Stats-Based CPU Sliders [MLB 12 Version]

[ralphieboy11]

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

I agree on the choppers around the plate. I was looking for that as well.

I just got another game under my belt with solid hits at 3. Pirates beat the Phillies, 4-1. McDonald over Hamels. It was nice to see McDonald(3rd starter) pitch well while Hamels struggled. McDonald pitched 7 innings giving only 1 run on a Rollins homer.

I saw one infield hit on a chopper around the plate. I think 3 is probably the cut off for solid hits. Anything lower and you will probably see too many of those. To my eye these last 2 games haven't looked that bad on the choppers around the plate with it at 3. I don't mind seeing a few, as long as I'm seeing more hits sort of "bleed" through the infield as opposed to being "smoked".

[nomo17k]

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

I agree on the choppers around the plate. I was looking for that as well.

I just got another game under my belt with solid hits at 3. Pirates beat the Phillies, 4-1. McDonald over Hamels. It was nice to see McDonald(3rd starter) pitch well while Hamels struggled. McDonald pitched 7 innings giving only 1 run on a Rollins homer.

I saw one infield hit on a chopper around the plate. I think 3 is probably the cut off for solid hits. Anything lower and you will probably see too many of those. To my eye these last 2 games haven't looked that bad on the choppers around the plate with it at 3. I don't mind seeing a few, as long as I'm seeing more hits sort of "bleed" through the infield as opposed to being "smoked".

Right... I think Steve also commented how Solid Hit 0 - 1 may have produced too many choppers that stayed around the plate. So 3 may be a good choice. I'll keep watching how fly ball / line drive / ground ball ratio changes once I begin with SH = 3.

If you are on PS3, you may want to use Replay Vault to check out how those hits actually look like. I don't really trust the game's identification of those types, and sometimes when you look at replays, you go "...was that a fly ball???" It's a quick way to check if you are running the game in the background.

How I wish there's replays on Vita!! Otherwise game looks very good so far.

[steviegolfballs]

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

Nomo, I am glad you are seeing offense.....my offense is way down through the first 20 games so this gives me patience and hope!

[nomo17k]

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

Nomo, I am glad you are seeing offense.....my offense is way down through the first 20 games so this gives me patience and hope!

With what Solid Hit setting? My recent games are mostly by 5th starters, so that's part of the reason. With those mediocre pitchers, there has been no shortage of offense! Quite a few > 10 hits games I'm seeing.

[steviegolfballs]

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

My hits aren't too far off but my HRS and doubles are quite low w hich leads to fewer runs.solid hits on 4, timing 6
Other than pickoffs at 0 and mgr. Hook at 6 my sliders are the same as yours.

[nomo17k]

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

Only 20 games in but will be trying Solid Hit 3 instead of 4. Wanna see how hit variety might increase...

As you see, there's been plenty of offense with Solid Hit 4 below.


Mar 31 test (Starter Stamina 10) 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: 3
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]

If anyone wants more offense just play games with the Astros involved. Boy, does their pitching stink for me so far! Using varying sliders that have never much deviated from Nomo's here, they just completed a 3 game home series with the Rockies.

9.64 ERA and .382 batting average allowed. Ouch.

[seanjeezy]

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

Alright Nomo, a little off-topic, but I have some results for the franchise sim sliders:

AVG w/ RISP is ridiculously low still, Texas had the highest average - .260, the rest were in the .220-.240 range

Last year there were 18 pitchers who threw at least 220 innings, 10 over 230, 2 over 240, and 1 over 250. with SP stamina at 6, I had 12 over 220, and only 1 over 230 (Sabathia 236.2).

16 starters last year had a sub 3.00 ERA, in my sim it was 13, 11 from the national league (8-8 IRL)

Jeff Samardzija appeared in a league high 90 games, and only two relievers had over 90 IP (exact same # as last year), this is with reliever stamina and manager hook at 4

What I take from this I could probably raise SP stamina 1 more click, and I did:

The maximum number of games pitched was 80, 4 relievers pitched over 90 innings, but one was a long reliever (Jeff Neimann)

18 guys pitched over 220 innings, 8 over 230 (5-3 AL), 0 over 240, etc.

2 guys had 8 complete games, but there were 60(!)guys with 4 or more, compared to 14 last year

18 guys with sub 3.00 ERAs (13-5 NL)

29 guys with at least 15 wins, compared to 20 from last year

21 guys with at least 40 saves compared to 8 from last year

The clutch thing is really a problem - "Captain Clutch" David Freese only batted .247 with RISP, and thats with a 99 clutch rating, while 2B Ramone Russel (LOL) bats .500 with a 33 clutch rating

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

Team Stats:

Lowest team ERA - 3.02 IRL - 3.02
Highest team ERA - 4.36 IRL - 4.92
Most complete games - 27 IRL - 27
Least complete games - 9 IRL - 8
Most blown saves - 29
Least blown saves - 11
Most hits allowed - 1571 IRL - 1568
Least hits allowed - 1330 IRL - 1260
Most runs allowed - 772 IRL - 860
Least runs allowed - 528 IRL - 529
Most BB allowed - 624 IRL - 580
Least BB allowed - 451 IRL - 404
Most K's - 1332 IRL - 1332
Least K's - 949 IRL - 940

Most holds - 66 IRL - 99
Least holds - 32 IRL - 39

The mean number of holds last year was 72, in my sim it was 50: Lots of late game lead changes, or the insane amount of complete games?

[nomo17k]

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

Alright Nomo, a little off-topic, but I have some results for the franchise sim sliders:

AVG w/ RISP is ridiculously low still, Texas had the highest average - .260, the rest were in the .220-.240 range

Last year there were 18 pitchers who threw at least 220 innings, 10 over 230, 2 over 240, and 1 over 250. with SP stamina at 6, I had 12 over 220, and only 1 over 230 (Sabathia 236.2).

16 starters last year had a sub 3.00 ERA, in my sim it was 13, 11 from the national league (8-8 IRL)

Jeff Samardzija appeared in a league high 90 games, and only two relievers had over 90 IP (exact same # as last year), this is with reliever stamina and manager hook at 4

What I take from this I could probably raise SP stamina 1 more click, and I did:

The maximum number of games pitched was 80, 4 relievers pitched over 90 innings, but one was a long reliever (Jeff Neimann)

18 guys pitched over 220 innings, 8 over 230 (5-3 AL), 0 over 240, etc.

2 guys had 8 complete games, but there were 60(!)guys with 4 or more, compared to 14 last year

18 guys with sub 3.00 ERAs (13-5 NL)

29 guys with at least 15 wins, compared to 20 from last year

21 guys with at least 40 saves compared to 8 from last year

The clutch thing is really a problem - "Captain Clutch" David Freese only batted .247 with RISP, and thats with a 99 clutch rating, while 2B Ramone Russel (LOL) bats .500 with a 33 clutch rating

Thanks for the numbers. Actually this is also a very nice way of adjusting Stamina & Manager Hook sliders.

In terms of IP, looks like Stamina = 7 makes sense, but that increased CGs a bit much... I actually notice this in my CPU vs. CPU games as well; I think AI managers tend to keep starters in the game if they are pitching shutouts or have allowed few runs, even if their pitch count is rather high. If you lower Manager Hook we might keep this in check, but that of course has effects on bullpen management as well.

I think the BA/RISP is a bug in the sim engine, but it doesn't actually affect the runs... meaning, the batters are actually hitting as well in RISP situations. The league stats are fine in terms runs, so I think somewhere in their code they are just miscalculating BA in RISP... For example, if they are counting *walks* in RISP as at-bats by mistake... think a typical .260 hitter in 500 at bats with 60 walks in a year. BA is computed like

130 Hits / 500 AB = .260.

but if by mistake the game is counting 60 walks as at-bats:

130 Hits / 560 AB = .232

That's about the kind of BA/RISP we are seeing, isn't it? Something along the line is what I suspect is happening....

[nomo17k]

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

...

Most holds - 66 IRL - 99
Least holds - 32 IRL - 39

The mean number of holds last year was 72, in my sim it was 50: Lots of late game lead changes, or the insane amount of complete games?

Yeah, I think (good) relievers in the game probably isn't as effective as IRL, and this probably shows up in your numbers as well.

Looking at the inning-by-inning scores that I did earlier, starters are much more effective after the first inning, which presumably is because they have time to build up their confidence after 1st inning. Most relievers don't pitch that long, so they often pitch with only half-assed confidence. That might make relievers even less effective in general.

But your numbers are from simmed games, right? I actually don't know there is a concept of pitcher/pitch confidence within the sim engine. So for simmed game, it might not be confidence that's the culprit.

Interesting.

[seanjeezy]

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

Yeah, I think (good) relievers in the game probably isn't as effective as IRL, and this probably shows up in your numbers as well.

Looking at the inning-by-inning scores that I did earlier, starters are much more effective after the first inning, which presumably is because they have time to build up their confidence after 1st inning. Most relievers don't pitch that long, so they often pitch with only half-assed confidence. That might make relievers even less effective in general.

But your numbers are from simmed games, right? I actually don't know there is a concept of pitcher/pitch confidence within the sim engine. So for simmed game, it might not be confidence that's the culprit.

Interesting.

There's also the problem where the CPU only uses their closer in save situations, whereas the closer is used in any tight situation in the 9th and sometimes the 8th inning in real life, Kimbrel, Valverde, Marmol, Axford, Storen, etc. - All of these guys had over 70 APP and 70 IP last year, in the Show they would be lucky to get 55 appearances.

Honestly I don't even start a serious franchise until OSFM v4 or whatever, the one that has MLB ratings edits, there are a lot of way overrated/underrated guys that need serious edits

[nomo17k]

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

There's also the problem where the CPU only uses their closer in save situations, whereas the closer is used in any tight situation in the 9th and sometimes the 8th inning in real life, Kimbrel, Valverde, Marmol, Axford, Storen, etc. - All of these guys had over 70 APP and 70 IP last year, in the Show they would be lucky to get 55 appearances.

Honestly I don't even start a serious franchise until OSFM v4 or whatever, the one that has MLB ratings edits, there are a lot of way overrated/underrated guys that need serious edits

Yeah, I think a lot of managerial decisions are still kinda cookie cutter and questionable. Something that has a lot of room for improvement.

[Halara]

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

Yeah, I think (good) relievers in the game probably isn't as effective as IRL, and this probably shows up in your numbers as well.

Looking at the inning-by-inning scores that I did earlier, starters are much more effective after the first inning, which presumably is because they have time to build up their confidence after 1st inning. Most relievers don't pitch that long, so they often pitch with only half-assed confidence. That might make relievers even less effective in general.

But your numbers are from simmed games, right? I actually don't know there is a concept of pitcher/pitch confidence within the sim engine. So for simmed game, it might not be confidence that's the culprit.

Interesting.

To start off, thanks for your work with this Nomo. I'm a manager mode player who's been lurking out here for years. I've been loving your sliders this year! Thanks for all the time you put in to this!

In regards to relievers confidence. I've noticed that if a reliever has been playing well and is coming off a good game, he will enter with a higher level of confidence. When they've come off a poor game or when they play in their first game of the season they have that mid level confidence. I think they might ever start with below mid level if they've been particularly cruddy in the last few games but I'm not 100% on that last bit.

- Hal

[nomo17k]

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

To start off, thanks for your work with this Nomo. I'm a manager mode player who's been lurking out here for years. I've been loving your sliders this year! Thanks for all the time you put in to this!

In regards to relievers confidence. I've noticed that if a reliever has been playing well and is coming off a good game, he will enter with a higher level of confidence. When they've come off a poor game or when they play in their first game of the season they have that mid level confidence. I think they might ever start with below mid level if they've been particularly cruddy in the last few games but I'm not 100% on that last bit.

- Hal

See my theory about MoM players being like earthworms seems correct... it just hasn't rained so they haven't come out on street to get noticed by SCEA. :P


Anyways, yeah I think you are correct about confidence carrying over adjacent games... don't know how the logic exactly works, but they do fluctuate game to game.

For relievers who only pitch an inning at most in a game, perhaps it takes too long for them to build up enough confidence to be as effective as they can be. Maybe in that sense confidence system might still be too strong.

[DarkSith777]

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

I humbly approach you with respect great one.......Would you recommend keeping pitcher confidence off? Also how much of a difference is there between the stats you accumulated with the default sliders and the set you are using now? Fielder reaction.....doesn't seem that lowering this causes the infielders to react too slowly? Will you have a definite set by opening day? I don't have the time to play around with sliders on my own anymore. Thank you great one!