Does Madden Cheat so CPU wins?

[roadman]

Quote: Originally Posted by Thunderhorse

And I am saying these momentum shifts can be logically explained when we witness them in the NFL, College football, etc. I maintain your reasoning here is lacking, because Madden's momentum shifts have no logical base in relation to the user's input throughout the game. But I like that you pointed out "similar". The use of that term is apt, considering Madden is similar to football but fails to actually be football. So I will concede I am debating something different than you are, and leave it at that.

Well, yeah, ok, when one says they agree to disagree, that usually means, let's leave it that because it isn't going anywhere.

I could also site there are players that play cpu vs cpu and or coach mode.

Also, I could also state what a 2k football developer once said, and that is the old line, once you put a controller in someone's hands, sim goes out the window.

[JayhawkerStL]

Quote: Originally Posted by Thunderhorse

And I am saying these momentum shifts can be logically explained when we witness them in the NFL, College football, etc. I maintain your reasoning here is lacking, because Madden's momentum shifts have no logical base in relation to the user's input throughout the game. But I like that you pointed out "similar". The use of that term is apt, considering Madden is similar to football but fails to actually be football. So I will concede I am debating something different than you are, and leave it at that.

Sigh.

It doesn't need the logical base you are looking for. The beauty of chaos theory and and how statistics work, all you need is the current system of dice rolls that affect outcomes over a large number of instances. I'm not kidding, that is all you need.

Tom Tippett was asked about adding clutch hitting and hot and cold streaks to Diamond Mind Baseball, probably the greatest and most accurate baseball sim ever developed. Seriously, based on the projections of each player, Tippett was able to use DMB to beat the vast majority of sports writers in predicting the upcoming season by playing it out 50 times and averaging the the results.

He refused to add clutch performance, and to have hot and cold performances. The problem was, without these artificially added results, the game would produce all the streaks and clutch hitting you see in real life. But if you added a boost, it would warp the hell out of the results. You would get more realistic variations in performance by just letting chaos dictate results than by trying to affect the results.

Leave it alone, and you get an accurate amount of no-hitters, hitting streaks, slumps, and magnificent comebacks. Yet, when any of these events happen to an individual, it feels scripted and wrong. It doesn't feel "random." But that's what "random" feels like. It's not an even distribution of events, its clumps of things happening and not happening.

Humans are TERRIBLE at picking up on randomness. We are hardwired to find and see patterns. We use those patterns to make decisions. Much of the modern world has been shaking off the misconceptions humans have of the world resulting from need to see patterns. Science overcomes this with data, using the scientific method to avoid getting trapped by anecdotal evidence.

Now, do I believe that modern action sports games on consoles are anywhere close to as accurate as text sims driven purely by data are? No. It's why I mostly roll my eyes at the insistence that these games be considered sims. They are so far off, it just seems like a silly argument. Fun>Realism.

But, that doesn't mean they are programmed to beat you. It just means that it will always feels that way. Humans are not capable of observing randomness in small samples. It always feels forced.

[NinersFan49]

Quote: Originally Posted by roadman

So, that could be similar to what the NFL experiences on Sunday's sometimes. Take for instance last weeks Browns vs Packers game. Browns, the worst team in the league, had a 21 to 7 lead going into the 4th quarter. The game started to guide towards the Packers direction in the 4th quarter and the game went into OT with the Packers prevailing and winning in OT. Same with the SB last year, a tale of 2 halves and close to 2.5 halves.

You're absolutely correct. The issue though, is whether certain events occur through dice rolls weighted by player rating differentials or by scripted outcomes. It's the way things play out on the field that is the most compelling, like for example slower players catching faster ones from behind, etc. Saying that the same thing has happened in the real NFL is irrelevant, the issue is random vs scripted.

[Thunderhorse]

Quote: Originally Posted by JayhawkerStL

Sigh. It doesn't need the logical base you are looking for. The beauty of chaos theory and and how statistics work, all you need is the current system of dice rolls that affect outcomes over a large number of instances. I'm not kidding, that is all you need. Tom Tippett was asked about adding clutch hitting and hot and cold streaks to Diamond Mind Baseball, probably the greatest and most accurate baseball sim ever developed. Seriously, based on the projections of each player, Tippett was able to use DMB to beat the vast majority of sports writers in predicting the upcoming season by playing it out 50 times and averaging the the results. He refused to add clutch performance, and to have hot and cold performances. The problem was, without these artificially added results, the game would produce all the streaks and clutch hitting you see in real life. But if you added a boost, it would warp the hell out of the results. You would get more realistic variations in performance by just letting chaos dictate results than by trying to affect the results. Leave it alone, and you get an accurate amount of no-hitters, hitting streaks, slumps, and magnificent comebacks. Yet, when any of these events happen to an individual, it feels scripted and wrong. It doesn't feel "random." But that's what "random" feels like. It's not an even distribution of events, its clumps of things happening and not happening. Humans are TERRIBLE at picking up on randomness. We are hardwired to find and see patterns. We use those patterns to make decisions. Much of the modern world has been shaking off the misconceptions humans have of the world resulting from need to see patterns. Science overcomes this with data, using the scientific method to avoid getting trapped by anecdotal evidence. Now, do I believe that modern action sports games on consoles are anywhere close to as accurate as text sims driven purely by data are? No. It's why I mostly roll my eyes at the insistence that these games be considered sims. They are so far off, it just seems like a silly argument. Fun>Realism. But, that doesn't mean they are programmed to beat you. It just means that it will always feels that way. Humans are not capable of observing randomness in small samples. It always feels forced.

Sigh,

New England beat Atlanta because they were the better team. There was nothing "random" about it.

And this thing about the CPU cheating is a non issue for me personally, because video game AI is not as creative as a human yet. Eventually the user will find the patterns and will exploit them if the AI does not have some kind of advantage to defend against that.

When it comes to one user against another user, we have a different issue at hand, and that is where my beef with random statistical dice rolls lies. It fails to respect one user's inputs in relation to the other's, and will actively reward one user who may not deserve to be rewarded because of some arbitrary formula put in place by the devs.

You seem to think I care about the computer cheating. Trust me, I could care less. I'm in the generation that grew up with MK I and II; That game cheated and it was great.

[NinersFan49]

The cheating matters to me to this extent:

A.If I'm in cvr1 or cvr0 and the cpu has a great wr vs my crappy cb and he beats him for a long TD, no problem.

B. If I have a great cb and the cpu has a crappy wr, who gets a really lucky dice roll and beats him for a long TD, no problem.

C. If the script says cpu will throw a long TD pass on this play, and makes it happen, that's a problem.

[JayhawkerStL]

Quote: Originally Posted by NinersFan49

The cheating matters to me to this extent: A.If I'm in cvr1 or cvr0 and the cpu has a great wr vs my crappy cb and he beats him for a long TD, no problem. B. If I have a great cb and the cpu has a crappy wr, who gets a really lucky dice roll and beats him for a long TD, no problem. C. If the script says cpu will throw a long TD pass on this play, and makes it happen, that's a problem.

To be clear, you are only making up C because you are butthurt over giving up a long TD. You have zero evidence that this is the case. Moreover, it really isn't the case.

Worse yet, this stupid, stupid rationalization doesn't even make you feel better. It just makes you bitter and unhappy. And then you run to the internet to spread it even further.

[JayCutlersCig]

Admittedly, the dice roll mechanics in M18, piss me off but also leaves me laughing at disbelief in the good luck sent my way. If they added better animation mechanics that logically support the dice roll, I’d be more supportive of it.


Sent from my iPhone using Operation Sports

[NinersFan49]

Quote: Originally Posted by JayhawkerStL

To be clear, you are only making up C because you are butthurt over giving up a long TD. You have zero evidence that this is the case. Moreover, it really isn't the case. Worse yet, this stupid, stupid rationalization doesn't even make you feel better. It just makes you bitter and unhappy. And then you run to the internet to spread it even further.

Thank you Dr Freud. Wow, must have struck a nerve there. Nope, all projection on your part. In A and B I said I didn't have a problem giving up the TD if it was randomized. And you don't have any evidence that C doesn't happen.

You've made a quick descent from high minded discussion of random statistical distribution to name calling.