I can't remember a game I've said that about in a long time. I can live with the way things are right now.

[nomo17k]

Re: Stats-Based CPU Sliders [MLB 12 Version]

I will be interested to see if they patch anything that makes a drastic change as well. I almost hope they don't really mess with anything major regarding gameplay. I can't remember a game I've said that about in a long time. I can live with the way things are right now.

Yup I feel the same way... hope it doesn't change too many things at once, since the game balance is pretty good. We have sliders to tune some aspects on our own as well.

I'd just like to see grounders slowing down a bit less especially when they are shots to the outfield, and high bouncing a bit toned down to reduce ground rule doubles. Otherwise I'm pretty happy with what I have for this year.

[nomo17k]

Re: Stats-Based CPU Sliders [MLB 12 Version]

Done 75 games with the Apr. 5 test set. This is basically the most up-to-date working set, except I was testing Starter Stamina = 10.

Stats are most where I want them to be, but I want to tally some numbers from previous test to see if I may want to make minor adjustments here and there.

Also, I want to see what Starter Stamina = 10 did to the game, and decide what value to use for that slider. Will be back with more thoughts after doing some number crunching...


##### Apr. 5 test set #####

Contact: 4
Power: 5
Timing: 6
Foul Frequency: 4
Solid Hits: 4
Starter Stamina: 10
Reliever Stamina: 5
Pitcher Control: 5
Pitcher Consistency: 4
Strike Frequency: 5
Manager Hook: 5
Pickoffs: 5
Pitch Speed: 10
Fielding Errors: 7
Throwing Errors: 5
Fielder Run Speed: 4
Fielder Reaction: 4
Fielder Arm Strength: 2
BR Speed: 6
BR Steal Ability: 4
BR Steal Frequency: 8
Wind: 4
Injury Frequency: 5

[ralphieboy11]

Re: Stats-Based CPU Sliders [MLB 12 Version]

I've been using these as well with success.

For what it's worth I've had starter stamina at 7. Seems ok. I've also bumped up reliever stamina to 7 and manager hook at 4.

It seems the decision to sub pitchers by the AI is a combination of best matchup scenario and the energy of the pitcher on the mound. By bumping up reliever stamina I was hoping to avoid the situation where a bullpen gets gassed over a stretch. You will see some games where the AI brings in a tired pitcher, has him pitch to one batter, and then brings in someone else. There ends up being too few guys in the bullpen and then the AI starts turning to next day starters.

I may bump up reliever stamina even more. My feeling is the relievers should perform to their ratings anyway. Stamina is more important for starters, because it requires them to take days off between starts.

[nomo17k]

Re: Stats-Based CPU Sliders [MLB 12 Version]

Thoughts on steal related sliders...

I've been using the following combination:

BR Speed: 6
BR Steal Ability: 4
BR Steal Frequency: 8

And this is what I have... 231 SB, 96 CS in 412 games, so in terms of per game per team, I have

SB: 0.56 (0.64)
CS: 0.23 (0.25)
SB%: 70.6% (72.3%)

So the number of steal attempts are still a tiny bit down, and to a lesser extent steal success % as well.

This is really a minor issue and any change isn't necessary. But eyeballing the previous tests, I notice steals are a bit down overall, despite the increase in Steal Frequency slider.

I think I'm going to increase BR Steal Frequency to 9. I could also reduce BR Steal Ability by one to increase the success rate, but I'm hoping the increased BR Steal Frequency will translate to good stealers attempting more steals, effectively increasing the success rate. So I'll keep that slider where it is.

[nomo17k]

Re: Stats-Based CPU Sliders [MLB 12 Version]

Error related sliders...

I've been using:

Fielding Errors: 7
Throwing Errors: 5

for most of my testing. And the combined per game per team stats after 341 games look like this:

FE: 0.31 (0.33)
TE: 0.28 (0.28)

which look fine so I'll keep those slider setting.

[bubs3141]

Re: Stats-Based CPU Sliders [MLB 12 Version]

so is the april 5 set your newest? Does it replace the mar 31 on page 1?

[nomo17k]

Re: Stats-Based CPU Sliders [MLB 12 Version]

What did Starter Stamina = 10 do for the game?

First of all, the main motivation for testing this was to see if the "6th inning scoring peak" could be alleviated by letting starters go deeper with more stamina. I revisit the figure that I made earlier (attached), and it does seem to delay the onset of the late-inning scoring peak from 6th to about 7th inning. Note that I had a lot fewer games for Stamina = 10 so data are of course noisier, but we can clearly see the trend pushed back to latter innings.

So, I would conclude part of the 6th inning problem is that starters run out of energy a bit too soon.

What's the value of Starter Stamina to use then?

With Stamina = 5, starters averaged 94.2 pitches per start.
With Stamina = 10, starters averaged 101.7 pitches per start.
The MLB ave is 96.5 pitches per start.

With Stamina = 5, starters averaged 5.80 IP per start.
With Stamina = 10, starters averaged 6.15 IP per start.
The MLB ave is 5.96 IP per start.

Based on the discussion in this post:

http://www.operationsports.com/forum...post2043524247

it looks like Stamina = 7 would make sense.

[nomo17k]

Re: Stats-Based CPU Sliders [MLB 12 Version]

Reliever usage...

On average, CPU uses about 4.15 pitchers per game, compared to 3.88 pitchers in MLB.

CPU obviously uses more than IRL. Haven't really tested the slider this year, but from last year's testing it might make sense to lower it again...

So I'll use Manager Hook = 4.

Unfortunately this (and increased starter stamina) probably leads to many more complete games by starters, but that's probably because CPU leaving them in a bit longer regardless of pitch count, if they are pitching well and they are taking leads by a comfortable margin.

I don't think you can tinker that aspect of AI, so this is a compromise.

[nomo17k]

Re: Stats-Based CPU Sliders [MLB 12 Version]

So this is the Apr. 9 release! I'd consider this is the "release" version.

#### Apr. 9 working set ####

Contact: 4
Power: 5
Timing: 6
Foul Frequency: 4
Solid Hits: 4
Starter Stamina: 7
Reliever Stamina: 5
Pitcher Control: 5
Pitcher Consistency: 4
Strike Frequency: 5
Manager Hook: 4
Pickoffs: 5
Pitch Speed: 10
Fielding Errors: 7
Throwing Errors: 5
Fielder Run Speed: 4
Fielder Reaction: 4
Fielder Arm Strength: 2
BR Speed: 6
BR Steal Ability: 4
BR Steal Frequency: 9
Wind: 4
Injury Frequency: 5

[nomo17k]

Re: Stats-Based CPU Sliders [MLB 12 Version]

I've been using these as well with success.

For what it's worth I've had starter stamina at 7. Seems ok. I've also bumped up reliever stamina to 7 and manager hook at 4.

It seems the decision to sub pitchers by the AI is a combination of best matchup scenario and the energy of the pitcher on the mound. By bumping up reliever stamina I was hoping to avoid the situation where a bullpen gets gassed over a stretch. You will see some games where the AI brings in a tired pitcher, has him pitch to one batter, and then brings in someone else. There ends up being too few guys in the bullpen and then the AI starts turning to next day starters.

I may bump up reliever stamina even more. My feeling is the relievers should perform to their ratings anyway. Stamina is more important for starters, because it requires them to take days off between starts.

I'm also going ahead with SS = 7 and MH = 4. Seems to make sense (except CGs but we cannot do anything reasonable about it).

In principle, I'd like to set Reliever Stamina to the same value as Starter's... they are after all same human being right... But I think a lot of guys use that slider to tweak how often you need to rest those guys between their appearances. I think that also make sense, but I feel it's more of personal choice thing than anything else, so I decided to leave it at 5 this time.

so is the april 5 set your newest? Does it replace the mar 31 on page 1?

Apr. 9 is the latest and greatest. Updated the first post.