Stats-Based CPU Sliders [MLB 18 Version Season-Mode Memorial Edition]

[NolanRyansSnowmonkey]

Re: Stats-Based CPU Sliders [MLB 18 Version Season-Mode Memorial Edition]

By the way, I don't remember it being discussed, but did you notice much difference after patch 1.08? I know a lot of slider creators are still tweaking after the patch, but it didn't seem to affect CPU vs CPU it seems?

[My993C2]

Re: Stats-Based CPU Sliders [MLB 18 Version Season-Mode Memorial Edition]

I'm going to roll with the OSFM v 2.0 but likely a set of Nomo's sliders that has more offense (I like the looks of the 4/18 set). I refuse to sim games to the end but hate CPU vs CPU extra inning games that go on and on. I had one a few years ago that went nearly 18 innings before it ended. LOL.

Last week I had a Fast Play non-QC CPU vs CPU Franchise game go about 18 innings and I was annoyed that it tied up my PS4 for at least an hour more than it should have. But then I had to remind myself that I choose to play CPU vs CPU games for my Franchise in all the games my team does not play. No one twisted my arm and forced me to play this way. LOL

However last month I played a full Broadcast non-QC MoM Franchise game Rockies at San Diego that went 16 innings and the worst part was that my team the Rockies not only lost the game (which took me close to 4 real world hours to complete), but going that long really did a number on my bullpen which affect the team for at least a week after this marathon game. Only savings grace is that there are extra days off built into the 2018 thanks to the collective bargaining of the MLBPA.

[BigOrangeVol4Life]

Re: Stats-Based CPU Sliders [MLB 18 Version Season-Mode Memorial Edition]

Last week I had a Fast Play non-QC CPU vs CPU Franchise game go about 18 innings and I was annoyed that it tied up my PS4 for at least an hour more than it should have. But then I had to remind myself that I choose to play CPU vs CPU games for my Franchise in all the games my team does not play. No one twisted my arm and forced me to play this way. LOL

However last month I played a full Broadcast non-QC MoM Franchise game Rockies at San Diego that went 16 innings and the worst part was that my team the Rockies not only lost the game (which took me close to 4 real world hours to complete), but going that long really did a number on my bullpen which affect the team for at least a week after this marathon game. Only savings grace is that there are extra days off built into the 2018 thanks to the collective bargaining of the MLBPA.

I've found with Nomo's settings in the past that feature more offense that there are still games where offenses are shut down and shutouts occur, but a team will usually break a tie early in the extra innings and games rarely go beyond 11 or 12 innings.

I love watching CPU vs CPU games using Nomo's sliders. I give the computer full control with trades, cuts, signings, etc. The game just comes to life with these.

While waiting on the rosters and patches to finalize, I haven't began a franchise yet with this year's game. Does anyone know if Spring Training games end in ties if they're still tied at the end of the 9th inning? In real life this happens, but the teams don't receive a tie in the standings, and it's as if they never played the game. The players' stats however are still recorded though. In past versions of this game, Spring Training games go to extra innings just as they do in the regular season and postseason. This is a modest change I'd like to see made by Sony to make the game further mirror the real game.

[NolanRyansSnowmonkey]

Re: Stats-Based CPU Sliders [MLB 18 Version Season-Mode Memorial Edition]

I've been too busy to even start my franchise yet. I plan on it this week. I can't wait to actually officially use these sliders.

[NolanRyansSnowmonkey]

Re: Stats-Based CPU Sliders [MLB 18 Version Season-Mode Memorial Edition]

Nomo, not sure if you have ever tried this in testing Human vs CPU, but do you know if switching teams every half-inning has any effect on sliders/stats?

For example, in testing, if i were to hit with the visiting team, then switch to the home team to hit, so that I am hitting for both teams against CPU pitching for the full game, would I get an accurate output to what my sliders are set at for human hitting and CPU pitching for both teams?

I know when I select my team before the game, I also select my Human Profile that carries my difficulty level etc. All other teams are set to a CPU profile for roster management, etc (30 team control- i have CPU also on the same difficulty level i am on, not sure if it matters or not). It doesn't specify the profile you are using in the in-game Switch Teams screen.
I'm assuming that Human Profile and Human sliders carry over with me when I switch teams, but sometimes I wonder if I'm wasting time.

I just started this, so I don't have enough stats to see a big enough picture.

Not sure if you've ever explored this, but I figured if anyone had an idea, you would be the one to ask...

Thanks

[nomo17k]

Re: Stats-Based CPU Sliders [MLB 18 Version Season-Mode Memorial Edition]

Nomo, not sure if you have ever tried this in testing Human vs CPU, but do you know if switching teams every half-inning has any effect on sliders/stats?

For example, in testing, if i were to hit with the visiting team, then switch to the home team to hit, so that I am hitting for both teams against CPU pitching for the full game, would I get an accurate output to what my sliders are set at for human hitting and CPU pitching for both teams?

I know when I select my team before the game, I also select my Human Profile that carries my difficulty level etc. All other teams are set to a CPU profile for roster management, etc (30 team control- i have CPU also on the same difficulty level i am on, not sure if it matters or not). It doesn't specify the profile you are using in the in-game Switch Teams screen.
I'm assuming that Human Profile and Human sliders carry over with me when I switch teams, but sometimes I wonder if I'm wasting time.

...

I think sliders and (difficulty) settings profile work independently. That means HUM/CPU sliders should work the way it says, i.e.,

In fact, when I collect stats from my own playing and focus on adjusting just hitting, for example, I'd switch teams to only hit for the game. I have never seen any issues and sliders "switch" with which team that I select for the inning. No issue there.

I'm less certain about how the (difficulty) profile carries over when you switch teams in the middle of a game. However, I think the profile is actually associated to the "controller" and not the team when each game starts, so I'd actually think the profile setting switches teams accordingly when you make the same change. If you have two controllers, you could easily test this.

[ericjwm]

Re: Stats-Based CPU Sliders [MLB 18 Version Season-Mode Memorial Edition]

I think sliders and (difficulty) settings profile work independently. That means HUM/CPU sliders should work the way it says, i.e.,

In fact, when I collect stats from my own playing and focus on adjusting just hitting, for example, I'd switch teams to only hit for the game. I have never seen any issues and sliders "switch" with which team that I select for the inning. No issue there.

I'm less certain about how the (difficulty) profile carries over when you switch teams in the middle of a game. However, I think the profile is actually associated to the "controller" and not the team when each game starts, so I'd actually think the profile setting switches teams accordingly when you make the same change. If you have two controllers, you could easily test this.

You have a good point! I have one controller.. yet if you set it for both wouldn't you think it would identify the settings regardless? It's looked uneven at times but I've never noticed that drastic a difference.

[ericjwm]

Re: Stats-Based CPU Sliders [MLB 18 Version Season-Mode Memorial Edition]

All I ask. Run 6 games with Max Scherzer starting and if he never gets 12+ strikeouts in one game...

[nomo17k]

Re: Stats-Based CPU Sliders [MLB 18 Version Season-Mode Memorial Edition]

All I ask. Run 6 games with Max Scherzer starting and if he never gets 12+ strikeouts in one game... it doesn't work.

This still isn't quite a sufficiently solid approach though. A quick calculation to see if what you are proposing tells something you'd like to show goes like this...

In 2017, Scherzer pitched 31 games, 3 of which were 12+-strikeout (that is 12 or more strikeouts) games. The probably for him to get 12+-strikeouts in a game is then p = 3./31 = 0.097 = 9.7%.

What we want to know is the probability of Scherzer pitching at least one 12+-strikeout game after n = 6 games.

A standard application of binomial distribution (the math formula is available all over the web) gives this probability to be about 46%.

So if I run an experiment using the real-life Scherzer from 2017 according to your design, it's basically a coin flip as to whether I see him having at least one 12+-strikeout game in 6 games.

Does that provide any convincing evidence to decide whether the strikeout numbers are correct or not? I say no... since a monkey flipping a coin would get this right or wrong just as frequently as a human being making a purely random guess.

[BigOrangeVol4Life]

Re: Stats-Based CPU Sliders [MLB 18 Version Season-Mode Memorial Edition]

Nomo, I saw in another thread where some were saying the default setting of "5" was too low for injuries over a course of time. A lot of posters said they had the number bumped up to at least "8". Are the frequency and number of injuries in the game compared to what goes on in MLB something you look at also? Have you read or heard this?

[nomo17k]

Re: Stats-Based CPU Sliders [MLB 18 Version Season-Mode Memorial Edition]

Nomo, I saw in another thread where some were saying the default setting of "5" was too low for injuries over a course of time. A lot of posters said they had the number bumped up to at least "8". Are the frequency and number of injuries in the game compared to what goes on in MLB something you look at also? Have you read or heard this?

This is a tricky question, since I do not have data on injuries in MLB... so I don't have any baseline to compare with to think they happen too often or not.

At one point I was interested in looking at the data, but it was a little harder to gather reliable data in easily accessible form, so I haven't really dug in myself. The tedium is in gathering good data to compare with, and if that's available I would have tried to adjust the Injuries slider.

But a couple years ago I believe someone (Mike Lowe? I don't exactly remember who) was trying to do this, though how rigorous things were done and such, I do not remember. You might find something relevant if you search the forum for the info.

[NolanRyansSnowmonkey]

Re: Stats-Based CPU Sliders [MLB 18 Version Season-Mode Memorial Edition]

Im curious to know what you have seen as far as how wind affects statistics? How would you go about setting the wind slider? Or is it more of something that needs to be adjusted visually?

I know wind can effect home run numbers, but, if wind was at 10 for example, it seems that the wind (as a league-wide average) would help a lot of balls leave the park, but also knock just as many balls down the other way. And likewise if it were at 5 or 0. Maybe i am over simplifying it by not accounting for other wind directions. But do you see wind affecting stats? Or mainly just how game feels and plays?

[KBLover]

Re: Stats-Based CPU Sliders [MLB 18 Version Season-Mode Memorial Edition]

I know wind can effect home run numbers, but, if wind was at 10 for example, it seems that the wind (as a league-wide average) would help a lot of balls leave the park, but also knock just as many balls down the other way. And likewise if it were at 5 or 0. Maybe i am over simplifying it by not accounting for other wind directions. But do you see wind affecting stats? Or mainly just how game feels and plays?

I think statistically, it might even out over time...however, there could be more benefit for mid power guys getting balls out.

Big boys like Gallo, Sano, Judge...if they connect, it will carry out most of the time.

Lighter guys like Gordon, Hamilton, Pedroia - they probably aren't going out either way.

The 55-65 guys? I think that's where it can matter for if it blows out hard enough.

So I keep it lower to minimize that effect, though perhaps I should test higher once I'm sure things are settling in nicely.

[NolanRyansSnowmonkey]

Re: Stats-Based CPU Sliders [MLB 18 Version Season-Mode Memorial Edition]

I dont usually like to bring things up like this without a pretty decent sample size, but its pretty glaring.

Like we talked about before, i am playing some franchise games switching teams every half inning, so i am always hitting.

I have played 5 games (so 10 games hitting).

All the offensive stats look really good overall.

I was just playing my 6th game and i jumped out to a 5-0 lead with the visiting team, with 11 hits to 3 hits. I immediately thought man it seems like all these games seem to be like this, with one team hitting and the other not so much. So i stopped and went to my spreadsheet.


The visiting team has won all 5 games. Here are the final scores, with the hits:

Game 1
9 runs /13 hits
1 runs /6 hits

Game 2
2 runs/12 hits
1 runs/3 hits

Game 3
7 runs/10 hits
2 runs /6 hits

Game 4
4 runs/8 hits
1 runs/4 hits

Game 5
6 runs/13 hits
2 runs /8 hits

Game 6 (through 5.5 innings)
5 runs/11 hits
0 runs/3 hits

At a glance they look fine. And they average out pretty good too. But so far the Visiting team is out hitting the home team but a huge margin.

Visiting Team (not including game 6):
5.6 R/G 11.2 H/G

Home Team (not including game 6):
1.4 R/G. 5.4 H/G

It looks like these numbers will be more drastic after this 6th game is finished. I play the exact same with both teams, with no bunting stealing etc.

Would you keep going with this, since the sample size is small enough? Or is this glaring enough in your opinion to scrap the experiment and go back to playing one team at a time?

[NolanRyansSnowmonkey]

Re: Stats-Based CPU Sliders [MLB 18 Version Season-Mode Memorial Edition]

I think statistically, it might even out over time...however, there could be more benefit for mid power guys getting balls out.

Big boys like Gallo, Sano, Judge...if they connect, it will carry out most of the time.

Lighter guys like Gordon, Hamilton, Pedroia - they probably aren't going out either way.

The 55-65 guys? I think that's where it can matter for if it blows out hard enough.

So I keep it lower to minimize that effect, though perhaps I should test higher once I'm sure things are settling in nicely.

Thank you. I was playing for a while with wind at zero, and HR have been low. So yeah maybe its those middle range guys that need some help from the wind. Almost all of my HR so far have came from those big guys. So what you said makes sense. I appreciate it.