Stats-Based CPU Sliders [MLB 17 Version]

[nomo17k]

Re: Stats-Based CPU Sliders [MLB 17 Version]

I think difficulty does matter because when I change to all star then hall of fame the game plays different

What difficulty u using to edit Ur sliders wI think I'll use that

I think it really shouldn't matter (which difficulty should CPU play on when they against each other, and how do you assign difficulty level to each when putting the game on CPU vs. CPU?). But my setting is usually just on All-star when playing CPU vs. CPU.

When looking at the spreadsheet the sliders in the top left. Does that change throughout the testing? I tried the sliders a few days ago but i wanted to see if they're updated on the spreadsheet?

Thanks

Each spreadsheet "tab" tests different set of sliders, so I create a new tab when I make changes to sliders, so that the status I see on each tab correspond to setting I want to test. For each tab, sliders are fixed.

[tgreer]

Re: Stats-Based CPU Sliders [MLB 17 Version]

What are you going with so far?Just like to see what other guys are using?


Sent from my iPhone using Tapatalk

Contact 6
Power 4
ST Stamina 7
Man Hook 3
FB Pitch Speed 10 (eye effect)
OS Pitch Speed 10 (")
Fld Err Inf 10
Fld Err Of 6
Thr Err Inf 8
Thr Err Of 8
Fld Run Speed 4
Fld Reac 4
IF Arm Strength 4
OF Arm Strength 4
Steal Ability 10
Steal Freq 6
Strike Freq 4
Wind 0
Everything else Default

I like what I am seeing...High scoring/Low scoring games. But I still may roll with Nomos as his are more in depth

[miamikb2001]

Re: Stats-Based CPU Sliders [MLB 17 Version]

Contact 6

Power 4

ST Stamina 7

Man Hook 3

FB Pitch Speed 10 (eye effect)

OS Pitch Speed 10 (")

Fld Err Inf 10

Fld Err Of 6

Thr Err Inf 8

Thr Err Of 8

Fld Run Speed 4

Fld Reac 4

IF Arm Strength 4

OF Arm Strength 4

Steal Ability 10

Steal Freq 6

Strike Freq 4

Wind 0

Everything else Default



I like what I am seeing...High scoring/Low scoring games. But I still may roll with Nomos as his are more in depth

Mine have alot of similar ones.Thanks for sharing


Sent from my iPhone using Tapatalk

[nomo17k]

Re: Stats-Based CPU Sliders [MLB 17 Version]

Does anyone know if human vs cpu is having the same high level of wild pitch frequency? I'm just hoping that it does so maybe they will do something about it and patch it. I also would like Nomo's opinion about this please. It really seems to be playing a big role in the games. I watched one cpu vs cpu game where Wade Davis came on to close the door with a 3 run lead and he gave up some hits then threw 3 wild pitches that eventually scored 4 runs and lost the game. Keep up the good work Nomo the game is still totally worth watching with your stats data sliders. thanks all

Based on my own experience playing HUM vs. CPU (or HUM vs. HUM online), I think the wild pitch issue is common across all modes, which isn't surprising since it's a similar gameplay engine after all.

Unless there already is a decent bug report, I think I want to file it so that it gets to the devs and that it's something we cannot really fix by tweaking the game as is.

[nomo17k]

Re: Stats-Based CPU Sliders [MLB 17 Version]

Just when I thought I was close on getting mine pinned down and ready to start my franchise thinking my numbers were looking good.....I come in here and see your latest test run. Only 2 of my numbers are closer to MLB avg then yours and not by much haha. Yours are looking good with this last set (WP numbers obviously are hard to fix) Anything you are concerned with "eye test" wise? I have 7 more games to run with mine to get to 50 then I am going to start with one of ours! haha

Aside from things that sliders cannot solve that we discusses in the thread, I think the game in general plays good.

Things that tend to fail my eye test so far tends to be those things that we cannot do much anyways... I think a major issue is that CPU makes a lot of errors that are actually scored as single. If you browse the replays at the end of the game, and if the entry says "deflected off XX," then there is a good chance that is a fielding error, but the game actually scores as a hit. That's a little unfortunate.

And I think quite a few hard hit line drives have too high a hang time (which basically means they have too much backspin) . I wouldn't totally mind seeing them once in a while, but I think the game might be producing a bit too many of those (some of which is likely the culprit of monsterous HRs.)

Another thing is that I think starting pitchers being pulled a bit earlier is probably true. I am jacking up starter stamina and lowering Manager Hook way down, but I cannot really manage to have them make more pitches per game than the MLB average (on average, though).

[miamikb2001]

Re: Stats-Based CPU Sliders [MLB 17 Version]

Aside from things that sliders cannot solve that we discusses in the thread, I think the game in general plays good.



Things that tend to fail my eye test so far tends to be those things that we cannot do much anyways... I think a major issue is that CPU makes a lot of errors that are actually scored as single. If you browse the replays at the end of the game, and if the entry says "deflected off XX," then there is a good chance that is a fielding error, but the game actually scores as a hit. That's a little unfortunate.



And I think quite a few hard hit line drives have too high a hang time (which basically means they have too much backspin) . I wouldn't totally mind seeing them once in a while, but I think the game might be producing a bit too many of those (some of which is likely the culprit of monsterous HRs.)



Another thing is that I think starting pitchers being pulled a bit earlier is probably true. I am jacking up starter stamina and lowering Manager Hook way down, but I cannot really manage to have them make more pitches per game than the MLB average (on average, though).

Definately agree lots of errors scored hits,in my games i'd say at least 1 a game sometimes 3 or 4.I think on non QC games starter's are generally pulled a little early but for the most part it's pretty good.On QC for me unplayable unfortunately as i like that mode.I like the different hit variety this year but alot of liners that hang up.


Sent from my iPhone using Tapatalk

[nomo17k]

Re: Stats-Based CPU Sliders [MLB 17 Version]

Definately agree lots of errors scored hits,in my games i'd say at least 1 a game sometimes 3 or 4.I think on non QC games starter's are generally pulled a little early but for the most part it's pretty good.On QC for me unplayable unfortunately as i like that mode.I like the different hit variety this year but alot of liners that hang up.

In non-QC games which all my test games are on, I agree that starter being pulled early isn't as big of an issue on surface. If I look at innings pitched, the average is fairly close to the MLB average.

But by watching some specific early pulls, I can understand how some people are seeing those peculiar early hooks (starter generally pitching well, low pitch count, but pulled anyways) and have the impression that manager hook is kinda quirky. I've seen that a couple times in my own games as well.

I think "on average" (which is an important qualification, because it's still possible that averaging numbers in broad strokes you may be washing away important special cases), the total innings pitched by starters will be slightly on a shorter side, but probably not by a big margin.

I think CPU manager is still not quite smart enough that there are certain decisions that they keep making which isn't realistic. It used to be something like CPU leaving a starter in when the team is leading by a big margin (like 5 runs or so) even if he has a crazy pitch count like 150 pitches. I think that may have been fixed, but there are new issues.

[miamikb2001]

Re: Stats-Based CPU Sliders [MLB 17 Version]

In non-QC games which all my test games are on, I agree that starter being pulled early isn't as big of an issue on surface. If I look at innings pitched, the average is fairly close to the MLB average.



But by watching some specific early pulls, I can understand how some people are seeing those peculiar early hooks (starter generally pitching well, low pitch count, but pulled anyways) and have the impression that manager hook is kinda quirky. I've seen that a couple times in my own games as well.



I think "on average" (which is an important qualification, because it's still possible that averaging numbers in broad strokes you may be washing away important special cases), the total innings pitched by starters will be slightly on a shorter side, but probably not by a big margin.



I think CPU manager is still not quite smart enough that there are certain decisions that they keep making which isn't realistic. It used to be something like CPU leaving a starter in when the team is leading by a big margin (like 5 runs or so) even if he has a crazy pitch count like 150 pitches. I think that may have been fixed, but there are new issues.

I agree the cpu Ai has improved in certain ways,but still has a way to go.I'm optimistic that were close to good cpu Ai maybe a year or two away from really good.Don't get me wrong i think its mostly good enough,i will play the game a ton this year


Sent from my iPhone using Tapatalk

[miamikb2001]

Re: Stats-Based CPU Sliders [MLB 17 Version]

I have seen some instances of the long reliever staying in too long in multiple games in a row,which is one of the reasons i have reduced reliever stamina too try and combat that somewhat.I have also seen the closer removed in situations it would almost never happen,although i have only seen that vary sparingly which is good.


Sent from my iPhone using Tapatalk

[nomo17k]

Re: Stats-Based CPU Sliders [MLB 17 Version]

I agree the cpu Ai has improved in certain ways,but still has a way to go.I'm optimistic that were close to good cpu Ai maybe a year or two away from really good.Don't get me wrong i think its mostly good enough,i will play the game a ton this year

I have seen some instances of the long reliever staying in too long in multiple games in a row,which is one of the reasons i have reduced reliever stamina too try and combat that somewhat.I have also seen the closer removed in situations it would almost never happen,although i have only seen that vary sparingly which is good.

That's one other thing that I think has changed from past years. I haven't seen as many seemingly random pitching changes. On the other hand, I still don't understand how CPU is making sense of situation when it does bullpen management. Doesn't play match up particularly well when it makes substitution, and often it leaves a reliever too long when there is clearly a better alternative in the bullpen...

Actually I remember closers being pulled used to be a fairly common problem a few years ago. That was fixed at some point, but it's either the same issue creeping back in or somehow the new implementation started to allow it to happen again. I've actually seen closers being pulled quite a few times, especially when they get in trouble and pitch counts go high.


I find it very unfortunate the devs don't seem to prioritize in improving CPU managers after quite a few disappointing years of little improvement... I think the baseball gameplay of The Show is something a lot of people rave about, and that is totally great thing since that's what a lot of people expect from a game.

But especially after watching last year's playoff, we cannot underestimate how engaging the chess match between human-like managers can be, but it's critically lacking in the Show.

How Francona/Roberts/Maddon over-relied on their aces/closers in critical games and how Bochy "panicked" when the Cubs started rallying on his relievers in that NLDS game... a lot of human elements there.

How managers manages is as much a part of baseball as players making plays, you know what I mean?

[tgreer]

Re: Stats-Based CPU Sliders [MLB 17 Version]

In watching most of the post game replays, the official scorekeeper of the games is definitely hitter friendly not pitcher. Agree there are too many errors recorded as hits.

Nomo with Arm Strength at 5 are you not seeing too many thrown out at home? I had it at 5 (prob too briefly) and there seemed to be too many. With it at 4 I have loved it, but again I did not have it at 5 for very long

[miamikb2001]

Re: Stats-Based CPU Sliders [MLB 17 Version]

That's one other thing that I think has changed from past years. I haven't seen as many seemingly random pitching changes. On the other hand, I still don't understand how CPU is making sense of situation when it does bullpen management. Doesn't play match up particularly well when it makes substitution, and often it leaves a reliever too long when there is clearly a better alternative in the bullpen...

Actually I remember closers being pulled used to be a fairly common problem a few years ago. That was fixed at some point, but it's either the same issue creeping back in or somehow the new implementation started to allow it to happen again. I've actually seen closers being pulled quite a few times, especially when they get in trouble and pitch counts go high.


I find it very unfortunate the devs don't seem to prioritize in improving CPU managers after quite a few disappointing years of little improvement... I think the baseball gameplay of The Show is something a lot of people rave about, and that is totally great thing since that's what a lot of people expect from a game.

But especially after watching last year's playoff, we cannot underestimate how engaging the chess match between human-like managers can be, but it's critically lacking in the Show.

How Francona/Roberts/Maddon over-relied on their aces/closers in critical games and how Bochy "panicked" when the Cubs started rallying on his relievers in that NLDS game... a lot of human elements there.

How managers manages is as much a part of baseball as players making plays, you know what I mean?

I agree wholeheartdly.It would make the game so much more immersive if manager's had there own seperate style which they would follow Last years playoffs were extremly interesting the way the bullpens were used.


Sent from my iPhone using Tapatalk

[orye74]

Re: Stats-Based CPU Sliders [MLB 17 Version]

Anyone seeing many blow outs? I'm seeing a lot of 2 runs games like 4-2, 6-4, 5-3, 8-6 (10 innings) and 1 11-7 game.

[tHurley2010]

Re: Stats-Based CPU Sliders [MLB 17 Version]

Pardon my sheer stupidity in advance. I'm wanting to start a MoM Franchise mode using your sliders, but am having a hard time finding the sliders you're using to test. Perhaps they're right in front of me, perhaps they're not posted at all, but I'd appreciate it if anyone could point out where I could find these sliders.

Thanks!

[nomo17k]

Re: Stats-Based CPU Sliders [MLB 17 Version]

Pardon my sheer stupidity in advance. I'm wanting to start a MoM Franchise mode using your sliders, but am having a hard time finding the sliders you're using to test. Perhaps they're right in front of me, perhaps they're not posted at all, but I'd appreciate it if anyone could point out where I could find these sliders.

Thanks!

On a sheet, if you put your cursor on the top-left cell named "sliders," a pop-up should show up with slider setting.