Stats-Based CPU Sliders [MLB 14 Version]

[kcsam]

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

Hey Nomo, where in your spreadsheet do you list out ERA? I must be missing it.

[nomo17k]

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

Quick question, any I may have missed this in an earlier post, but what skill level are you using in the game settings for these sliders or does that even matter?

The difficulty level does not matter in CPU vs. CPU games. That's what has been communicated by the devs in the past, and I don't think I have ever saw the difference when I tried to use different difficulty levels. I leave them at All-star though.

Alright, well, let me ask it to you like this: As of right now, if you were going to fire up a franchise with any of the sliders you have developed thus far, which ones would you run with? I'm asking because I value your opinion and the data you've collected. To me, it looks like the 4/26 set may provide a slightly better result based off real MLB averages. Would you agree?

I'll just have a "currently recommended" set in the first post then. I don't always run with what I think I should be using, I often use a slider value one above or below (or even at extremes) where I think it should be, just to look for the effects the slider change causes. It's easier to find effect when you vary a slider to an extreme value.

And lastly, what have you found with the Stolen Base slider set at 0 in your 4/26 set? I tried a game last night with Stolen Base Frequency and Success maxed out and no one attempted a SB in the first 4 innings with 5 base runners, so not sure what exactly those sliders affect. Guess I need to have a much greater sample size than 4 innings.

Things like SBs that happen just once or none at all for single game tends to be tougher to nail down because of the sample size and the nature of how the events happen...

I'm still confused by the SB aspects of the game. I'm setting the slider to 0 to see if it is reversed (as it has been in the past), and I am not seeing that much difference from when I was using 6 for the slider. The overall success rate has improved a bit, but not as much as I need to bring it to the MLB ave (about 73%). So, unfortunately after this set is done I have to set it to 10 to see if that actually brackets the slider setting that I should converge to.

That would only happen if the slider is still reversed though. It's unfortunate one of the harder sliders to nail down might be reversed. It takes far longer to get data I need.

Hey Nomo, where in your spreadsheet do you list out ERA? I must be missing it.

ERA should come out pretty close to the average run scored. So you should just look at the column "R" or "RBI" for an estimate. Things like ERA should just come out correct if most other stats are in line with the reality.

[kcsam]

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

This is a placeholder for an upcoming thread talking mostly about how to brew coffee so that it doesn't taste too sour. Another focus will be on what fast food chain cooks the best-tasting french fries. There will be occasional tangent on MLB 14 The Show sliders as well.


The on-going data collection can be followed here:

https://docs.google.com/spreadsheet/...=sharing#gid=2


Meanwhile, here is the "currently recommended" slider set:

This set is based on the collected data through April 26. Everything is at 5 (default) except:

Foul Frequency: 3
Solid Hit: 7
Starter Stamina: 7
Pitcher Control: 3
Pitcher Consistency: 3
Strike Frequency: 4
Pitch Speed: 5
Fielding Errors: 7
Throwing Errors: 3
Fielder Run Speed: 3
Fielder Reaction: 4
Fielder Arm Strength: 4
BR Steal Ability: 5

Note that I am setting Pitch Speed and BR Steal Ability both at 5, because the steal aspect of the game has not been looked at thoroughly yet. I'm just hoping the game yields realistic steal rate and success rate at default at least. With Pitch Speed at 10 which I usually use for realism, this has not been the case.

What made you decide to drop Power from 6 to 5? Based off your data from 4/26, it would appear the move up to 6 was more in line with the MLB HR totals than when you had at 5 on the 4/20 set. Maybe I'm missing something, but thought I'd mention it.

[nomo17k]

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

What made you decide to drop Power from 6 to 5? Based off your data from 4/26, it would appear the move up to 6 was more in line with the MLB HR totals than when you had at 5 on the 4/20 set. Maybe I'm missing something, but thought I'd mention it.

I'm actually looking at HR/H (i.e., what fraction of safe hits become HRs), because per game stats can sometimes be misleading.

I increased Power because it was down quite a bit for the 4/20 set, but HR/H = 8.3% (11.1% in MLB), but now with Power 6 it's a bit inflated (13.3%). I couldn't really believe lowering Strike Frequency by one reduced HRs that much (yes, that was effectively the only change I was introducing from the set prior to it), so I'm just guessing it was a very unlucky set where I saw less HRs.

It's sometimes interesting how the pitcher command related sliders don't yield effects that we expect from what they seem to do.

[rjackson]

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

Sorry this is off-off-topic but real quick: pitch type order--is X the most thrown or the highest quality pitch in a pitcher's arsenal? Seeing a trailer video, Bumgarner had his slider attached to X but it is his K pitch, not the most thrown so it got me wondering. If anyone not a dev knows, it is Nomo.

Just off-topic: I used these sliders last year in HUMvCPU playing and it wirked perfectly. It made me not play more like the CPU but made me play BETTER and more realistically. I became more patient but more aggresive in middle-middle pitches.

On topic: the best pizza I ever had was at Charley's in Orvietta, Italy. A lit of not good pizza in Rome, but this place was unbelievably good. Great gelato place less than a block away as well.

[nomo17k]

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

Sorry this is off-off-topic but real quick: pitch type order--is X the most thrown or the highest quality pitch in a pitcher's arsenal? Seeing a trailer video, Bumgarner had his slider attached to X but it is his K pitch, not the most thrown so it got me wondering. If anyone not a dev knows, it is Nomo.

Actual numbers of pitches thrown by Bumgarner (just for the primary/secondary):

37/29
28/39
26/37
34/18
25/59
26/44

So it looks like it's not just the "highness" in the pitch selection that determines how frequently a pitch is thrown. From the stats it appears more of the time Bumgarner threw the secondary more than the primary, but not always. It looks fairly dynamic.

I think breaking pitches, because they are harder to command and therefore harder to buildup in-game pitch command, tend not be used as often as fastballs, given all other things equal.

Just off-topic: I used these sliders last year in HUMvCPU playing and it wirked perfectly. It made me not play more like the CPU but made me play BETTER and more realistically. I became more patient but more aggresive in middle-middle pitches.

On topic: the best pizza I ever had was at Charley's in Orvietta, Italy. A lit of not good pizza in Rome, but this place was unbelievably good. Great gelato place less than a block away as well.

I wish I could go visit Europe. I need some change of scenery......

Anyways, not very surprising there are not many good pizza in Rome, just as how pizza/bagels in NY are often overrated.

But I can definitely say the best Mario the Plumber that I have seen is Italian, and he looked like this:

[rjackson]

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

Very good points on pitch selection. As for fries, KFC wedges are the best of all the fries but are underrated as they tend to come with chicken and not burgers.

[bcruise]

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

Great to see this thread finally getting back on topic.

I'm a fan of curly fries, myself.

[nomo17k]

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

Very good points on pitch selection. As for fries, KFC wedges are the best of all the fries but are underrated as they tend to come with chicken and not burgers.

I agree KFC wedges are underrated. KFC does spices very well, so that applies to their fries as well. Unfortunately I tend to go with mushed potatoes as a side at KFC. That's another factor. They have better side dish offerings.

Great to see this thread finally getting back on topic.

I'm a fan of curly fries, myself.

Not all curly fries are created equal though...



Off topic, it looks like the BR Steal Ability slider is still reversed. It's after only about 30 games, but reversing the trend will take unusually lucky run of successful steals.

Unfortunately, it also looks like the slider doesn't give us a dynamic range in SB% necessary to correct the base stealing issues. Too many base stealers (or failed hit & run attempts) are still resulting in caught stealings. So at the max Pitch Speed, we may be out of luck as far as getting SB% correct this year.

Things might be different at slower pitch speed, but it would still take a 25% increase in SB% rate, and I'm not sure lowering Pitch Speed can account for that much change in SB%. In my past experiments, I don't think varying Pitch Speed caused that much change.

So we might be out of luck with base stealers this year. Or perhaps should add the info to the reporting and hope another fix gets in for a potential patch.

[seanjeezy]

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

Sorry this is off-off-topic but real quick: pitch type order--is X the most thrown or the highest quality pitch in a pitcher's arsenal? Seeing a trailer video, Bumgarner had his slider attached to X but it is his K pitch, not the most thrown so it got me wondering. If anyone not a dev knows, it is Nomo.

Actually, Bumgarner's most used pitch over the last 3 years on Fangraphs (and therefore MLB Gameday) is his "slider". In the grand scheme of things order isn't that important though, the CPU will use the pitches it has the most confidence in.

[kcsam]

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

I agree KFC wedges are underrated. KFC does spices very well, so that applies to their fries as well. Unfortunately I tend to go with mushed potatoes as a side at KFC. That's another factor. They have better side dish offerings.





Not all curly fries are created equal though...



Off topic, it looks like the BR Steal Ability slider is still reversed. It's after only about 30 games, but reversing the trend will take unusually lucky run of successful steals.

Unfortunately, it also looks like the slider doesn't give us a dynamic range in SB% necessary to correct the base stealing issues. Too many base stealers (or failed hit & run attempts) are still resulting in caught stealings. So at the max Pitch Speed, we may be out of luck as far as getting SB% correct this year.

Things might be different at slower pitch speed, but it would still take a 25% increase in SB% rate, and I'm not sure lowering Pitch Speed can account for that much change in SB%. In my past experiments, I don't think varying Pitch Speed caused that much change.

So we might be out of luck with base stealers this year. Or perhaps should add the info to the reporting and hope another fix gets in for a potential patch.

What are you setting base stealing at now after your latest findings? I'm still playing with pitch speed at 5 like you suggested in your last slider update. It would be great if they add a patch for this. I don't recall if there's a report out there yet for it.

[Heroesandvillains]

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

What are you setting base stealing at now after your latest findings? I'm still playing with pitch speed at 5 like you suggested in your last slider update. It would be great if they add a patch for this. I don't recall if there's a report out there yet for it.

It seems to me based on his data that even if the slider is reversed, and even if you were to play with a zeroed out pitch speed, steal success is going to be low this year (without another patch) no matter what you do.

Nomo's showing a 53% success rate on his 04/26 set. And I think this is his best if I'm not mistaken.

[nomo17k]

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

What are you setting base stealing at now after your latest findings? I'm still playing with pitch speed at 5 like you suggested in your last slider update. It would be great if they add a patch for this. I don't recall if there's a report out there yet for it.

I've been testing BR Steal Ability at 0 and 10 to hopefully bracket the realistic SB% within that allowable range. With Pitch Speed at 10, that's simply not possible it seems.

I've never tested below the Pitch Speed below 5, so I don't know how much that would help. Who knows, decreasing the pitch speed that much might help some. (I doubt it, but it will be better.)

Another option is to increase BR Speed. But that has side effects on other results (infield hits and XBHs), so I'm not very positive about making that sort of change.

[kcsam]

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

I've been testing BR Steal Ability at 0 and 10 to hopefully bracket the realistic SB% within that allowable range. With Pitch Speed at 10, that's simply not possible it seems.

I've never tested below the Pitch Speed below 5, so I don't know how much that would help. Who knows, decreasing the pitch speed that much might help some. (I doubt it, but it will be better.)

Another option is to increase BR Speed. But that has side effects on other results (infield hits and XBHs), so I'm not very positive about making that sort of change.

It's worth a shot. Are you still running your test cpu games? If so, maybe you can drop pitch speed to 0 and then max out steal frequency and steal ability and see how it plays out. I agree with not fooling with base runner speed. That'll throw off a lot of other stuff.

[bcruise]

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

Got your sliders some nice exposure on twitch nomo. I played a franchise game on the PS4 with them and it was a thriller, crazy game and right down to the wire. And all the stats (save one, my throwing errors because of some difficulties with Pure Analog) ended up looking very realistic like it could actually happen in a Cards-Cubs game.

They really do play well human vs. CPU, just as they do CPU only.

Archive (yes 2 hours): http://www.twitch.tv/bcruise/c/4203262