Is online completely meaningless?

[Skilletor]

what do u mean "lol at people playing this offline" kid? ill smash u in mvc3, ttt2 vf5 or doa5 kid

Money match at EVO, kid? All of the above games, kid.

But my comment was more directed at the pithy turnouts this game is getting. Kid.

Of course it's hilarious that somebody who gets hit by DJ's laser 3 times thinks he could beat anybody at Tekken.

 

[Awesmic]

DJ? I thought that was Eddie Winslo-- I mean, Gordo. He's the only one I know that resembles that dude from Street Fighter.

Waitaminut... lasers? Ohhhh, you're talking about Super Emo. That winged wrist-cutter's ridiculous, but yeah, that laser can be sidestepped hella easy. And I only know two moves with Zafina.

 

[WebHead]

Money match at EVO, kid? All of the above games, kid.

But my comment was more directed at the pithy turnouts this game is getting. Kid.

Of course it's hilarious that somebody who gets hit by DJ's laser 3 times thinks he could beat anybody at Tekken.

fight me online boy. ttt2 and vf5 both have great netcodes. u scared broski

 

[shinryu]

That's a normative, epistemic question. It means whatever you make it mean. If you have fun online, it means it's entertaining. If you beat Master online, it means you are good at playing DOA on a level where you can't rely on not being buffer-trapped by lag. It's not the same thing as an offline match, as you have to worry about compensating for lag-related issues, whereas offline you can really focus on timing and free-canceling, etc. However, that doesn't mean it's a trivial chore playing online. It's simply a different challenge in certain regards. Beating Master online does not necessarily mean you could beat him offline, but that doesn't mean it "doesn't matter."

Sports analogy: You can beat Michael Jordan at basketball, if the court is outdoors, and the weather is shitty, and additionally someone shoots the ball out of the air randomly with a shotgun. This means you are awesome at outdoor shitty weather skeet shooting basketball. It's the most popular sport nowhere ever at anytime and totally counts! Get the fuck out of here.

You understand how I can't really take that seriously? This is a fucking fighting game, not Being and Nothingness. You're playing a version of the game where a completely random variable frequently decides between a successful counter or combo and disaster and where things that are drastically bad ideas in the game as designed are now actually better ideas than the strategies that are optimal in offline. If you think that's "beating" someone at anything you must think you can "beat" someone at roulette. You sure as shit ain't playing poker anymore online.

 

[Skilletor]

fight me online boy. ttt2 and vf5 both have great netcodes. u scared broski

Skillzilla81 on XBL or PSN. I have both games on both systems.

 

[Brute]

Sports analogy

If "victory" online were as dictated by luck and exogenous factors as you proposed, better players would have a very negligible advantage over worse players, indicating their win ratio would be nearly identical (being within at minimum one standard deviation). However, that's just not the case. I'm sure if you looked at Master, Mr Wah or Rikuto's online win rate, it would be higher than yours. Furthermore, if you challenged them to a few matches, more likely than not, they would consistently win those. While random factors do affect online play, it's not enough to entirely trivialize the match.

 

[Tones]

Screwing around online gives me ideas.

 

[shinryu]

If "victory" online were as dictated by luck and exogenous factors as you proposed, better players would have a very negligible advantage over worse players, indicating their win ratio would be nearly identical (being within at minimum one standard deviation). However, that's just not the case. I'm sure if you looked at Master, Mr Wah or Rikuto's online win rate, it would be higher than yours. Furthermore, if you challenged them to a few matches, more likely than not, they would consistently win those. While random factors do affect online play, it's not enough to entirely trivialize the match.

Ok, let me put it differently: assuming I am only average player, what would you posit my win ratio offline would be against them? How would you adjust that for online? Would you have to apply a giant random factor to that equation, probably? What if I've learned to lag abuse and they haven't/don't? You see the problem? In VF5FS online I have been beaten by people I can assure you would never touch me offline because of shit connections. I can hang pretty well with guys who won pools at Evo in that game, I am not scrubby. But since you can't reliably beat 2p with elbow in lag half the game goes to hell. It's the same thing here.

It's not at all that "luck" is the whole of offline play, it's that it's any of it. There's very, very little systemic randomness in (good) offline fighting games. Hasn't really been any since SFII. Once you break the rules of frame data the game isn't the game anymore, and lag breaks those rules inconsistently You wouldn't watch basketball where the ball got blown out of the air randomly and you wouldn't play poker where the decks randomly got extra cards added or pulled, would you? To use another sports analogy: online is not going from hardcourt tennis to Wimbledon, where your skills transfer over but the bounce is different and you must compensate. Online is going from hardcourt to a mud field in Serbia randomly strewn with old landmines. Your skills transfer over but holy shit your leg just got blown off sry ggpo.

 

[Brute]

Ok, let me put it differently: assuming I am only average player, what would you posit my win ratio offline would be against them?

Very low. Around 10%

How would you adjust that for online? Would you have to apply a giant random factor to that equation, probably?

Very close (within 1 S.D.). Perhaps a little higher. Around 15%.

What if I've learned to lag abuse and they haven't/don't? You see the problem?

They can and often do learn to adjust to the tragedies of online lag as well, despite being even better offline.

It's not at all that "luck" is the whole of offline play, it's that it's any of it. There's very, very little systemic randomness in (good) offline fighting games. Hasn't really been any since SFII. Once you break the rules of frame data the game isn't the game anymore, and lag breaks those rules inconsistently

Frame data isn't everything. If it were, Helena would suck. And she doesn't. There's loads of meta-game and "yomi" outside the frame data (which btw doesn't even tell which attack crush, can be delayed, etc.).
And the fact that "randomness is almost entirely absent in good, offline fighting games" is ridiculous. All fighters rely on you anticipating your opponent and behaving accordingly. That involves guessing, and when you guess (ie: no certainty) there is always some randomness involved.

You wouldn't watch basketball where the ball got blown out of the air randomly and you wouldn't play poker where the decks randomly got extra cards added or pulled, would you? To use another sports analogy: online is not going from hardcourt tennis to Wimbledon, where your skills transfer over but the bounce is different and you must compensate. Online is going from hardcourt to a mud field in Serbia randomly strewn with old landmines. Your skills transfer over but holy shit your leg just got blown off sry ggpo.

Those would arguably be the only times I'd watch those sports because they're fucking boring and seeing shit get turned upside-down would please the inner troll spirit of my being.
But please stop using "sports analogies." You have little concept of what makes an accurate analogy. I get the point you're trying to convey, but it's simply not indicative/comparable of the bullshit you propose in these "analogies."

 

[shinryu]

Very low. Around 10%


Very close (within 1 S.D.). Perhaps a little higher. Around 15%.


They can and often do learn to adjust to the tragedies of online lag as well, despite being even better offline.



Frame data isn't everything. If it were, Helena would suck. And she doesn't. There's loads of meta-game and "yomi" outside the frame data (which btw doesn't even tell which attack crush, can be delayed, etc.).
And the fact that "randomness is almost entirely absent in good, offline fighting games" is ridiculous. All fighters rely on you anticipating your opponent and behaving accordingly. That involves guessing, and when you guess (ie: no certainty) there is always some randomness involved.


Those would arguably be the only times I'd watch those sports because they're fucking boring and seeing shit get turned upside-down would please the inner troll spirit of my being.
But please stop using "sports analogies." You have little concept of what makes an accurate analogy. I get the point you're trying to convey, but it's simply not indicative/comparable of the bullshit you propose in these "analogies."

For the record I don't think this is worth taking to the PMs, since you deserve a public excoriation for the rambling pseudo-intellectual horseshit you're spewing. DOA players, look upon your fellows and cringe at our scene.

First off, I'm genuinely curious how you're estimating SDs for win chances, here. Do you actually mean that the point estimate for proportion of online wins for any given player would be within the 68% confidence interval for the point estimate of offline wins assuming a binomial distribution of wins and losses in both cases? You think these are going to be essentially identical distributions? That's fucking bold. I might advise you to back off that or back yourself up with some data and a model and not numbers out of your ass. For the record, I am proposing no such model; I am simply saying that any such model estimating win percentage through any method (simple binomial estimation, logistic regression, Monte Carlo estimation, whatever you like) would necessarily have an online-induced randomness component and the very existence of such a component is a problem. Furthermore, though this is at best a combination of personal experience and anecdotal evidence, there seems to be a preponderance of opinion that predisposes me to think such a component would be fairly large. But so far I have assumed you know what the fuck you're talking about with regard to statistics. I suspect that's being too kind.

Secondly: Learn to fucking read. I said "SYSTEMIC" randomness, that is to say randomness inherent to the system of the game. If I am at +4 after a guard break and I am Tina and I am offline I know for a fucking graven in stone god given fact that my 6p will beat any character's p if it comes out on first frame. This is written in the SYSTEM and is NOT RANDOM. I am making the bet that they will not mid punch hold, yes, but there is no IN GAME, IN BUILT, AS DESIGNED RANDOMNESS which would prevent my 6p from beating every character's p given the situation. p(my p wins over their p) = 1. Online, that is no longer true; AN ELEMENT OF SYSTEMIC RANDOMNESS INDEPENDENT OF PLAYER CHOICE IS INTRODUCED BY ONLINE PLAY. My 6p might or might not in fact come out on first frame, and I can no longer make informed decisions (or bets, if you prefer) about my play. p(my p wins over their p) = 1 - (online bullshit factor). Moreover, as I cannot be appraised of the lag situation moment to moment or assured of its consistency between connections, I cannot easily adapt my play to compensate and strategies that are suboptimal offline become much more viable. Clear enough? I bolded the IMPORTANT PARTS for you this time. You may find certain analogies between what I'm saying here and the influence of online in a hypothetical model in the first paragraph, assuming your hydrocephalus isn't acting up any worse than usual today.

Finally: If an analogy gets the point across, it's not a bad analogy. An example, SAT style: Helena is to unsafe as dipshit is to you. Think that one should be pretty clear, no?

 

[Brute]

For the record I don't think this is worth taking to the PMs, since you deserve a public excoriation for the rambling pseudo-intellectual horseshit you're spewing. DOA players, look upon your fellows and cringe at our scene.
First off, I'm genuinely curious how you're estimating SDs for win chances, here. Do you actually mean that the point estimate for proportion of online wins for any given player would be within the 68% confidence interval for the point estimate of offline wins assuming a binomial distribution of wins and losses in both cases? You think these are going to be essentially identical distributions? That's fucking bold. I might advise you to back off that or back yourself up with some data and a model and not numbers out of your ass.

That's my theory. But theory is hypothetical. So no, I don't have data to back that up. But from obersvation, good players offline do well online, too. That indicates that there is some correlation between the two. Being good offline makes you better online, so likely vice versa as well.

Secondly: Learn to fucking read. I said "SYSTEMIC" randomness, that is to say randomness inherent to the system of the game. If I am at +4 after a guard break and I am Tina and I am offline I know for a fucking graven in stone god given fact that my 6p will beat any character's p if it comes out on first frame. This is written in the SYSTEM and is NOT RANDOM. I am making the bet that they will not mid punch hold, yes, but there is no IN GAME, IN BUILT, AS DESIGNED RANDOMNESS which would prevent my 6p from beating every character's p given the situation. p(my p wins over their p) = 1. Online, that is no longer true; AN ELEMENT OF SYSTEMIC RANDOMNESS INDEPENDENT OF PLAYER CHOICE IS INTRODUCED BY ONLINE PLAY. My 6p might or might not in fact come out on first frame, and I can no longer make informed decisions (or bets, if you prefer) about my play. p(my p wins over their p) = 1 - (online bullshit factor). Moreover, as I cannot be appraised of the lag situation moment to moment or assured of its consistency between connections, I cannot easily adapt my play to compensate and strategies that are suboptimal offline become much more viable. Clear enough? I bolded the IMPORTANT PARTS for you this time. You may find certain analogies between what I'm saying here and the influence of online in a hypothetical model in the first paragraph, assuming your hydrocephalus isn't acting up any worse than usual today.

As I said, it is a different game. A more flawed one. And I never said randomness (hetereoskedasticity if we want to jerk off the intellectual crowd, though I don't really identify myself as an intellectual) plaws no part in hindering online. It does. I just don't think it renders it obsolete.

Finally: If an analogy gets the point across, it's not a bad analogy. An example, SAT style: Helena is to unsafe as dipshit is to you. Think that one should be pretty clear, no?

Well, somebody got mad. xD

 

[Yaguar]

No... because without it I can bet DOA would not have got even a fraction of the sales it did.

 

[shinryu]

It's a good thing you don't classify yourself as an intellectual.

1) Heteroscedasticity IS NOT randomness. Jesus wept. For reference: http://en.wikipedia.org/wiki/Heteroscedasticity. (Yes, you can argue that there is a different variance in outcomes from online play vs offline play, but this is exactly the opposite of what you're arguing in the first place. So it looks a lot like you used a big scary word and used it wrong, frankly).

2) If you're seriously proposing that an average player has a 10% chance in beating a top-tier player offline, the binomial probability of someone winning, say, 15 out of 100 games at 10% is about 4%. That's a significant deviation from expectation. Furthermore, if the point estimate of that probability is 10% from 100 offline games, the normally approximated 1 SD confidence interval for that prediction is +/- 3%. Putting aside possible inaccuracy from the approximation, 15% is just barely, barely sneaking in under the 2 SD 95% confidence interval (I'm going to assume you don't want a one-way confidence interval, and if so I'd love to see your logic there), definitely way outside 1 SD. The "model" you propose actually suggests that online has a suggestively significant effect on the odds of victory, also suggesting you pulled numbers out of your ass and/or don't know anything about statistics and are just spouting horseshit you don't understand. Either of those tend to piss me off and I suspect you're doing both.

 

[Brute]

It's a good thing you don't classify yourself as an intellectual.
1) Heteroscedasticity IS NOT randomness. Jesus wept. For reference: http://en.wikipedia.org/wiki/Heteroscedasticity. (Yes, you can argue that there is a different variance in outcomes from online play vs offline play, but this is exactly the opposite of what you're arguing in the first place. So it looks a lot like you used a big scary word and used it wrong, frankly).

Exactly my point. Heteroscedasticity does not play a large enough roll in affecting online play. In this case the difference of variables would be the "random" things you say invalidate online play, such as buff-traps, input lag, etc. My point is that good players find a way to adjust to these things while still implementing core mechanics

2) If you're seriously proposing that an average player has a 10% chance in beating a top-tier player offline, the binomial probability of someone winning, say, 15 out of 100 games at 10% is about 4%. That's a significant deviation from expectation. Furthermore, if the point estimate of that probability is 10% from 100 offline games, the normally approximated 1 SD confidence interval for that prediction is +/- 3%. Putting aside possible inaccuracy from the approximation, 15% is just barely, barely sneaking in under the 2 SD 95% confidence interval (I'm going to assume you don't want a one-way confidence interval, and if so I'd love to see your logic there), definitely way outside 1 SD. The "model" you propose actually suggests that online has a suggestively significant effect on the odds of victory, also suggesting you pulled numbers out of your ass and/or don't know anything about statistics and are just spouting horseshit you don't understand. Either of those tend to piss me off and I suspect you're doing both.

I said "maybe higher" because it's true that differences can account for more than that, especially if one is unadjusted to more to one style of play than the other. You're right. The numbers I provided are misleading. I apologize.
Of course I pulled numbers out of my ass. You asked me to "guess" in a hypothetical scenario. Where else could I pull them from? To my knowledge, no one has run any data on this. xD

 

[shinryu]

God, this is truly not worth the time I'm spending on it, but on the off-chance you're just young and bright, let me restate:

Pulling numbers out of your ass looks bad. If nobody's run data. say that first. My question was rather or not you would expect a large online component to inform your estimate, not whether or not you would put up exact numbers for such a model. Apologies if that was unclear, but in context I think it was fairly stated. You clearly do not expect a significant online component and I disagree. I would suspect most people would disagree, again entirely anecdotally.

Randomness is not heteroscedacticity. It isn't. Randomness is not dispersion, period. Don't say one is the other or people who know what they're talking about will call you on it, trust me on this one. Additionally, either the distribution of estimates for probability of victory in offline or online play are homoscedastic or heteroscedastic, depending on whether their variance is significantly different. But the random factor of online play is not necessarily influential on the variance of the distribution of the probability estimates of the outcome even if it is highly influential on the outcomes themselves we make those estimates from. It's entirely possible that the online/offline victory probability estimate distributions are homoscedastic but not identical if their means are significantly different, and our argument is more about a mean shift, ultimately.

I refer to the online component as "random" since moment to moment the effect is largely unpredictable to the player and again, metaphorically, makes the game more akin to roulette rather than poker. In the aggregated data, though, one could have a model taking something like connection bars into account as a factor variable (say, a logistic model regressing p = logit(player_skill_1 + player_skill_2 + connection_quality)) and that could produce a worthwhile model of the influence of the three variables. The question is whether or not the connection quality variable would be significant. Again, I strongly suspect it would, and it would not influence that estimate in a direction that would tend to weight player skill more highly stratifying on worsening connection strengths. If the point of the game is to demonstrate which player has better skill then that outcome would strongly suggest offline isn't "meaningful" for determining that sort of thing. I strongly suspect even an online skill variable would be less influential than the connection quality if we split "offline" and "online" skill components out for each player. To go a little reductio ad absurdum, play anybody in the world on 1 bar and see how well you think that represents your skill level or theirs. This is all hypothetical, but the mechanics of the situation in terms of the effects of lag suggest that this is the case, to the extent that skill is dependent on the lack of variability in the frame data (and it is dependent on that, whether or not you in fact intellectually know the precise frame data or merely that one move beats another). Hence, online play seems less meaningful.

This is getting deep into the statistical weeds, but this is a sore spot with me. TL;DR: Precision in terminology does count, and if we're going to take victory as indicative of skill then online play is almost necessarily less meaningful as a test for such skill.

 

[Skilletor]

I don't know nothing about no stats, but my girl is going through her PhD level statistics class and all of that sounds familiar.

Two points Shinryu.

 

[CrimsonCJ]

God, this is truly not worth the time I'm spending on it, but on the off-chance you're just young and bright, let me restate:

Pulling numbers out of your ass looks bad. If nobody's run data. say that first. My question was rather or not you would expect a large online component to inform your estimate, not whether or not you would put up exact numbers for such a model. Apologies if that was unclear, but in context I think it was fairly stated. You clearly do not expect a significant online component and I disagree. I would suspect most people would disagree, again entirely anecdotally.

Randomness is not heteroscedacticity. It isn't. Randomness is not dispersion, period. Don't say one is the other or people who know what they're talking about will call you on it, trust me on this one. Additionally, either the distribution of estimates for probability of victory in offline or online play are homoscedastic or heteroscedastic, depending on whether their variance is significantly different. But the random factor of online play is not necessarily influential on the variance of the distribution of the probability estimates of the outcome even if it is highly influential on the outcomes themselves we make those estimates from. It's entirely possible that the online/offline victory probability estimate distributions are homoscedastic but not identical if their means are significantly different, and our argument is more about a mean shift, ultimately.

I refer to the online component as "random" since moment to moment the effect is largely unpredictable to the player and again, metaphorically, makes the game more akin to roulette rather than poker. In the aggregated data, though, one could have a model taking something like connection bars into account as a factor variable (say, a logistic model regressing p = logit(player_skill_1 + player_skill_2 + connection_quality)) and that could produce a worthwhile model of the influence of the three variables. The question is whether or not the connection quality variable would be significant. Again, I strongly suspect it would, and it would not influence that estimate in a direction that would tend to weight player skill more highly stratifying on worsening connection strengths. If the point of the game is to demonstrate which player has better skill then that outcome would strongly suggest offline isn't "meaningful" for determining that sort of thing. I strongly suspect even an online skill variable would be less influential than the connection quality if we split "offline" and "online" skill components out for each player. To go a little reductio ad absurdum, play anybody in the world on 1 bar and see how well you think that represents your skill level or theirs. This is all hypothetical, but the mechanics of the situation in terms of the effects of lag suggest that this is the case, to the extent that skill is dependent on the lack of variability in the frame data (and it is dependent on that, whether or not you in fact intellectually know the precise frame data or merely that one move beats another). Hence, online play seems less meaningful.

This is getting deep into the statistical weeds, but this is a sore spot with me. TL;DR: Precision in terminology does count, and if we're going to take victory as indicative of skill then online play is almost necessarily less meaningful as a test for such skill.

This is kinda a gem on a site in which the term "randomness" is tossed out with a lot of vigor and awfully little meaning. I do think you are essentially correct, though I would suggest that there is more of an "online" skill component that determines how one holds up under lag than not. But even it doesn't answer entirely for variables that are simply going to be outside of perception; if lag was consistent in and of itself, that'd be one story, but it is not, and not even necessarily consistent between the players. (Taking inventory on its apparent patterns does help, though. But this only further distances the offline game from the online game.)

Though it would have been better without the degrading tone. The argument speaks for itself. And, to a point, Brute actually didn't equate the two terms, thus the quotation marks and the use of dispersion rather than randomness as a choice.

 

[Prince Adon]

I don't know what this dumb argument is about but point is online can't be taken seriously. Lag screws up everything, from being able to punish to being able to react accordingly. Competitive players should only use online as a form of practice. Other than that online is only good for some fun friendly matches with friends.

DOA5 netcode is a peice of sh**. People can win with the most garbage stuff. It's not even worth using for practice that's how bad it is. DOA5 online means absolutely nothing to me.

 

[VirtuaKazama]

To me, Offline matches is where it's at. Every Wednesday (if I have the time) I would end up playing people at Xanadu or the Gameroom at my college (either for matches or just to train someone).

 

[SilverKhaos]

Online is the lifes blood of games nowadays. Way more people can and do play at home online, than the ones who can spend time and money running around to offline scenes. Without online, the FGC would be pretty much dead nowadays. Or least even more of a niche demographic than it was before. With online, though, everyone gets to play fighters.