Stats-Based CPU Sliders [MLB 16 Version]

[nomo17k]

Re: Stats-Based CPU Sliders [MLB 16 Version]

My "theory" was that "watching" a game was different then SIMing a game using a different engine but that is probably from long past experience (I actually golfed on a green screen in the olden days).

Legacy Madden SIMed games were not actually played games but some random algorithm to create stats.

To be honest I was alarmed that the stats I was getting after 9 MoM games were not as bad as I imagined. They SEEMED really low but it's still Middle April so . . . . . .

One thing I noticed was that stealing in rain looked like the baserunners were running in mud - they seemed a LOT slower. More errors and BB also but it was only one game.

I'm afraid most sports games are like that, having two different engines, one for realistic (quasi)physics-based engine to fully simulate the "process" and simulator which really only simulates results (and not the "process" itself). People just want to quickly do away with part of the games that they are not involved, so to speed things up the simulator is used to "pad" realistic results without actually fully playing out the process.

I actually prefer to have an option of playing everything out with the full gameplay engine, and managed to do this last year with MLB 15:

2430 CPU vs. CPU games


When you compare full-season CPU vs. CPU results against what you typically get with simmmed seasons, you definitely notice the differences. But to the game's credit, it's not THAT wildly different.

[Christian9]

Re: Stats-Based CPU Sliders [MLB 16 Version]

I was messing around with this, and what I am doing is starting the game and Fast Forwarding to the end. Found that pitch counts while looking at Pitcher Analysis is WAY off. So then tried simming a half inning at a time, but still same issue. The pitch count when looking at Game Log looks right, but not sure what I’m doing wrong to get such conflicting results. Any help would be appreciated.

[Christian9]

Re: Stats-Based CPU Sliders [MLB 16 Version]

No one else has experienced this?

[tgreer]

Re: Stats-Based CPU Sliders [MLB 16 Version]

Here is where I stand after 15: Obviously these numbers reflect facing the ACES pitching since I started with Day 1 in Season

Only changes from pure Default:
Contact+1
Solid Hits +1
SB Ability +1

SHOW MLB
Runs: 96 3.20 4.18
Hits: 232 7.73 8.64
Errors: 12 0.40
2B: 35 1.17 1.69
3B: 8 0.27 0.18
HR: 29 0.97 0.96
GIDP: 36 1.20 0.76
SB: 9 0.30 0.54
CS: 8 0.27 0.21
BB: 74 2.47 2.92
K: 243 8.10 7.67
AVG: 232-997 .233 .253

Offense ugly but could be due to facing #1's...not enough SB attempts or success rate but its early. Nomo do you see anything I should change or want me to pump out 75 with these settings? On to game #16 now

[Grinder12000]

Re: Stats-Based CPU Sliders [MLB 16 Version]

No one else has experienced this?

I did that last year and found Pitch Counts were meaningless fast forwarding. As Nomo said - must be a different engine used.

[Christian9]

Re: Stats-Based CPU Sliders [MLB 16 Version]

I did that last year and found Pitch Counts were meaningless fast forwarding. As Nomo said - must be a different engine used.

Thanks for the response.

I see what you're saying. Just odd that they're not matching up. I agree could be different engine generating different results, but this must be a glitch/bug.

[Grinder12000]

Re: Stats-Based CPU Sliders [MLB 16 Version]

I see what you're saying. Just odd that they're not matching up. I agree could be different engine generating different results, but this must be a glitch/bug.

My gut feeling is that this is a known "glitch" which is very low on the TO DO lists. Too bad.

[nomo17k]

Re: Stats-Based CPU Sliders [MLB 16 Version]

I was messing around with this, and what I am doing is starting the game and Fast Forwarding to the end. Found that pitch counts while looking at Pitcher Analysis is WAY off. So then tried simming a half inning at a time, but still same issue. The pitch count when looking at Game Log looks right, but not sure what I’m doing wrong to get such conflicting results. Any help would be appreciated.

I did that last year and found Pitch Counts were meaningless fast forwarding. As Nomo said - must be a different engine used.

Thanks for the response.

I see what you're saying. Just odd that they're not matching up. I agree could be different engine generating different results, but this must be a glitch/bug.

My gut feeling is that this is a known "glitch" which is very low on the TO DO lists. Too bad.

I have indeed been seeing the same thing with pitch count, but I think that's not really a glitch but the limitation of how the simulation engine plays.

This is mostly my (educated) guessing, but what's happening is that, given that the simulation engine only simulates the result of an at-bat, what shows up in Pitcher Analysis is the pitch result of an at-bat and nothing else; it simply cannot show anything else since the simulation engine does not really simulate other pitches.

Yet the overall pitch count must be realistic, since otherwise the game does not take account for pitcher energy loss due to pitch count. So, in simulated portion of at-bats, I think the game is actually adding appropriate amount of pitches (or generating count), artificially to take account the average of 3 - 4 pitches per at-bat. That probably is why the overall pitch count from the game log seems okay, whereas Pitcher Analysis you only see the last pitch of each at-bat.

[nomo17k]

Re: Stats-Based CPU Sliders [MLB 16 Version]

Here is where I stand after 15: Obviously these numbers reflect facing the ACES pitching since I started with Day 1 in Season

Only changes from pure Default:
Contact+1
Solid Hits +1
SB Ability +1

SHOW MLB
Runs: 96 3.20 4.18
Hits: 232 7.73 8.64
Errors: 12 0.40
2B: 35 1.17 1.69
3B: 8 0.27 0.18
HR: 29 0.97 0.96
GIDP: 36 1.20 0.76
SB: 9 0.30 0.54
CS: 8 0.27 0.21
BB: 74 2.47 2.92
K: 243 8.10 7.67
AVG: 232-997 .233 .253

Offense ugly but could be due to facing #1's...not enough SB attempts or success rate but its early. Nomo do you see anything I should change or want me to pump out 75 with these settings? On to game #16 now

If you are largely satisfied with how the gameplay "feels", then you might want to stay with the setting for a while longer. I'd expect offense to go up a bit with Contact +1 and Solid Hit +1 as others have been reporting.

[Grinder12000]

Re: Stats-Based CPU Sliders [MLB 16 Version]

I posted this warning in it's own thread but since this is the one I follow

The IN GAME Team Stats such as BABIP as NOT accurate and are only an average of the averages. They add up all the BABIP for players and divide by 12 (which is ludicrous).

it's like that for all averages (such as Team ERA ??)

[nomo17k]

Re: Stats-Based CPU Sliders [MLB 16 Version]

The initial 75-game set with default sliders (+ max pitch speed) is complete. The summary of what I will be trying to improve with slider adjustments follows (target MLB number in parentheses):

• BA: .233 (.253)
  • The batting average is a bit low most likely due to BABIP being on the low side.
  
  
• BABIP: .274 (.299)
  • BABIP is low most likely due to relative lack of line drives. Hit type fractions (fly balls, line drives, ground balls) are 42.1%/18.4%/39.6% (34.1%/21.0%/45.0%), so the game currently is producing way more fly balls which tends to bring BABIP lower.
  
  
• BB%: 8.39% (7.72%)
  • Walks per plate appearance might be slightly higher than in MLB. This might be adjusted by increasing early-count strike %, which is slightly low, with the first pitch strike % being 58.1% (60.7%) and the early count strike % being 60.9% (62.1%).
  
  
• HR/H: 13.5% (11.1%)
  • HRs are slightly inflated compared to other XBHs, with 2B/H 18.4% (19.6%) and 3B/H 1.53% (2.08%). Most HRs are fly balls, so this might be adjusted by decreasing the fly ball fraction which currently is inflated. Overall XBH fractions seem pretty decent, though.
  
  
• SB%: 49.0% (71.5%)
  • The steal success rate is substantially down, partly due to the Pitch Speed sliders being maxed out. Typically, matching Pitch Speed and BR Steal Ability slider values made SB% respectable in the past, but with the more than 20% gap not sure maxing out BR Steal Ability brings up SB% enough. Will see...
  
  
• GIDPs per BIP: 3.85% (2.97%)
  • GIDPs per balls in play seems slightly inflated, despite ground ball fraction in the game is actually lower compared to MLB. It's always been one of harder things to correct, but with one GIDP per game, it doesn't nearly as bothering as it has in the past.
  
  
• IFE per balls in play (that are outs): 0.75% (1.38%)
  • Infield fielding errors per balls in play (which resulted in outs) seems down.
  
  
• OTE per balls in play (that are outs): 0.04% (0.14%)
  • Outfield throwing errors per balls in play (which resulted in outs) seems down. Long throws are way more accurate this year, and as such they are not resulting in errors as much.
  
  
  

[nomo17k]

Re: Stats-Based CPU Sliders [MLB 16 Version]

Based on the initial set, I'll be trying these adjustments:

Foul Frequency: 4
Solid Hit: 8
Pitcher Control: 4
Strike Frequency: 7
Fastball PS: 10
Offspeed PS: 10
Fielding Errors In: 8
Throwing Errors Out: 10
BR Steal Ability: 10

Those sliders not listed are all at 5 (default).

Note that I've made bigger adjustments than what I feel might be needed, intending to overshoot the target slider value and its effect. It is so much faster to be able to say "with Solid Hit 5 the line drive fraction is lower than what I like it to be but with Solid Hit 7 the line drive fraction is higher than what I want... hence the right value is between 5 - 7 which means I should probably use 6". If I only make a small adjustment (e.g., by increasing Solid Hit to 6) and don't "bracket" the desired effect (e.g., not seeing LD fraction increase as much as I need), then I have to keep making small adjustments and that will take much longer.

So these values are still for testing, not intended to be used for real.

[DoubleTxL]

Re: Stats-Based CPU Sliders [MLB 16 Version]

What difficulty are these sliders used with?

[nomo17k]

Re: Stats-Based CPU Sliders [MLB 16 Version]

What difficulty are these sliders used with?

Difficulty settings do not matter in CPU vs. CPU games. I'm just setting all to All-star though.

[tgreer]

Re: Stats-Based CPU Sliders [MLB 16 Version]

Here is where I stand after 15: Obviously these numbers reflect facing the ACES pitching since I started with Day 1 in Season

Only changes from pure Default:
Contact+1
Solid Hits +1
SB Ability +1

SHOW MLB
Runs: 96 3.20 4.18
Hits: 232 7.73 8.64
Errors: 12 0.40
2B: 35 1.17 1.69
3B: 8 0.27 0.18
HR: 29 0.97 0.96
GIDP: 36 1.20 0.76
SB: 9 0.30 0.54
CS: 8 0.27 0.21
BB: 74 2.47 2.92
K: 243 8.10 7.67
AVG: 232-997 .233 .253

Offense ugly but could be due to facing #1's...not enough SB attempts or success rate but its early. Nomo do you see anything I should change or want me to pump out 75 with these settings? On to game #16 now

Things are definitely looking better through 30 games offensively as we are seeing some of the #2 and #3 starters. Also SB attempts picking up to real life pace (success rate still not there) but moving closer. Hopefully Runs and Hits continue upwards as we hit back end of rotation guys.

Contact+1
Solid Hits+1
SB Ability +1

Thru 30
SHOW MLB
R: 205 3.42-4.18
H: 493 8.22-8.64
E: 28 0.47
2B: 93 1.55-1.69
3B: 11 0.18-0.18
HR: 59 0.98-0.96 *Extra Base Hits Looking Good*
DP: 58 0.97-0.76
SB: 27 0.45-0.54
CS: 20 0.33-0.21
BB: 160 2.67-2.92
K: 464 7.73-7.67
AVG: 493-2033 .242-.253