Stats-Based CPU Sliders [MLB 14 Version]

[Bobhead]

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

Interesting - so basically, it depends on how many line drives you're getting now?

I.E. if you have a high contact slider (lots of line drives), raising solid hits would just give you even more line drives - cutting down what little flyballs you'd get? So lowering it would likely give you diverse hit types again, including more flies which are still likely well-hit (high contact slider), so more HR?

I've done 0 testing on that. It's just... conjecture?

What I know is that I play on Hall of Fame, and in previous years, when I increased Solid Hits, I got a pronounced increase in Home Runs (among other things).

So while I want to, and have reason to believe the theory proposed by another user. My own experience tells me it's not a "straight-line" type of situation.

[Bobhead]

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

Nomo, any final word on whether you think the baserunning ability slider is reversed again?

My own limited experienced seems to suggest it is, but my experience is very limited. I've only played about 10 games this year (really busy!). So I figured I'd check with you.

On a side note, I also think the Steal Frequency slider might be reversed. I dont remember if that was always a thing or not. Then again we've never really had a reason to ever touch this slider. But this year I think this slider might be the key to fixing these pickoff shenanigans, so I've been playing around with it. I set it to 9 and see a huge drop in attempted steals/overall aggressiveness by the CPU.

So yeah... what are your thoughts and observations? Reversed, or no?

[Cowboys63]

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

Nomo,
Don't your sliders do basically the same thing that the hybrid rosters do. I've been thinking about using the hybrid roster, but I like to download and use legend players from my lifetime as well and there doesn't seem to be good information (or I can't find it) on re-rating downloaded legends. But if the goal is to get accurate stats, your sliders are designed to accomplish that as well with the regular ratings....correct? If so, I believe I will stick with non-hybrid rosters. I play manager only mode.

Thanks for your hard work and in advance for your reply.

[Heroesandvillains]

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

I haven't seem Nomo on here since that (unfortunate and expected) argument with another OS'er a little while back.

He deserves to be treated better as he does us all a great service and he's also a real nice (and brilliant) guy IRL.

I hope he didn't decide to pull the plug on going public with his testing. That'd be a real downer.

[kcsam]

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

I haven't seem Nomo on here since that (unfortunate and expected) argument with another OS'er a little while back.

He deserves to be treated better as he does us all a great service and he's also a real nice (and brilliant) guy IRL.

I hope he didn't decide to pull the plug on going public with his testing. That'd be a real downer.

For sure, I totally agree. I'm new to the game, so his sliders and info on them was greatly beneficial to me. Also, his testing was an awesome asset. I relied on hit heavily.

Didn't know he had an argument with someone. Could he have been banned? If so, that would be tremendously unfortunate. I hope to see him back soon. Still trying to find some realistic sliders for QC. Currently running with his latest ones with non QC.

[KBLover]

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

Didn't know he had an argument with someone. Could he have been banned? If so, that would be tremendously unfortunate.

His name is still blue so I don't think he's been banned, and I don't think he'd say anything that would lead to it.

Hopefully, he's just really busy or such.

[Heroesandvillains]

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

Guys! LOL! I was too vague. Didn't mean to sound cryptic.

It was nothing like you're imagining.

Let's drop the topic and cross our fingers he returns. If you want to see the back and forth it wouldn't be too tough to find.

I was just "hoping" out loud. Let's not derail the thread any further out of respect for our OS friend Nomo.

[nomo17k]

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

Been playing full games the past two days, and there is definitely a difference in the way the CPU plays versus QC. Fewer walks, less deep counts, and more balls are also put in play. I also think fewer "meat balls" are thrown since there are fewer full or three ball counts. Of course, it could be all in my head due to such a small sample of games played.

Looking back at how things were flowing with QC, I'll most likely stick with full games from now on. The way the game flows feels more realistic. Plus the pitcher versus batter matchups are more interesting. Kind of a downer seeing Kimbrel versus McCutchen in a close game only to have him start with an 0-2 count or even 3-0 count.

I agree. I had played a lot of cpu vs cpu games with QC and I hardly ever saw high scoring/hit games. I played my first regular game last night with the A's vs. White Sox, and the A's dropped 8 runs in the 1st inning while only recording 1 out! The final was 11-1. A's had 18 hits. By far the most runs and hits I had ever seen in QC.

So yeah, the engine definitely plays differently on QC. Not sure sliders can be changed around to reflect a more realistic game with QC.

According to what the devs (Ramone) have said I think what the game really wanted to achieve with QC is to shorten the game time. I think how the counts are generated in QC tends to push toward deeper counts to force things to happen (i.e., ball put in play, striking out, or walking).

I think a better way of taking advantage of QC while playing realistically, if you actually play (and not just watch as in CPU vs. CPU), is to have a mindset of putting in play the first pitch that you can reasonably do so. If you stay patient and selective, then you will necessarily go deep in count all the time. The difference is that when you are not using QC, then there are quite a few at-bats in which (you think) you see a hittable pitch and that one is put in play. With QC, you won't see this happening because early counts are often skipped.

On one hand, QC reduces the amount of pitches you can do something with in an at-bat by starting with a deeper count. But on the other hand, QC also prevents us from doing something earlier in an at-bat. There is clearly some trade off there.

In turn, I think QC really isn't tuned right now to play realistically in CPU vs. CPU.

[nomo17k]

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

How would I go ab out using these sliders for my franchise? Would I use the same number for both human and cpu?

That's what I usually do. You can simply go to the Slider Vault and download this slider.

But this slider set is meant to be used for CPU vs. CPU games. If you actually control players, your mileage may vary.

[nomo17k]

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

I love this thread. It's always great.

Edit: Just finished reading the entire thread.

Throw me in the boat of people that think pickoffs are a suspect in this base stealing issue. I've only played a handful of games and have already seen numerous pickoffs. This is with default sliders mind you.

I also am inclined to agree with the suggestion that increasing the Solid Hits slider could lower Home Runs. What I want to add is that I think this function is a parabolic one. If you set Solid Hits to 0, Contact to 0, etc... at that point, I believe increasing Solid Hits at that point would create a net increase in home runs. Yes you may have fly balls turned to line drives, but you also have a ridiculous number of poorly hit balls that are now turned into hard hits, etc... I say this because I play on higher difficulties, and I am very confident that given the hitting variables on Legend or Hall of Fame (and I think I had Contact at 3 one year), increasing the Solid Hits slider produces a net increase in home runs.

But stevie's theory definitely makes sense. As nomo said, based on my (and his) understanding of the SH slider, it would probably turn fly balls to line drives, under the right conditions.

I think your case is complicated by the fact that you use Zone hitting. I remember that you were complaining a bit about how MLB 13 felt "floaty" in how the game appeared to produce more fly balls and popups.

That might actually tell you something, that you might have a tendency to hit the lower part of the pitch with Zone, so that you tend to hit more fly balls and therefore more home runs.

If that's the case, you may actually be seeing your tendency to hit, and not the effect of Solid Hit slider.

Over the last couple years, I have varied Solid Hit slider to different values, and it's effect has been pretty consistent with it changing the fraction of line drives in place of fly balls and ground balls.

Also nomo, have you tried tinkering with the BR Steal Frequency slider to fix the stealing issue? It's the slider I've been playing with and it seems to be doing the trick.

It might just be me but it also appears to be reversed this year. I don't remember if that was a thing in previous years.

I haven't seen anything wrong with the number of steal attempts. They aren't that off at default so I haven't bothered changing. I don't believe it is reversed, as it never has been. I haven't tested though.

[nomo17k]

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

One, this is awesome. I love to study the stats. That said, how do you get all of your stats? Manually entry? From team pages (along the bottom)?

I'd love to do more analyzing of the stats myself, but unlike something online where I can just write a web scrape script, it's not (at least to me) seemingly that simple with The Show.

Personally, if I could download the individual stats it would solve the historic player card issue for me, as I'd just load them into Excel or my SQL database to oogle at, er smile like a creep at.

Is there any update on the sliders vs stats at this point? Would love to hear where your progress is at.

I have problems with batting averages, both in having individuals get over .315 for a batting crown (would prefer league winners at least .330 as over 90% of batting crowns are over .330) and lg batting averages around .235-240. This is done mostly simming.

As I get to where most of the MLB players are gone, seasons 15+ the lg batting averages converge to a higher mean (around .265-.270), but the individuals aren't much higher...occasionally see a .330.

On Quick Counts, I don't get many 3-2 counts. I do get a ton of 1-2 counts. The most common. I don't play that often (3-4 games a week) so it may be user error, but I tend to average 12-16 K's (as a batter) per game. A bit high and have to really work to get the 4-5 BB's. It seems it's heavily based on the confidence of the pitcher early in the game. If he walks a few and gives up 2-3 hits in the 1st two innings, he's likely to walk more, if he doesn't - I'll be lucky to walk 2 times in the game.

I've then tinkered with sliders quite a bit and played QC games, however, at least for my somewhat limited playing, it seems that "pitcher confidence" early on dictates the BB/SO's for while they're in there more than any slider.

I look at box scores and pitcher analysis screen after each CPU vs. CPU game to tally all the numbers and record them into spreadsheet.

I think the idea of online franchise is partly to enable more stats tracking. Too bad online doesn't work well this year. It's fair to say most guys want better historic stats tracking. It's just too minimal at its current state.

For playing, I think I'm fairly satisfied with the last slider set, so for CPU vs. CPU I don't think there will be much change... the only thing that's still very off is steals, but that's not really correctable with sliders... I made a compromise by lowering Pitch Speed, which should work for CPU games.

Sim stats have gotten more stable I think. At least it's better than the last few years, in which the offensive stats really got inflated.


QC... I haven't gotten in as many games using that mode, so I'm still holding my judgement. But I would have preferred a system in which we can forward pitch by pitch or so. I'm very patient when playing, so it hasn't really given me the advantage of shorter game. I'd still typically see around three pitches per at-bat. Shorter, but not very drastically.

[nomo17k]

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

Nomo,

Did you ever do any work on what the effect of bumping up a player's power would do to the stats as opposed to tuning the sliders? I'm doing well in my test with everything except I just can't get the HRs where I want them. 5 is about.15 too low and 6 is way too much. Just wanted to see if you had anything before I start playing with it.

I think power is generally easy to understand. I am not a fan of editing attributes for the sake of adjusting the game, but power attribute will pretty much do the same thing as the power slider does, just a lot more finer.

I don't know after how many games you are citing the number, but if it's a small sample then I wouldn't worry about it. Or if the offense overall is a bit low and not just the power, then I may actually lower Pitcher Control by one or two, to increase hittable pitches.

[nomo17k]

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

So I plug each number in for both human and CPU, I can use any hitting/pitching interface I chose to and any difficulty will work with these? If I am wrong, someone please correct me.

I responded to your earlier post too.

The slider is for CPU vs. CPU, so your mileage may vary for your own gameplay. And as the difficulty level doesn't matter in CPU vs. CPU games (it's said to be roughly at All-star), your mileage may also vary if your skills are at different level.

In general, you shouldn't worry too much about sliders you are using. Find your difficulty level first (either by manually adjusting difficulty level till you play realistically or by using Dynamic Difficulty), and if you feel something is really missing, you should just adjust that particular slider that neutralizes your tendency.

What my slider does is to neutralize the tendency of CPU (to what MLB stats indicate). So if your skills and approach are similar to CPU, then this slider set would work well for you. If not, then it's not going to work as well...

[nomo17k]

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

Interesting - so basically, it depends on how many line drives you're getting now?

I.E. if you have a high contact slider (lots of line drives), raising solid hits would just give you even more line drives - cutting down what little flyballs you'd get? So lowering it would likely give you diverse hit types again, including more flies which are still likely well-hit (high contact slider), so more HR?

I've done 0 testing on that. It's just... conjecture?

What I know is that I play on Hall of Fame, and in previous years, when I increased Solid Hits, I got a pronounced increase in Home Runs (among other things).

So while I want to, and have reason to believe the theory proposed by another user. My own experience tells me it's not a "straight-line" type of situation.

I haven't explicitly tested this kind of combination, but I would think things would probably happen as KBLover describes.

Increasing the Contact slider increased the chance of producing line drives, fouling off pitches (that would otherwise lead to whiffs), discipline (i.e., easier to check swing), all at once. But if you actually decrease the Solid Hit slider by a (unknown) correct amount, then I think it is possible to keep the ratio of fly ball/line drive/ground ball roughly constant while letting the Contact slider affect the other aspects.

Bobhead's experience is harder to interpret because he uses Zone, and the PCI placement has a lot bigger effect on the hit variety than what sliders can do.

Nomo, any final word on whether you think the baserunning ability slider is reversed again?

My own limited experienced seems to suggest it is, but my experience is very limited. I've only played about 10 games this year (really busy!). So I figured I'd check with you.

On a side note, I also think the Steal Frequency slider might be reversed. I dont remember if that was always a thing or not. Then again we've never really had a reason to ever touch this slider. But this year I think this slider might be the key to fixing these pickoff shenanigans, so I've been playing around with it. I set it to 9 and see a huge drop in attempted steals/overall aggressiveness by the CPU.

So yeah... what are your thoughts and observations? Reversed, or no?

I don't think it's BR Steal Frequency. Although I did feel out of the box CPU attempts to steal a bit more than it should, but the overall, the numbers aren't really that off. What really stood out was that way too many steals (or failed hit-and-run) end in caught stealing. Unfortunately, this cannot be fixed by the BR sliders as well...

Pickoff cannot be fixed by slider. I think there's something off about how CPU takes a lead this year. You just need to

Overall I think something a bit funky has happened with steal-related portion of the game this year.

[nomo17k]

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

Nomo,
Don't your sliders do basically the same thing that the hybrid rosters do. I've been thinking about using the hybrid roster, but I like to download and use legend players from my lifetime as well and there doesn't seem to be good information (or I can't find it) on re-rating downloaded legends. But if the goal is to get accurate stats, your sliders are designed to accomplish that as well with the regular ratings....correct? If so, I believe I will stick with non-hybrid rosters. I play manager only mode.

Thanks for your hard work and in advance for your reply.

So long as the roster is using the attribute rating distribution that is not too far off from the default SCEA roster, the stats should come out okay. I use the default SCEA roster when I do CPU vs. CPU games.

A good check is to first simulate the entire season by the roster of your choice. Look at the end-of-the-season stats. Do this may be a couple times. If the stats look fine, then the roster should be okay. If not, then it's the player editing issue and sliders won't fix them.