Stats-Based CPU Sliders [MLB 17 Version]

[orye74]

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

Do you really have manager hook at 9'right now?

[nomo17k]

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

Do you really have manager hook at 9'right now?

I'm just testing reliever usage right now. I think for the most realistic starter/bullpen usage, Manager Hook might need to be changed once the starter gets pulled. To avoid excessive early hooks, it needs to be set lower when starters are pitching, but once they are gone, it might be better increased to some higher value so that CPU manager doesn't leave each reliever too long.

[orye74]

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

I tried the manager hook on 9 for 1 game but I think they definitely pulled the Rays starting pitcher too early. Is it me or is anyone else seeing visiting teams winning about 75-80% of the games?

[nomo17k]

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

I tried the manager hook on 9 for 1 game but I think they definitely pulled the Rays starting pitcher too early. Is it me or is anyone else seeing visiting teams winning about 75-80% of the games?

Yeah, Manager Hook needs to be kept lower for starters... otherwise there will be too many unrealistic early hooks... though it could actually be useful for playoff situations. Unfortunately, even then, the situation for early hook doesn't make sense at all. I've seen pitchers getting pulled before 5 IP even on the winning side. Very weird logic...

You must be just unlucky (or sampling of teams is biased) to see visiting teams winning so often. I don't think there's any home/away advantage in the game. In real-life, though, there is home field advantage even in baseball. To me, home-field advantage feels a bit weird in baseball, but nonetheless data appear to indicate it exists...

[doctorhay53]

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

Having the opportunity to score last is a pretty big HFA. You can end the game without giving the other team a chance to answer. It's equivalent to always having the last possession in football or basketball.

I also saw a recent article about the NBA showing that as players have matured and taken better care of their bodies (and avoided night life on road trips), HFA has declined a bit. Apparently a big reason for the 'classical' HFA was traveling arrangements and night life, but as teams have improved the flying experience and taught players to avoid alcohol and such, HFA has declined (this is NBA specific, but probably the same situations apply in all sports).

But the MLB is unique in the sense that home-field gives a tangible, well-defined HFA by giving the home team the last chance to score.

[orye74]

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

I see you keep dropping fielder run speed. Have you tried upping the BR speed to 6 instead? i did this and get a decent amount of doubles without taking too much away from double plays.

[nomo17k]

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

Having the opportunity to score last is a pretty big HFA. You can end the game without giving the other team a chance to answer. It's equivalent to always having the last possession in football or basketball.

I also saw a recent article about the NBA showing that as players have matured and taken better care of their bodies (and avoided night life on road trips), HFA has declined a bit. Apparently a big reason for the 'classical' HFA was traveling arrangements and night life, but as teams have improved the flying experience and taught players to avoid alcohol and such, HFA has declined (this is NBA specific, but probably the same situations apply in all sports).

But the MLB is unique in the sense that home-field gives a tangible, well-defined HFA by giving the home team the last chance to score.

That's an interesting story about night life! I wonder if travel magazine or something could come up with a ranking of "America's best night life" by comparing the amount of home field advantage by city...

I actually don't know the "bottom of the 9th advantage" holds up scrutiny... what if we switch home/visitor for baseball games, and do we still see it? In this day and age, someone might have already looked at the data. I've only read articles like this one on the topic: http://sportsworld.nbcsports.com/hom...age-important/

[nomo17k]

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

I see you keep dropping fielder run speed. Have you tried upping the BR speed to 6 instead? i did this and get a decent amount of doubles without taking too much away from double plays.

What I am trying to do is to increase BABIP by trying to make balls in play fall for hit a little more. That's why I'm trying increasing Solid Hit to convert some fly balls/ground balls into more line drives... (though I may have hit this year's version of "weirdness threshold" with experimenting with the slider values at 5, 6, and 7.... if the transition were smooth across those values, I would've expected 6 would be just right, but 5 and 6 have yielded similar results but BABIP just jumped when it was set to 7... I don't quite understand this mechanics but I've seen it happen the last couple years.)

Fielder Speed isn't a major adjustment, and I just did it because I don't think it changes much, but at least it is in a "right" direction to cause the effect that I want to see more (i.e., balls in play fall for more base hits). I don't want to decrease it too much though, as at some point outfielders start having issues catching up to shots in the gap, turning singles into doubles rather awkwardly.

[doctorhay53]

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

Here's an interesting post on the topic of HFA in baseball:

http://www.fangraphs.com/blogs/is-ho...ng-endangered/

Also, interesting stuff about batting last:

http://retrosheet.org/Research/PankinM/Pankin-sabr42posters.pdf

Anyway, just wanted to say keep up the good work.

[tgreer]

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

Think I am going to go with this 5/8 set for my MoM. I am ready to start! lol...Numbers looking good, I know you said you were tinkering with a few things still like Manager Hook. What would you recommend? 7-9? Anything else giving you concerns? Thanks Nomo

[nomo17k]

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

Think I am going to go with this 5/8 set for my MoM. I am ready to start! lol...Numbers looking good, I know you said you were tinkering with a few things still like Manager Hook. What would you recommend? 7-9? Anything else giving you concerns? Thanks Nomo

I wouldn't say that the numbers have stabilized for the 5/8 set, but it does look okay so far... at least I'm seeing what I've wanted to see, and not much of red-flag moments. If things keep this way, I'm likely to go with it at least as the beta set. I'm just watching if I'm toning down the offensive boost by Solid Hit 7 with Pitcher Control 3 now... If I'm successful, then I should see less offensive breakout games (which gives sloppy feel to the games) than I saw in the 5/5 set, and also the batting average should be right around where I want it (.255). So far, it's been a nice mix of low and high scoring games so I'm encouraged... but that's the kind of thing I'm specifically watching for.

I'd keep Manager Hook lower than the default, perhaps 1 or 2, since otherwise I think you'll probably see questionable early hooks (like CPU pulling starter in 5th inning when the team is up).

[tgreer]

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

I wouldn't say that the numbers have stabilized for the 5/8 set, but it does look okay so far... at least I'm seeing what I've wanted to see, and not much of red-flag moments. If things keep this way, I'm likely to go with it at least as the beta set. I'm just watching if I'm toning down the offensive boost by Solid Hit 7 with Pitcher Control 3 now... If I'm successful, then I should see less offensive breakout games (which gives sloppy feel to the games) than I saw in the 5/5 set, and also the batting average should be right around where I want it (.255). So far, it's been a nice mix of low and high scoring games so I'm encouraged... but that's the kind of thing I'm specifically watching for.

I'd keep Manager Hook lower than the default, perhaps 1 or 2, since otherwise I think you'll probably see questionable early hooks (like CPU pulling starter in 5th inning when the team is up).

Yes thats what i was trying to say but I went the wrong way with it. I will more then likely go with 2 based on your results

[bcruise]

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

Any thoughts on whether wild pitches have been tuned any further with 1.06 nomo? Nothing in the notes, but they didn't mention it there last time they were tuned either.

Yours is the only data I can take seriously on this since you have a strong base set of games and are therefore able to track how much it's really changed since then. And because there's none of that pesky user input factor to get in the way (and that's a HUGE factor in WP's).

Thanks!

[nomo17k]

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

Any thoughts on whether wild pitches have been tuned any further with 1.06 nomo? Nothing in the notes, but they didn't mention it there last time they were tuned either.

Yours is the only data I can take seriously on this since you have a strong base set of games and are therefore able to track how much it's really changed since then. And because there's none of that pesky user input factor to get in the way (and that's a HUGE factor in WP's).

Thanks!

Will see... I just installed the patch v1.06, so I don't have data yet. WPs probably being the last/biggest remaining issue that require help from patch, I'm certainly interested in if there's been any improvement there.

[floydpinkster]

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

First Game after 1.06 patch with Nomo's Sliders
This game is so Broken.
This is Ricky Vaughn Pre-Glasses kinda stuff.
Howell 3 WP
Tillman 3 WP
Jimenez 1 WP
3 HBP in the game as well