An experiment on CPU offensive AI and sliders.....

[Brian SCEA]

Re: An experiment on CPU offensive AI and sliders.....

It just doesn't make any sense that on max legend I gave up a MAXIMUM of 4 runs and 7 hits with these experiments using the same techniques and concentration level that I used for HOF maxed sliders that I got killed on, absolutely killed.

I understand that the sample size isn't huge, but with similar experiennces from different users, i think something is going on in the legend difficulty. I could be wrong, but I don't wanna play on that level knowing that I also could be right.

anyways, this is no longer an issue for me on HOF level maxed sliders. ( I was just demolished by the arizona cardinals 12-2 after 6 innings using roy halladay). this after 5 or so games on maxed HOF sliders which I was demolished in 4 times.

On max legend sliders when I hit poorly I consistently threw 2 run 5 hit performances with Josh towers against the phillies offense.

anyways, thanks for the comments, and I highly appreciate your interaction with us.

As far as what you're seeing, just to verify you feel this is happening in Legend, but Hall of Fame is working just fine? I.e. that due to the way you're playing, that Hall of Fame is actually harder than Legend? Isn't this really the same conclusion as the guy who said that Rookie was harder than Hall of Fame? You yourself acknowledge it's not a lot of games either way.

Wouldn't you also agree that there is nothing I can say to convince you otherwise? If you read my posts, I haven't even disagreed with your conclusions - only talked about your method of experimenting.

All the experiments listed so far have one thing in common - everyone can get different results from them and draw conflicting conclusions. The experiments themselves could be more objective, because the experimenter/observer is the one generating all the data. Maybe you'll even confirm your previous results but find a different explanation for them.

In the Clever Hans example, even the most skeptical scientists who didn't believe the horse was psychic couldn't help but prove the horse was psychic due to the flawed experiment. It's a proven and repeatable phenomenon, and it's not a question of open mindedness. More to the point, everything you've described here is so subjective that no evidence to the contrary would be sufficient.

You've said this happens on Legend, but not on Hall of Fame. Wouldn't you agree there will definitely be people who feel this is happening in Hall of Fame? Wouldn't you disagree with them?

It's actually not really surprising for a skilled player to win several games in a row on Legend with max sliders. The game would be broken and predetermined if that didn't happen.

If there were 1 million people starting a season, some people are going to win 20 games in a row. Some could win 10 in a row, lower their difficulty level, and lose 10 in a row.

[Brian SCEA]

Re: An experiment on CPU offensive AI and sliders.....

Exactly, this thread has little to do with sliders and a whole lot of A.I. Could a mod move this back to the main page?

Thanks

Isn't this thread really about comeback code when using sliders?

[Brian SCEA]

Re: An experiment on CPU offensive AI and sliders.....

I understand you were commenting mainly on the experiment and thanks for taking time to read and comment on my post. I think the CPU batter AI is perhaps the biggest issue with the game. I think some of the batting AI problem is due to meter pitching, it is far too accurate. Therefore, I guess, you have to program the CPU to hit good pitches better than they should be able to since even poor control pitchers can be pretty accurate with the meter, especially with fastballs. The CPU doesn't swing and miss enough against pitchers that just have great stuff. If you throw a great curveball or slider and it gets a good chuck of the plate, the CPU is very likely to get a good rip at it. It is almost impossible to pitch well with Carlos Marmol using classic pitching, the CPU tears him apart because his location is obviously not good, but he has great stuff and shouldn't have to locate well. Play on Legend or HoF and try pitching with Marmol using classic pitching. Marmol gives up runs in real life not because he is hit but because he is wild. If Marmol throws strikes, batters just aren't going to hit him. In this game, you don't feel dominant pitching with a strikeout pitcher because the CPU hitting is too location based. So pitching with someone who has overpowering stuff just doesn't feel right in the game because you really can't overpower the CPU.

So you feel that it's not that the CPU chases too often, but hits too well when they chase? Particularly against the best pitches?

As for as the CPU pitching accuracy, we're obviously comparing against a wide range of user skill but there are definitely many users more accurate than the CPU. It would be more apparent if we showed where the CPU was originally aiming. We don't because it's not realistic.

[pberardi]

Re: An experiment on CPU offensive AI and sliders.....

What on earth are we talking about here?

If you pound the outer corner with your fastball and complain the cpu is hitting laser like bullets for fouls or whatever, then throw a changeup for g&*ds sake!

I have no problem striking the cpu out on HOF or getting them to chase. The cpu will start to time your fastball just like you would the cpu. Throw some high heat out of the zone once in awhile because it's easier to hit the low fastball.

You have to throw balls in this game and you have to change speeds. Yes sometimes the cpu will hit a pitch that's ridiculous but you also have to lose once in awhile too.

And stop hitting "x" to speed up the game. You tend to throw and not pitch when you do not take your time and think about your next pitch. If a hitter has two hits off you and there are men on base, pitch around him. I get more DP's this way.

Pitching is an art not a science.

That said, I make plenty of mistakes pitching.

[ParisB]

Re: An experiment on CPU offensive AI and sliders.....

What on earth are we talking about here?

If you pound the outer corner with your fastball and complain the cpu is hitting laser like bullets for fouls or whatever, then throw a changeup for g&*ds sake!

I have no problem striking the cpu out on HOF or getting them to chase. The cpu will start to time your fastball just like you would the cpu. Throw some high heat out of the zone once in awhile because it's easier to hit the low fastball.

You have to throw balls in this game and you have to change speeds. Yes sometimes the cpu will hit a pitch that's ridiculous but you also have to lose once in awhile too.

And stop hitting "x" to speed up the game. You tend to throw and not pitch when you do not take your time and think about your next pitch. If a hitter has two hits off you and there are men on base, pitch around him. I get more DP's this way.

Pitching is an art not a science.

That said, I make plenty of mistakes pitching.

Amen

The game plays just fine.

[Chef Matt]

Re: An experiment on CPU offensive AI and sliders.....

Amen

The game plays just fine.

Same here. I admit last year I had a few issues with how the game felt so I didn't play it. But I don't see any of this at all this year.

Mind you, it was really going to be hard for me to get over that "gun shy" feeling of scripted gameplay, but I can officially say I see nothing of what I at least thought I saw last year.

Allstar Default. btw.

[brewersfan08]

Re: An experiment on CPU offensive AI and sliders.....

Really? This is good news...so everything set on default all-star? I also have pitch confidence turned off btw

[Chef Matt]

Re: An experiment on CPU offensive AI and sliders.....

Really? This is good news...so everything set on default all-star? I also have pitch confidence turned off btw

Yes default All-Star. Here are a few of my last scores.

3-0 Shutout win
7-1 drubbiing win
5-6 loss
13-5 loss this one killed, they ht everything.
6-2 win
6-1 win
8-3 loss

As you can see all varied scores.

I get a perfect amount of all stats, HR, 2b, BB, K's etc.. It really plays perfectly for me.

And as you can see I had a shutout and a nice 7-1 game I lead the entire way without a single threat. That honestly never happened for me last year. Not one game did I feel comfortable with a substantial lead in '09.

And in that 7-1 game after I took the big lead I just pitched to contact and they didn't threaten to take over at all. It was a great game

[baseball66]

Re: An experiment on CPU offensive AI and sliders.....

Yes default All-Star. Here are a few of my last scores.

3-0 Shoutout win
7-1 drubbiing win
5-6 loss
13-5 loss this one killed, they ht everything.
6-2 win
6-1 win
8-3 loss

As you can see all varied scores.

I get a perfect amount of all stats, HR, 2b, BB, K's etc.. It really plays perfectly for me.

And as you can see I had a shutout and a nice 7-1 game I lead the entire way without a single threat. That honestly never happened for me last year. Not one game did I feel comfortable with a substantial lead in '09.

And in that 7-1 game after I took the big lead I just pitched to contact and they didn't threaten to take over at all. It was a great game

Are you using classic or meter pitching?

[Chef Matt]

Re: An experiment on CPU offensive AI and sliders.....

Are you using classic or meter pitching?

I use meter. I tried classic in a few exhibition games because alot of the Vets around here swear by it but I got hosed badly. So I went back to meter. And I use no guess pitch, and no vibration or ball marker for pitching.

[baseball66]

Re: An experiment on CPU offensive AI and sliders.....

I use meter. I tried classic in a few exhibition games because alot of the Vets around here swear by it but I got hosed badly. So I went back to meter. And I use no guess pitch, and no vibration or ball marker for pitching.

Ok, thanks. I would like to stick with default but i use classic pitching and that's a no go because classic is more erratic compared to meter.

To compensate i have upped my control and consistency sliders.

I'm going to try and keep the sliders the same for both me and the cpu so as to not cheat in eithers favor. I'll see how that plays out. The global sliders shouldn't matter in that regard as they affect both teams equally.

Ok, lets play some Baseball.

[Phoenixmgs]

Re: An experiment on CPU offensive AI and sliders.....

So you feel that it's not that the CPU chases too often, but hits too well when they chase? Particularly against the best pitches?

As for as the CPU pitching accuracy, we're obviously comparing against a wide range of user skill but there are definitely many users more accurate than the CPU. It would be more apparent if we showed where the CPU was originally aiming. We don't because it's not realistic.

The CPU hits fine when they do actually chase. They just don't chase enough. The CPU doesn't swing and miss enough on quality strikes either. The CPU does have times where it consistently hits well executed pitches on the corners too well. My theory is that is mainly due to the ability to locate too well with meter pitching, and if the CPU wasn't able to hit well located pitches hard, then meter pitching would just be too easy. I may be better with the meter than others, but several people post with the issue of not walking enough people with meter pitching. Power pitchers don't have to have great location to be successful in real life. However, in this game, the CPU hits pitches that catch a good amount of the plate consistently well no matter if it's a power pitcher or a 5th starter. A nasty breaking ball that's around the middle of the plate is still nasty. I don't think the batting AI properly takes in consideration how hard some pitches/pitchers are to hit. In my opinion, location plays too much a factor into how well the CPU hits. Don't get me wrong some pitchers don't have the stuff to pitch well without locating, but some pitchers do have to stuff to do that. Try pitching with a power pitcher that has great stuff but has less than average control (for example, Carlos Marmol) using classic pitching on Legend or HoF and you're just not going to pitch well in this game.

What on earth are we talking about here?

If you pound the outer corner with your fastball and complain the cpu is hitting laser like bullets for fouls or whatever, then throw a changeup for g&*ds sake!

You have to throw balls in this game and you have to change speeds. Yes sometimes the cpu will hit a pitch that's ridiculous but you also have to lose once in awhile too.

I'm fully aware of how to pitch. I pitch with classic pitching so I throw plenty of balls. I throw at most 2 fastballs in row. If I get up 0-2 with 2 fastballs, I'm throwing a waste changeup or something and then I come back with a high and tight fastball for the K.

[baseball66]

Re: An experiment on CPU offensive AI and sliders.....

Yup, it's almost like the ball needs to be a smaller object when the cpu swings against pitchers who have great swing and miss stuff (high k/9 ratio).

[ParisB]

Re: An experiment on CPU offensive AI and sliders.....

Power pitchers don't have to have great location to be successful in real life. However, in this game, the CPU hits pitches that catch a good amount of the plate consistently well no matter if it's a power pitcher or a 5th starter. A nasty breaking ball that's around the middle of the plate is still nasty. I don't think the batting AI properly takes in consideration how hard some pitches/pitchers are to hit. In my opinion, location plays too much a factor into how well the CPU hits. Don't get me wrong some pitchers don't have the stuff to pitch well without locating, but some pitchers do have to stuff to do that. Try pitching with a power pitcher that has great stuff but has less than average control (for example, Carlos Marmol) using classic pitching on Legend or HoF and you're just not going to pitch well in this game.

This post isn't about the experiment, but just wanted to comment as part of the discussion I politely completely disagree with that statement(s). Major league hitters punish anything over the plate, and can adjust to it after facing the pitcher a couple of times. No pitcher will survive long if he can't locate. The nastiness is only effective when he's fresh on the scene and until hitters start getting more comfortable and figuring him out by laying off the junk and wait for the pitch over the plate to punish it.

And I also think you have some interesting expectations from Marmol. He's not the unhittable machine you seem to perceive him as. In real life when he's over the plate, he gets hit and hit hard, just like any major league pitcher. K-Rod is another good example. When he first came onto the scene he was electric with his nasty slider and fooling hitters left and right. They eventually figured out to lay off the nasty and wait for him to come over the plate and he went through a transitional period where he struggled and his nasty slider would get rocked when he came over the plate and couldn't pinpoint it on the corners.

[Phoenixmgs]

Re: An experiment on CPU offensive AI and sliders.....

This post isn't about the experiment, but just wanted to comment as part of the discussion I politely completely disagree with that statement(s). Major league hitters punish anything over the plate, and can adjust to it after facing the pitcher a couple of times. No pitcher will survive long if he can't locate. The nastiness is only effective when he's fresh on the scene and until hitters start getting more comfortable and figuring him out by laying off the junk and wait for the pitch over the plate to punish it.

And I also think you have some interesting expectations from Marmol. He's not the unhittable machine you seem to perceive him as. In real life when he's over the plate, he gets hit and hit hard, just like any major league pitcher. K-Rod is another good example. When he first came onto the scene he was electric with his nasty slider and fooling hitters left and right. They eventually figured out to lay off the nasty and wait for him to come over the plate and he went through a transitional period where he struggled and his nasty slider would get rocked when he came over the plate and couldn't pinpoint it on the corners.

I don't expect Carlos Marmol to be unhittable in the game. I do expect him to be hard to hit. Here are his H/9 and K/9 stats in the last 3 years:

2007 - H/9: 5.3 - K/9: 12.5
2008 - H/9: 4.1 - K/9: 11.7
2009 - H/9: 5.2 - K/9: 11.3

I've traded Marmol the past 2 years in this game because the CPU unrealistically hits him hard. Marmol is a pitcher that can catch a good chunk of the plate and have good hitters swing and miss. If Marmol gives up runs, it's because he is wild not because he gets hit.

Check out this site:

http://baseballanalysts.com/archives/2010/02/there_are_two_t_1.php

The pitchers in that power pitcher chart CAN throw the ball by you without having good location. A power pitcher can dominate a good hitting lineup; just look at playoff games where guys like Josh Beckett dominated the game. It wasn't because Josh was painting corners, it was because of his stuff.

Somebody like Mark Buehrle has to locate to dominate. And, the average pitcher has to locate to pitch well. But there are those few pitchers that can just throw the ball by MLB hitters, and those pitchers aren't represented well in this game.

Brandon Webb is another pitcher that fits into the power pitcher formula in a way. When Webb has his good sinker working, you're just gonna pound it into the ground. The batter knows it's coming and knows where he's gonna throw it but he just ain't gonna hit it hard.