Stats-Based CPU Sliders [MLB 12 Version]

[steviegolfballs]

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

Thank you for the feedback... now you are always a step ahead of me haha.

It's good to know that's what's going to happen if you lower SH too much; I didn't like how quite a few batted balls stay around the home plate in previous years game. We don't see that as often in 12, which is a good thing.

But if necessary we could eventually use just a click down (or two) if that actually reduces GIDPs. I don't see that affectineg the rest of the game in any major way.

Since I increased Timing to 6, I've been keeping track of ground ball/line drive/fly ball fractions, and they seem to be fine (except line drives, which have always been a couple % lower than the MLB ave... not a big deal).

Visually, what I feel is that the game is a tiny bit poppy (though stats have been fine)... meaning line drives are almost always hard hit... there can be more very weakly hit grounders, etc. New ball physics is great, but I still think the game can increase "poor hit" variety. That probably also leads to less GIDPs.

If I could, I want to increase weaker hits variety, without compromising the rest, which a click down with Solid Hit might do.

I agree with your thoughts, I hope there can be a compromise with the solid hit sliders and hit variety. I am working with your march 25 sliders right now. Let me know if you have thoughts and I can help test.

[steviegolfballs]

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

Nomo, what are some of the extra settings you are using? I have played a few games with your sliders from march 25 and the strikeouts seem to be very low. I use perfect umpires and all-star settings.

[nomo17k]

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

Nomo, what are some of the extra settings you are using? I have played a few games with your sliders from march 25 and the strikeouts seem to be very low. I use perfect umpires and all-star settings.

Difficulty settings don't matter... I just don't see any evidence that they do. I've been trying various combinations for a while but I don't see any changes other than expected from natural variance... and you see from the stats that I have posted so far that they don't vary much.... (I've been changing the settings here and there, but the stats are all close to MLB ave.) If in doubt, I'd use All star though.

Strikeouts do fluctuate a lot, since pitchers with high K/9 ratings get more whiffs while contact pitchers don't as often... so I'd be careful what kind of pitchers you are using in your games.

For an instance, in my test with the Mar. 25 set, Ks per game has been 7.23, a bit high compared to the MLB ave of 7.04. But all is still okay, since I've had a few very long extra inning games which skewed my stats a tiny bit. If you compute K/AB (fraction of at bats ending up in Ks), it's 20.4% in my games compared to 20.7% of the MLB ave. So it's pretty close.

[steviegolfballs]

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

That's reassuring, thanks. I don't have the sample size needed to make changes yet but I'm only averaging about 5.9 K's per team game so I am a bit concerned. The other thing that might make a difference is that I believe you are testing in season mode, so your batters may have a slight fatigue factor while I am testing in exhibition mode where all are at full strength. In your games the pitchers are more rested than the batters and that could possibly lend itself to a few more K's.

[nomo17k]

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

That's reassuring, thanks. I don't have the sample size needed to make changes yet but I'm only averaging about 5.9 K's per team game so I am a bit concerned. The other thing that might make a difference is that I believe you are testing in season mode, so your batters may have a slight fatigue factor while I am testing in exhibition mode where all are at full strength. In your games the pitchers are more rested than the batters and that could possibly lend itself to a few more K's.

I doubt stamina would have THAT much effect, but then it might be...

One quick (but not rigorous at all) way to test if you are seeing a low K number for your given (small) sample size is to do something like this. Say your 5.9 Ks per game is over 10 games.

Should I get more games in before making any big conclusion?

Say what if the above number is due to pure luck... and what if just the next game a trend changes, and the team collects 12 Ks (which is reasonable; happened in more than 9 games in my > 60 games so far). Then your per game average jumps up:

(59 + 12) / 11 = 6.45 Ks per game

So all it takes is just one game of high Ks performance to bring the stats to what you expect to be normal.

This exercise basically shows you that the sample size isn't big enough yet to worry about things.

[steviegolfballs]

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

I doubt stamina would have THAT much effect, but then it might be...

One quick (but not rigorous at all) way to test if you are seeing a low K number for your given (small) sample size is to do something like this. Say your 5.9 Ks per game is over 10 games.

Should I get more games in before making any big conclusion?

Say what if the above number is due to pure luck... and what if just the next game a trend changes, and the team collects 12 Ks (which is reasonable; happened in more than 9 games in my > 60 games so far). Then your per game average jumps up:

(59 + 12) / 11 = 6.45 Ks per game

So all it takes is just one game of high Ks performance to bring the stats to what you expect to be normal.

This exercise basically shows you that the sample size isn't big enough yet to worry about things.

Agreed, I totally buy in to the large sample philosophy (which is why I am here in your post). I notice trends but you are right, I was low on doubles and then just had a game with 7 so it happens.

[Bobhead]

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

Agreed, I totally buy in to the large sample philosophy (which is why I am here in your post). I notice trends but you are right, I was low on doubles and then just had a game with 7 so it happens.

7 doubles??? Please tell me this was a CPU game. I'm lucky if I hit 2 in 1 game, while the CPU hits them at-will. I'm sure there's a solution somewhere, though. I'm just venting.

[nomo17k]

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

So I've played all games in April at this point.

Since not many people are crazy enough to play every single games on schedule as CPU vs. CPU, I thought some might be interested in how the stats look like if you never sim games. So I'll post some that I find interesting.

Here's the standings, for a starter.

[nomo17k]

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

Here's the National League.

[nomo17k]

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

Batting average and HRs.

[nomo17k]

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

RBIs and SBs.

[nomo17k]

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

Ws and SVs.

[nomo17k]

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

ERA and SOs.

[nomo17k]

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

Doc Halladay considering retirement.

[nomo17k]

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

I've been emphasizing recently that I'm here at OS mainly to discuss the Show.... and LOVE.

What's better time to look for that special someone than a Friday evening?

Is there any better place than OS to find that lady of my dream?

Not likely...

So, in order to close in on what true love means for myself, I decided to look into one aspect of home plate umpire tendencies.

Home Plate Ump   Strike %   Games  Total Pitches  Strikes   K/BB
----------------------------------------------------------------
Patrick Johnson      63.8     15       4437        2830     2.43
Dutch Nelson         63.2     13       3574        2260     2.67
Earl Hendricks       63.2     19       5366        3392     2.28
Freddie Ferguson     63.0     17       5036        3173     2.39
Ricky Holliday       62.9     22       6405        4031     2.29
Porter Callahan      62.9     17       5062        3183     2.55
Woody Keller         62.7     14       4113        2579     2.33
Mike Winters         62.7     16       4777        2995     2.00
CB. Hernandez        62.5     17       4958        3098     1.77
Joe MacDonald        62.5     14       4056        2533     1.94
James Gibson         62.4     22       6715        4191     2.30
Eric Summersgill     62.2     17       4847        3014     2.55
Jeff Carlson         62.1     20       6038        3751     2.04
Tom Reyburn          62.0     21       6082        3770     2.04
Marvin Knight        62.0     19       5860        3632     1.86
Earl Cooper          61.9     17       5130        3178     1.78
Clyde Washington     61.9     13       3902        2417     2.69
Wally Hughes         61.8     17       4923        3041     2.06
Kerwin Thompson      61.6     18       5140        3166     1.88
Rusty Valentine      61.0     15       4423        2696     1.56

There haven't been that many games called by each umpire, but it does look like the differences exist already. Needless to say, it's not just the umpires that affect those numbers... pitchers do probably have stronger impacts on these stats. But at the same time, there are a lot of pitches recorded already.

Does this match your experiences with home plate umpires? Does Patrick Johnson call more strikes than Rusty Valentine?

Most importantly, where's my love?