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

[KBLover]

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

I actually still doubt Control works as a direct modifier to batter's PCI. Since it influences the overall offensive output, it feels that it has a direct influence, but I think the correlation is spurious. Of course I cannot make this claim with any certainty, but it doesn't fit at all with what B Ma has said about how the Control/Consistency/Strike Frequency sliders work... and I still think his explanation and implementation make sense mostly, so breaking away from it appears very unreasonable thing to do.

I don't know "how" it works, but if I raise Control, CPU offense dies, even if the pitch itself is badly located. They swing and miss more, poorer contact labels, more chases. This is using Classic so no stick skills involvement from me.

Likewise if I lower Control, the CPU makes more contact, hits the ball harder and chases less. Same if I raise CPU control against me hitting (using Directional so minimal influence from me). My hitters get bad contact, miss more, etc.

Control seems to impact the "execution" of the pitch so that whatever calculations go on under the hood to determine the hit chance (in batter analysis) and such are affected.

Control impacts accuracy of the physical location of the pitch, but it sure seems to do something as well to what the hitters do and how they respond to the pitch, which impacts the results they get. It might not impact size of PCI, but seems to impact how they use it and how accurately they judge pitches and the zone itself.

[NolanRyansSnowmonkey]

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

Trying to decide if I should just not worry about batted ball types at all in my human vs CPU set. I know the main thing is the other numbers and averages, but sure would be nice to get batted ball types closer (I know i can get them closer than CPU vs CPU because of where I place my pci). The bad thing about them not lining up is you have to find a way to get the ground ball batting average to be higher, since there generally won't be as many ground balls hit in real life. So more balls have to get through the infield in order to get babip in the correct range.

I guess that's the main thing that's bothering me is all the seeing-eye singles and other ground ball hits. And line drives are at about 16% instead of 20% (even when I score them how i understand them).

All other numbers look pretty good right now. I think if i could just get line drives to 20% i can live with ground ball and fly ball ratios being off.

I've only played 9 games with this exact set but the numbers are:

(Human hitting only)
AB. 33.7 (LG Avg 34.1)
R. 4.44 (4.48)
H. 8.44 (8.71)
RBI 4.33 (4.27)
BB 2.8 (3.11) in real games im a bit more patient than when im testing but still close. Easy fix.
SO 8.00 (8.03)
1B. 5.67 (5.53)
2B. 1.33 (1.70)
3B. 0.00 (0.18)
HR 1.44 (1.16)
HBP 0.56 (0.34)
SF 0.33 (0.25)
GIDP 0.67 (0.77)
Avg .247 (.255) too low...
Babip .279 (.300) Too low...
SO% 21.2 (21.1)

One game had 0 doubles and 4 HR so without that game the numbers are almost exact.

I know it's only 9 games and at this point the numbers can change very quickly but so far so good.

I'm trying to figure out how to get my babip up to .300 without throwing everything off. I may have to lower fielder reaction but there goes the ground ball average even higher lol.

The odd thing about this set is it was just an experiment in foul frequency. I tried it at 0, 5, and 10 just to learn something about it. (I haven't learned anything much lol) I "might" can see some difference but not even sure. But somehow with foul frequency at 10 (which I would have never tried with an actual set) the numbers started falling into place close to what they should be.

Is this coincidence? Is there anything unrealistic or off by having it at 10? Ive never seen any slider set do this. Not even sure what it's accomplishing. I'm assuming it's bringing babip up a small amount due to less weak hits landing fair? I really don't like being super far off of default on hitting sliders but not sure why I even feel that way.

[KBLover]

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

The bad thing about them not lining up is you have to find a way to get the ground ball batting average to be higher, since there generally won't be as many ground balls hit in real life. So more balls have to get through the infield in order to get babip in the correct range.

I guess that's the main thing that's bothering me is all the seeing-eye singles and other ground ball hits. And line drives are at about 16% instead of 20% (even when I score them how i understand them).

I feel your pain. That's actually the trade-off I'm going to make this year if I can't get the LD rate accurate (and LDs in the GAP more than just simple singles, which then impacts 2B+3B rate since FBs are unreliable for that given they turn into hits the least...or into HR).

I was able to get LD rate around 20% last year. Haven't been able to consistently this year. I'm guessing the new hitting system is part of the reason.

I hate the FB/GB ratio being off because that, too, impacts BABIP since FB are the least in-play hits (and in The Show, maybe even less so than IRL because of how non-HR flies are rendered most of the time). Part of why I have to make GB "too good" is because of this.

I'm trying to figure out how to get my babip up to .300 without throwing everything off. I may have to lower fielder reaction but there goes the ground ball average even higher lol.

My fielder reaction is 1 right now, fielder speed 0. I contemplated moving reaction to 2, but after some more games, both me and CPU had lower GB BABIPs so I'm staying at one. Having some bloopers, flares, and "funky" hits find grass is nice, not to mention not tracking down every foul pop. Also makes Reaction and Speed ratings more important/highlighted. I see a real difference between guys that rob hitters blind or range all over the place and ones that can only "turn and chase" because they know they can't get there at all.

The odd thing about this set is it was just an experiment in foul frequency. I tried it at 0, 5, and 10 just to learn something about it. (I haven't learned anything much lol) I "might" can see some difference but not even sure. But somehow with foul frequency at 10 (which I would have never tried with an actual set) the numbers started falling into place close to what they should be.

Is this coincidence? Is there anything unrealistic or off by having it at 10? Ive never seen any slider set do this. Not even sure what it's accomplishing. I'm assuming it's bringing babip up a small amount due to less weak hits landing fair? I really don't like being super far off of default on hitting sliders but not sure why I even feel that way.

It's funny. I had Foul Frequency at 10 in MLB 16 and 0 last year and this year.

I don't know how exactly it works, but with it at 0, I find the field "opens up" more as more timings and pitch locations stay in play unless poorly hit or too far off. Then, you get fouls or misses (based on hitter Vision and where the pitch vs swing is physically, I'm guessing).

On the other hand, more of those poorer hits will also stay in play. A grounder near the 3B line might shoot past, barely stay fair at 0 for a double...as would a 3-hop chopper that he can charge and throw out the batter.

That's what it seems to do based on "eye test". Like I said, I know of no inner workings details.

[statdude]

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

Has anyone had an injury occur in a CPU vs CPU game yet?

[nomo17k]

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

Has anyone had an injury occur in a CPU vs CPU game yet?

I explicitly turn off injuries for testing (to avoid losing too many key players... though I guess I could test the Injuries slider had I used manual injuries), so I don't know if it has occurred or not.

But aside from bugs it should occur. Last year I tested the Injuries slider for a while and it worked as advertised.

[CoachCraven]

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

Has anyone had an injury occur in a CPU vs CPU game yet?

I've had maybe two or three in mine.

[NolanRyansSnowmonkey]

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

My fielder reaction is 1 right now, fielder speed 0. I contemplated moving reaction to 2, but after some more games, both me and CPU had lower GB BABIPs so I'm staying at one. Having some bloopers, flares, and "funky" hits find grass is nice, not to mention not tracking down every foul pop. Also makes Reaction and Speed ratings more important/highlighted. I see a real difference between guys that rob hitters blind or range all over the place and ones that can only "turn and chase" because they know they can't get there at all.

Yeah i originally had fielder speed at 2 with Contact 5, Power 5, and Solid Hits 3, and my Babip started out ok but over time it got lower and lower, down to .260 or so. My ground ball average got down to about .170 even on fielder speed 2. (Keep in mind I play on Legend). Line drives were at 20%. Home runs were ok, but the ball seemed a bit water-logged on a lot of hits.

I considered lowering fielder speed to 1 or 0, but still had the power problem. So I tried raising only Contact to 6. My Babip went up, and balls were getting through the infield better, but home runs dropped (as we discussed earlier). After many games with only 1 combined home run, I moved power up one notch. Now my ground ball average was about .330, lol. And for some unknown reason my Line Drive % has hovered around 16%. But home runs were near perfect.

So I raised fielder reaction from 2 to 3. With still too many ground balls getting through, I raised it to 4 which brought ground ball average to about .235.

After raising fielder speed notch by notch to nerf doubles, I ended up at default 5.

Everything now lined up, with only 3 exceptions. Batting average was a bit low, BABIP a bit low, (back to too many fly balls and not enough ground balls) and line drive rate at 16%.

I tried bringing up line drive % by raising solid hits to 4, which worked, but Home Runs dropped a lot. So back to 3.

So, full circle, I may have to drop Fielder Reaction back to 3 to allow more ground balls through and just live with line drives at 15%. I may even have to try dropping fielder speed to 4, but for now enough weak fly balls are still finding grass.

So, that's the only reason I have had to keep fielder speed higher. 5,5,4,3,3 with fielders at 2,2 was playable but numbers were off. With Contact and power raised I've had to bring Fielder sliders up more than I normally would. It hasn't looked bad so far, since more balls are hit harder, but we will see after more games.

The other thing is, can I keep CPU offense potent enough on Legend to overcome Fielder speed and reaction?

[KBLover]

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

Everything now lined up, with only 3 exceptions. Batting average was a bit low, BABIP a bit low, (back to too many fly balls and not enough ground balls) and line drive rate at 16%.

As a Directional player, I'm starting to influence down constantly unless in certain situations (sometimes with a pure power hitter, sac fly situation, must avoid DP situation like 1 out, 2 on, and I need runs).

That's helped dramatically increase my LD on average. It's still varying, but I've gotten as high as 30% LD rate (even if a lot of them find gloves lol - funny how that happens).

I don't know how to translate that to a Zone player, let alone one on Legend (Legend Directional sounds like a masochistic exercise). Maybe keep PCI high? Keep it above the ball as much as possible? I don't know.

Granted, I did have solid hits quite high (8) and I'm reluctant to lower it. For one, I won't always influence down and I don't want to invite pops and choppers to happen too much. For two, solid hits impacts carry, so it's keeping the non-HR extra bases good.

HR/H have been solid. HR/FB are a touch low, but reasonable, especially given that this team only has Gallo and Sano for real power threats. LD rate 23% (or 27% if you exclude fouls...not sure how it's done in real stringing/categorizing so I track both ways). LD BABIP is low, my GB BABIP is low, FB BABIP is a couple notches higher (.170 vs .130 MLB), overall right at average with .300 BABIP.

So far, I'm happy. Some games make me question it (one time scored 12 on 17 hits, last game 4 runs, 4 hits, .150 BABIP), but that's probably just baseball

[NolanRyansSnowmonkey]

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

Im noticing something that maybe has been in the game for years but im not sure. It's the fact the PCI is not where I put it each time. I kept noticing that I would be set up high and away for example, then I would hit a fly ball on a ball - below where I placed my pci. I would see the feedback and think, man how did I move my pci that much just pushing the swing button?? I thought it was just me accidentally moving it.

I noticed it happening opposite way also - low pci but hit a ground ball on a pitch above the pci.

I notice it most on 2-0 and 3-1 pitches where I know exactly where my pci is - dead middle. When the pitch is dead middle and you swing perfectly and chop it to the pitcher, and the feedback shows your pci in the upper third of the zone, when you didn't touch the stick at all, you have one angry snowmonkey.

I went into a human vs human game and threw fastballs down the middle to myself using classic for about 140 pitches and hit with the other controller without ever touching the stick. The pci feedback shows the pci moving up and down and left and right with no influence of the stick, and seems totally random which direction it moves (toward and away from the ball)

I thought maybe early swings moved it one way, and late another, but it's not the case. I had it move both directions on both.

I thought maybe it's because I'm on legend so I switched to beginner and saw the exact same thing, it's just more noticeable on legend due to the much smaller pci.

Has anyone ever noticed this before? I know KB plays directional but im just confused on how im supposed to hit if its doing this. Im still on 1.07 by the way. I've never noticed this in the Show but maybe I've just been blind to it. Either way, doesnt make much sense to me. Maybe there is a good reason I'm not aware of...

I think the patch has something to do with pci but not sure if it addresses that..

[nomo17k]

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

Im noticing something that maybe has been in the game for years but im not sure. It's the fact the PCI is not where I put it each time. I kept noticing that I would be set up high and away for example, then I would hit a fly ball on a ball - below where I placed my pci. I would see the feedback and think, man how did I move my pci that much just pushing the swing button?? I thought it was just me accidentally moving it.

...

The PCI movement due to timing has been in the game for years, and works exactly the way bcruise in another thread (v1.08 patch related). If your swing timing is late, then the PCI tends to appear down from where you placed it. If your swing timing is early, then it tends to appear up from where you placed it.

That is because bat-ball contact occurs in the 3D space, but the PCI only shows the 2D slice of place where contact actually happened. Since the swing plane tends to be that of slightly upper swing, when you are early on a pitch, you tend to swing over it, often producing ground balls to the pull side. When you are late on a pitch, you tend to swing under it, often producing fly balls to the away side.

But I think the PCI moves a little wonky at time in this years game... for example, our ability to move PCI pre-pitch is limited to the strike zone, but depending on swing timing, the swing feedback often shows PCI well outside the strike zone. I am not sure if this is meant to be due to the devs removing the PCI movement "lock" once the swing starts (that is what the devs indicated in a pre-release stream, I think?), or some sort of bug. It actually moves so freely people with some skills online rarely strikes out this year.

Zone hitting this year feels very arcadey.

[nomo17k]

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

Trying to decide if I should just not worry about batted ball types at all in my human vs CPU set. I know the main thing is the other numbers and averages, but sure would be nice to get batted ball types closer (I know i can get them closer than CPU vs CPU because of where I place my pci). The bad thing about them not lining up is you have to find a way to get the ground ball batting average to be higher, since there generally won't be as many ground balls hit in real life. So more balls have to get through the infield in order to get babip in the correct range.

I guess that's the main thing that's bothering me is all the seeing-eye singles and other ground ball hits. And line drives are at about 16% instead of 20% (even when I score them how i understand them).

All other numbers look pretty good right now. I think if i could just get line drives to 20% i can live with ground ball and fly ball ratios being off.

I've only played 9 games with this exact set but the numbers are:

(Human hitting only)
AB. 33.7 (LG Avg 34.1)
R. 4.44 (4.48)
H. 8.44 (8.71)
RBI 4.33 (4.27)
BB 2.8 (3.11) in real games im a bit more patient than when im testing but still close. Easy fix.
SO 8.00 (8.03)
1B. 5.67 (5.53)
2B. 1.33 (1.70)
3B. 0.00 (0.18)
HR 1.44 (1.16)
HBP 0.56 (0.34)
SF 0.33 (0.25)
GIDP 0.67 (0.77)
Avg .247 (.255) too low...
Babip .279 (.300) Too low...
SO% 21.2 (21.1)

One game had 0 doubles and 4 HR so without that game the numbers are almost exact.

I know it's only 9 games and at this point the numbers can change very quickly but so far so good.

I'm trying to figure out how to get my babip up to .300 without throwing everything off. I may have to lower fielder reaction but there goes the ground ball average even higher lol.

The odd thing about this set is it was just an experiment in foul frequency. I tried it at 0, 5, and 10 just to learn something about it. (I haven't learned anything much lol) I "might" can see some difference but not even sure. But somehow with foul frequency at 10 (which I would have never tried with an actual set) the numbers started falling into place close to what they should be.

Is this coincidence? Is there anything unrealistic or off by having it at 10? Ive never seen any slider set do this. Not even sure what it's accomplishing. I'm assuming it's bringing babip up a small amount due to less weak hits landing fair? I really don't like being super far off of default on hitting sliders but not sure why I even feel that way.

Your numbers look awfully good, especially given that you use Zone. Given that they are from only 9 games, even those numbers you say low are actually all within "margin or error", so it's too early to make any conclusion.

If I were you, I'd just continue with the same setting for a while to see if any issue remains over a more extended time.

I think one thing the Fould Frequency slider does is to work as a modifier to Vision ratings. If you set it very high, it's very likely that it is helping you to have more deeper counts, which means both walks and strikeouts can potentially increase.

[NolanRyansSnowmonkey]

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

Your numbers look awfully good, especially given that you use Zone. Given that they are from only 9 games, even those numbers you say low are actually all within "margin or error", so it's too early to make any conclusion.

If I were you, I'd just continue with the same setting for a while to see if any issue remains over a more extended time.

I think one thing the Fould Frequency slider does is to work as a modifier to Vision ratings. If you set it very high, it's very likely that it is helping you to have more deeper counts, which means both walks and strikeouts can potentially increase.

Thanks, yeah doubt the numbers will stay that good but here is to hoping.

My fouls per game, and fouls per swing are just about spot on. My fouls per pitch are just a bit high due to me swinging a little too much (51% of pitches instead of 46%) which comes down to not being quite as patient when testing game after game.

[NolanRyansSnowmonkey]

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

The PCI movement due to timing has been in the game for years, and works exactly the way bcruise in another thread (v1.08 patch related). If your swing timing is late, then the PCI tends to appear down from where you placed it. If your swing timing is early, then it tends to appear up from where you placed it.

That is because bat-ball contact occurs in the 3D space, but the PCI only shows the 2D slice of place where contact actually happened. Since the swing plane tends to be that of slightly upper swing, when you are early on a pitch, you tend to swing over it, often producing ground balls to the pull side. When you are late on a pitch, you tend to swing under it, often producing fly balls to the away side.

But I think the PCI moves a little wonky at time in this years game... for example, our ability to move PCI pre-pitch is limited to the strike zone, but depending on swing timing, the swing feedback often shows PCI well outside the strike zone. I am not sure if this is meant to be due to the devs removing the PCI movement "lock" once the swing starts (that is what the devs indicated in a pre-release stream, I think?), or some sort of bug. It actually moves so freely people with some skills online rarely strikes out this year.

Zone hitting this year feels very arcadey.

Yeah i thought it might be due to timing. That would make sense. But it does it on perfect timing also. And like I said it moved down with late timing, and up on the next pitch with late timing. I swung at 4 consecutive middle middle pitches, taking note of good timing, without touching the stick, and I got a pop up, a grounder, a chopper, and line drive. It doesnt take much with a relatively small pci.

Like you said it's a bit jumpy because I've played hundreds of games on zone on previous versions but it's never been a factor. I found more complaints on a different forum about this same issue of it moving the wrong way.

I may have to increase Contact just to make up for this. I've been wondering why it's felt so inconsistent with good swings. It's very noticeable on Legend. I just love legend because it's the only way I've found to get realistic user strikeouts.

[KBLover]

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

I may have to increase Contact just to make up for this. I've been wondering why it's felt so inconsistent with good swings. It's very noticeable on Legend. I just love legend because it's the only way I've found to get realistic user strikeouts.

Keep in mind that there are degrees of "Good" timing.

Anything in that green area on the timing graphic is "Good" but there's a "Perfect" spot in there too. Anything else is technically Early or Late. "Just Early" and "Just Late" are on the border of the green and yellow wedges.

So, and perhaps especially on Legend, if this system is very sensitive, even "Good" timing might cause it to move.

[NolanRyansSnowmonkey]

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

Keep in mind that there are degrees of "Good" timing.

Anything in that green area on the timing graphic is "Good" but there's a "Perfect" spot in there too. Anything else is technically Early or Late. "Just Early" and "Just Late" are on the border of the green and yellow wedges.

So, and perhaps especially on Legend, if this system is very sensitive, even "Good" timing might cause it to move.

Very true. Thats a good point. Slightly early "good" could cause it to move in an opposite direction than slightly late "good timing".

Still don't know why one late swing would move the pci up on one pitch and down on another. And vice versa. I do know it seems way off compared to previous years.