Stats-Based CPU Sliders [MLB 14 Version]

[nomo17k]

Re: Stats-Based CPU Sliders [MLB 14 Version]

Hi Nomo. Have you made any updates to these sliders based on your test findings? And also, are these meant to be used with the stock roster file that came with the game out of the box?

The latest slider set was uploaded to the Slider Vault some time ago; not sure if that is the same as what I listed in the first post in this thread, but I'll check the next time I'm on the game. The results were pretty stable so I haven't made any change since.

If you are talking about the set with QC, I haven't gone further than whatever I have in my spreadsheet. I think it will take a major effort to find the slider configuration that yield realistic stats on CPU vs. CPU games. I think there's something about the way QC is generated that tends to put CPU batters more at disadvantage, resulting in less offense on QC games.

[kcsam]

Re: Stats-Based CPU Sliders [MLB 14 Version]

The latest slider set was uploaded to the Slider Vault some time ago; not sure if that is the same as what I listed in the first post in this thread, but I'll check the next time I'm on the game. The results were pretty stable so I haven't made any change since.

If you are talking about the set with QC, I haven't gone further than whatever I have in my spreadsheet. I think it will take a major effort to find the slider configuration that yield realistic stats on CPU vs. CPU games. I think there's something about the way QC is generated that tends to put CPU batters more at disadvantage, resulting in less offense on QC games.

No, I wasn't talking about QC. I was talking about regular played games. Are you using the base rosters with these or something from the roster vault? I'm curious in knowing if your sliders make a difference in what roster file you use.

[nomo17k]

Re: Stats-Based CPU Sliders [MLB 14 Version]

No, I wasn't talking about QC. I was talking about regular played games. Are you using the base rosters with these or something from the roster vault? I'm curious in knowing if your sliders make a difference in what roster file you use.

Sorry I missed the roster part of your question...

Unless the roster maker is doing something really weird that substantially deviates from how SCEA generally rates players, the game should play fine. Something like OSFM rosters should be fine because at the MLB level they don't change ratings from default much.

For testing I use the default SCEA roster that comes with the game.

[BrianU]

Re: Stats-Based CPU Sliders [MLB 14 Version]

I saw there was a patch released a few days ago but did not see anything in the patch notes about 'fixing' the SB% using a higher pitch speed. I know sometimes they fix things without putting it in the notes, is there any improvement to this area since the patch? My gut tells me it's the same but I thought I would ask before I dive in again.

Having to play with a low pitch speed just to get an accurate SB% really ruins some of the fun for me.. It's such a glaring issue for an otherwise simulation game, it really stands out as being an aberration. I know SCEA Brian was posting in the thread earlier about the stealing issues and even fixed a part of it but sadly from the data it clearly still is a big problem. People have done tests to determine that Pitch Speed 10 is the most realistic representation of MLB pitching in terms of how long it takes the ball to reach the plate from the pitchers hand it's too bad you have to sacrifice the accuracy in one area of the game to make another one accurate.

Hopefully SCEA gives this data a good look and can catch the problem for next year. Most of the numbers are looking great but I feel the simulations they run to get their inhouse numbers could use a reworking if such a glaring problem was able to exist.

[nomo17k]

Re: Stats-Based CPU Sliders [MLB 14 Version]

I saw there was a patch released a few days ago but did not see anything in the patch notes about 'fixing' the SB% using a higher pitch speed. I know sometimes they fix things without putting it in the notes, is there any improvement to this area since the patch? My gut tells me it's the same but I thought I would ask before I dive in again.

Having to play with a low pitch speed just to get an accurate SB% really ruins some of the fun for me.. It's such a glaring issue for an otherwise simulation game, it really stands out as being an aberration. I know SCEA Brian was posting in the thread earlier about the stealing issues and even fixed a part of it but sadly from the data it clearly still is a big problem. People have done tests to determine that Pitch Speed 10 is the most realistic representation of MLB pitching in terms of how long it takes the ball to reach the plate from the pitchers hand it's too bad you have to sacrifice the accuracy in one area of the game to make another one accurate.

Hopefully SCEA gives this data a good look and can catch the problem for next year. Most of the numbers are looking great but I feel the simulations they run to get their inhouse numbers could use a reworking if such a glaring problem was able to exist.

I haven't tested the SB aspect after the patch and don't know if anything has been done about the steal issue. Of course it's unfortunate but unless there is another patch (I doubt it), hoping for a fix is hopeless. And I think SCEA usually doesn't patch game balance issues unless something is "broken." The steal balance is bad, but is it broken??? Well.....

You don't have to worry about SCEA looking at my data for the upcoming release; they (Brian) run their own and track quite a bit more stats to tune the game. I just do my own to add a bit more walks.

A couple things were a bit off this year on default sliders (BABIP and steals) but I think the devs were really pressed for time in targeting three platforms at once. MLB 15 will probably just be back playing solidly on default.

[Bobhead]

Re: Stats-Based CPU Sliders [MLB 14 Version]

Funny how LD% is spot on for the CPU because I still find it to be drastically too low for a human player.

[nomo17k]

Re: Stats-Based CPU Sliders [MLB 14 Version]

Funny how LD% is spot on for the CPU because I still find it to be drastically too low for a human player.

With Zone, I actually find that you have A LOT of control over the type of ball you want to put in play. Of course you cannot always do what you desire especially when you fall behind, but that's what makes the pitcher-batter interaction so compelling in this game.

You have been complaining of over production of fly balls/popups, but personally I find that to be an issue only when I don't get myself ready for pitches thrown hard or high. I correct that tendency by pre-locating my PCI rather high and reacting to lower pitches, and it's been working well for me.

[Bobhead]

Re: Stats-Based CPU Sliders [MLB 14 Version]

With Zone, I actually find that you have A LOT of control over the type of ball you want to put in play. Of course you cannot always do what you desire especially when you fall behind, but that's what makes the pitcher-batter interaction so compelling in this game.

That was true in previous years but I really don't agree anymore. I find that even when I'm right on top of the ball all game, I'm getting home runs, and doubles and triples in the gaps but no actual LDs.

I feel like the fact I use a poorly ranked team offensively (the Mets) has something to do with it, but it shouldn't. Even the worst teams in baseball still have a LD% upwards of 19%. When I first got the game I ran a small number of test games (about 20+ or so) and I found LD% to be more or less spot on. But I used each team equally. I believe you do the same in your test games?

So I'm thinking that the issue is that BABIP an LD% are tied too much to contact ratings, when in reality, both are pretty consistent across all of MLB (it's the whiff ratings, OBP, and power that really define an offense)

[nomo17k]

Re: Stats-Based CPU Sliders [MLB 14 Version]

That was true in previous years but I really don't agree anymore. I find that even when I'm right on top of the ball all game, I'm getting home runs, and doubles and triples in the gaps but no actual LDs.

I feel like the fact I use a poorly ranked team offensively (the Mets) has something to do with it, but it shouldn't. Even the worst teams in baseball still have a LD% upwards of 19%. When I first got the game I ran a small number of test games (about 20+ or so) and I found LD% to be more or less spot on. But I used each team equally. I believe you do the same in your test games?

So I'm thinking that the issue is that BABIP an LD% are tied too much to contact ratings, when in reality, both are pretty consistent across all of MLB (it's the whiff ratings, OBP, and power that really define an offense)

How much different your in-game FB/LD/GB % from the MLB average?


Since the majority of safe hits are line drives, hitters who generates more of them tend to have higher BABIPs. BABIP for hitters vary quite a bit among different hitters compared to that of pitchers. Tony Gwynn had a career BABIP of about .340, for example.

I don't think the game can have it differently... the game needs some mechanism representing hitter's propensity to generate line drives (easiest way to adjust batting average), which roughly translates to higher BABIP as well as batting average:



(which is borrowed from here.)

Of course line drive are not the only factor in high batting average, someone like Ichiro produced tons of safe hits from grounders, but that's more a product of speed and his mastery of hitting to the opposite field. And that's exactly what the game hasn't really been able to reproduce well, unlike safe hits through line drives.

[Bobhead]

Re: Stats-Based CPU Sliders [MLB 14 Version]

I don't have an exact number because I'm not doing the computer spreadsheet stuff this year, but at the beginning of the year I checked the pitch analysis screen after every game, and I never once got an LD% at or above 19%, not even near 19%.... Every single game it would be like 4/50... 7/56, etc...I'd estimate that my LD% averaged around 11.

[kcsam]

Re: Stats-Based CPU Sliders [MLB 14 Version]

Hey Nomo. Have you done any tests in regards to the injury slider? I see a lot of different slider positions for injury, and am not quite sure which gives the most realistic count. I currently play on 8, but I'm beginning to think that may be a little too high. Do you or anyone else have any suggestions?

[harleyhorse8]

Re: Stats-Based CPU Sliders [MLB 14 Version]

Hey Nomo. Have you done any tests in regards to the injury slider? I see a lot of different slider positions for injury, and am not quite sure which gives the most realistic count. I currently play on 8, but I'm beginning to think that may be a little too high. Do you or anyone else have any suggestions?

That one's tough. Either 6 or 7 is good, but 6 is on the low side and 7 is a little high.

[kcsam]

Re: Stats-Based CPU Sliders [MLB 14 Version]

That one's tough. Either 6 or 7 is good, but 6 is on the low side and 7 is a little high.

Thanks for the suggestion. I guess I'll go with 7 initially. 8 seemed like way too many. Although, I hardly saw any in my played games at 8. Just a lot in the simmed ones. Mostly in the minors too, which is strange.

[Heroesandvillains]

Re: Stats-Based CPU Sliders [MLB 14 Version]

Thanks for the suggestion. I guess I'll go with 7 initially. 8 seemed like way too many. Although, I hardly saw any in my played games at 8. Just a lot in the simmed ones. Mostly in the minors too, which is strange.

The sim-engine and gameplay engine are two different animals.

I personally play with it at 7 and sim games with it at 5.

[kcsam]

Re: Stats-Based CPU Sliders [MLB 14 Version]

The sim-engine and gameplay engine are two different animals.

I personally play with it at 7 and sim games with it at 5.

Oh wow, so you play at 7 and then switch to 5 when simming? Are there any other adjustments you make between playing and simming?