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

[Unspoken55]

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

Hey Nomo I love your work. One thing that seems to not be so accurate all of the time is a stacked team dominating most of the time over weak teams. I was wondering if you ever test this way? It's one of the way I love playing cpu vs cpu and I seem to always be disappointed with how often I lose. So far this year seems a bit better based on the very limited amount I have played, but my expectations are low.

[nomo17k]

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

Hey Nomo I love your work. One thing that seems to not be so accurate all of the time is a stacked team dominating most of the time over weak teams. I was wondering if you ever test this way? It's one of the way I love playing cpu vs cpu and I seem to always be disappointed with how often I lose. So far this year seems a bit better based on the very limited amount I have played, but my expectations are low.

I am not sure if I am understanding you fully, but you mean you have one team stacked up with highly rated players and play them against mediocre team?

Actually, one thing I always enjoy doing is to have a minor league team play against an MLB team... In the past I typically used the Astros since they were so bad (today it would be the Padres...) and played them against a better AAA or AA team... it's always interesting, since it kinda demonstrates how attribute rating separation created play quality difference in these levels.

When I do this, the worst MLB team still beats a minor league team in the vast majority of games... So I actually don't know if I agree with your observations.

[Unspoken55]

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

What I do is take the best at each position and play out a season. With this team in the past it seemed as if I would lose to much. I have not used your sliders this year yet, but have just used hall of fame difficulty and it has been 👌. But I will add you sliders, and play some games and see how it feels.

Sent from my GT-P5210 using Operation Sports mobile app

[tgreer]

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

Nomo it is strange that you are seeing so few errors on the Infield even on your most current set. This is what I have been running with last 15 games for sliders and avg stats in ()

IF Fld Err: 10 (MLB 0.24/Mine 0.33)
IF Thr Err: 8 (MLB 0.24/Mine 0.06)

[nomo17k]

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

Nomo it is strange that you are seeing so few errors on the Infield even on your most current set. This is what I have been running with last 15 games for sliders and avg stats in ()

IF Fld Err: 10 (MLB 0.24/Mine 0.33)
IF Thr Err: 8 (MLB 0.24/Mine 0.06)

I think 15 games is way too small a sample size for fielding error stats to stabilize... those are events that happen less than once (0.24 is about once every four games in fact), so depending on you are in a stretch of (un)luckiness, these numbers can easily fluctuate above and below what we "want" to see.

By now, I think I'd bet my rent payment (I do not own a house) or grandma (I do not have a girl friend) that you will end up seeing less fielding errors even if you jack the slider up to 10.

I think there is some general issue of that particular slider not affecting the propensity for errors enough, and looking more in detail (replays and logs), there are a class of playes that the game should record as errors but instead scoring as singles (which decreased the number of recorded errors).

I've noticed an increase in the latter kind of "scoring errors" the past couple years (some of which I probably have uploaded on YouTube: https://www.youtube.com/user/baseball17k/videos), and I'm trying to see if counting them as errors would bring up the numbers more inline with the real-life numbers. I don't think it has been particularly promising though, so I'm more inclined to think that the fielding error slider is not very sensitive this year.

[nomo17k]

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

Browsing the YouTube account I've been using primarily to post in-game videos to report issue, I realized I've captured the results of a full season of 2430 CPU vs. CPU games from MLB 15.

Spoiler

<div style="position:relative;height:0;padding-bottom:56.25%"><iframe src="https://www.youtube.com/embed/5gMd74G6meE?ecver=2" width="640" height="360" frameborder="0" style="position:absolute;width:100%;height:100%;le ft:0" allowfullscreen></iframe></div>

That was a year where I think the major (unsolved) issues were steal success being a little down, and bunting was too frequent and successful. But other than that the game was tuned pretty well I think, and it's interesting to compare these stats to what we get when we "sim" games, which is what most people do.

Take a look if you ever wonder what happens if you "never" sim any games and let CPU play the game out when you are not controlling. I think the results look pretty good.

[miamikb2001]

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

I think 15 games is way too small a sample size for fielding error stats to stabilize... those are events that happen less than once (0.24 is about once every four games in fact), so depending on you are in a stretch of (un)luckiness, these numbers can easily fluctuate above and below what we "want" to see.



By now, I think I'd bet my rent payment (I do not own a house) or grandma (I do not have a girl friend) that you will end up seeing less fielding errors even if you jack the slider up to 10.



I think there is some general issue of that particular slider not affecting the propensity for errors enough, and looking more in detail (replays and logs), there are a class of playes that the game should record as errors but instead scoring as singles (which decreased the number of recorded errors).



I've noticed an increase in the latter kind of "scoring errors" the past couple years (some of which I probably have uploaded on YouTube: https://www.youtube.com/user/baseball17k/videos), and I'm trying to see if counting them as errors would bring up the numbers more inline with the real-life numbers. I don't think it has been particularly promising though, so I'm more inclined to think that the fielding error slider is not very sensitive this year.

I was gonna mention that too,i've seen a number of plays that should have been scored errors scored as hits......So it appears The Show has generous offensive minded Official Scorers lol!


Sent from my iPhone using Tapatalk

[nomo17k]

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

I was gonna mention that too,i've seen a number of plays that should have been scored errors scored as hits......So it appears The Show has generous offensive minded Official Scorers lol!

These are all scored as hits in the game:

Spoiler

<div style="position:relative;height:0;padding-bottom:56.25%"><iframe src="https://www.youtube.com/embed/4oQI-Ml2EYY?ecver=2" width="640" height="360" frameborder="0" style="position:absolute;width:100%;height:100%;le ft:0" allowfullscreen></iframe></div>

<div style="position:relative;height:0;padding-bottom:56.25%"><iframe src="https://www.youtube.com/embed/Kt47N99BtVk?ecver=2" width="640" height="360" frameborder="0" style="position:absolute;width:100%;height:100%;le ft:0" allowfullscreen></iframe></div>

<div style="position:relative;height:0;padding-bottom:56.25%"><iframe src="https://www.youtube.com/embed/jDhmMQb7Exk?ecver=2" width="640" height="360" frameborder="0" style="position:absolute;width:100%;height:100%;le ft:0" allowfullscreen></iframe></div>

<div style="position:relative;height:0;padding-bottom:56.25%"><iframe src="https://www.youtube.com/embed/MQvW5W6ZoDY?ecver=2" width="640" height="360" frameborder="0" style="position:absolute;width:100%;height:100%;le ft:0" allowfullscreen></iframe></div>

<div style="position:relative;height:0;padding-bottom:56.25%"><iframe src="https://www.youtube.com/embed/_VLKS6kro_Y?ecver=2" width="640" height="360" frameborder="0" style="position:absolute;width:100%;height:100%;le ft:0" allowfullscreen></iframe></div>

<div style="position:relative;height:0;padding-bottom:56.25%"><iframe src="https://www.youtube.com/embed/Sk6dj6l1Umk?ecver=2" width="640" height="360" frameborder="0" style="position:absolute;width:100%;height:100%;le ft:0" allowfullscreen></iframe></div>

<div style="position:relative;height:0;padding-bottom:56.25%"><iframe src="https://www.youtube.com/embed/ZDnsomPc8l8?ecver=2" width="640" height="360" frameborder="0" style="position:absolute;width:100%;height:100%;le ft:0" allowfullscreen></iframe></div>

A couple could be considered a hit (in the eyes of beholder), but there are quite a few obvious blunders that should be scored as errors.

I think whether a play gets scored as error is somehow tied with fielding animations, but I think there are what are meant to be fielding error animations that are not flagged as errors.

[jripper09]

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

also with fielder error at 10 there are more bobbles and then recovery throw runner out. that wouldnt happen if slider was normal. plus more bobbled balls in general but like you showed in youre display there not recorded as errors. maybee the AI gets confused when this happens because most of the time it is scored a hit. if on 5 the computer scores the error or non error correctly. i think the fielding error slider should just be called the bobble slider.

[miamikb2001]

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

also with fielder error at 10 there are more bobbles and then recovery throw runner out. that wouldnt happen if slider was normal. plus more bobbled balls in general but like you showed in youre display there not recorded as errors. maybee the AI gets confused when this happens because most of the time it is scored a hit. if on 5 the computer scores the error or non error correctly. i think the fielding error slider should just be called the bobble slider.

I pretty much agree with your post completly.Hopefully a small tweak in a patch.


Sent from my iPhone using Tapatalk

[Brownsfans25]

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

What difficulty r u using all start on hall of fame

[Brownsfans25]

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

[QUOTE=Brownsfans25;2048781321]What difficulty r u using all start or hall of fame[/QUOT
E]misspelled Uru

[nomo17k]

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

What difficulty r u using all start on hall of fame

Difficulty level does not have any effects in CPU vs. CPU games.

also with fielder error at 10 there are more bobbles and then recovery throw runner out. that wouldnt happen if slider was normal. plus more bobbled balls in general but like you showed in youre display there not recorded as errors. maybee the AI gets confused when this happens because most of the time it is scored a hit. if on 5 the computer scores the error or non error correctly. i think the fielding error slider should just be called the bobble slider.

I think it is actually exactly that, Fielding Error slider adjust the frequency of bobbles and fly balls being dropped. I'm still not entirely sure those are happening as frequently as they should given the Error slider is maxed out.

But what is definitely happening is that many of those bobbles are not properly scored as errors by the game. I don't know if that's *the* issue but it is a significant part of why fielding error number is down.

[nomo17k]

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

As far as I know, only Pitcher Consistency slider is effective in controlling WPs (and HPBs for that matter). Blocking attribute could be edited as well, but that's something of last resort if you are into editing...

I don't think fielding sliders/attributes have anything to do with wild pitches (not that they shouldn't, but I haven't seen enough evidence that they do).

Out of boredom (though I gotta be done with my tax filing really soon...), I did some experimenting with the Block attribute of catchers... and I think the reason why wild pitches are so high and Pitcher Consistency slider doesn't really change WPs, at least a significant portion of it, is due to the Block attribute rating not really changing effectiveness of catchers blocking balls in the dirt.

Basically all catchers, regardless of their Block attribute, don't handle pitches in the dirt well this year.

I was gonna capture a video showing the difference (which I just didn't capture properly...), and I think when you see it you'd probably agree that the way Block 99 and Block 0 catchers handle wild pitches are really about the same.

In the past, I think catchers rated highly in Block tended to handle pitches in the dirt cleanly more often, but this years game, no matter what Block attribute is, the catcher really doesn't handle them any differently.

If you are curious I'd encourage you trying this... I suppose we should create a bug report...

[tgreer]

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

The 4/7 set seems to be trending in the right direction ...Walks look low, but everything else it seems each time I look is improving