EPTRace balance in progaming - Page 3
| Forum Index > BW General |
|
lazz
Australia3119 Posts
| ||
|
baubo
China3370 Posts
On July 30 2009 21:01 RowdierBob wrote: That sounds like speculation. As I said, I'm curious to see the stats. I could be wrong as you say, but I think PvZ in particular has become much more balanced post-Bisu. You just want to see stats? Isn't that what TLPD is for? | ||
|
sudo.era
United States300 Posts
The reason terran is most dominant versus zerg early on: new zergs fail to understand that ZvT is about trying to keep the terran in his base until you get defilers and/or ultras. You can surprise the terran and try to win before he gets too many siege tanks/vessels, or even get lucky with a dodge + counter-attack. But otherwise you just want to contain and rush to hive asap. It's a hard lesson to learn when your understanding of the game is essentially "1. expand more than he does, 2. get units, 3. upgrade, 4. er... win?". And at this level of understanding, noob terrans will steamroll most noob zergs. The same goes for some Protoss play, I'm sure. The funny thing is that, as I've heard some others say before, I believe psi storm is the most powerful ability in the game (next up being dark swarm), and it's the ability that causes me to lose most often -- yet ZvP is my best win ratio. Learning to pick off high templars pre-battle, and then predict where storms will be from remaing HT's, and then overwhelm with your army is a hard lesson to learn, just like needing hive for ZvT. And yet, I do definitely feel like ZvP is the easiest to win for me. About pro level, Savior said in an interview that zerg is, overall, the most difficult race to play. And given he's better at every race than anybody in this forum, I tend to believe him. Alot of people would misinterpret this as meaning that zerg has the most difficult time winning, but that's not really the case. At that level of skill, you should have the ability to pull off more difficult maneuvers by reflex alone.What he referred to was more the level of control for zerg, overall - not just for specific strategies - takes the highest level of attention and action. This doesn't mean there aren't specific builds and strategies the other races have that don't trump zerg in this respect. Just talking generally. ^^The same goes for comments on Protoss, but in the opposite way. Protoss, lately, have the hardest time winning, but overall take the least control. I'm not saying this with bias, just stats. Unless you see Protoss pulling off really high-apm strats, they tend to require less action to get things done. With two players at the same level of skill (in the pro leagues), it's very common to see 200-250 EAPM protoss players beat 350-400 EAPM zerg. I think I know the least about terran, except that TvP seems like a terran camp-fest - putting the burden on toss players to indirectly engage the gigantic siege line. And like I said before, TvZ is almost always in the terran's favor until hive tech. So yeah, theres... all of that. Feel free to disagree. | ||
Plexa
Aotearoa39261 Posts
On July 30 2009 17:16 CaptainPlatypus wrote: Almost zeroThanks  Any chance of TLPD getting an export function so I don't need to make 147 player entries in an excel table by hand to do it? ;-) Bug pop about it, im sure he'll help out | ||
|
errol1001
454 Posts
About pro level, Savior said in an interview that zerg is, overall, the most difficult race to play. Ugh.. who cares? Also, I don't, and I suspect most people don't, think that APM or EAPM = skill. | ||
|
sudo.era
United States300 Posts
On July 30 2009 23:25 errol1001 wrote: Ugh.. who cares? Also, I don't, and I suspect most people don't, think that APM or EAPM = skill. I do, and apparently you don't. Maybe it'd be more succinct to say, "I don't care." Skill is interpretive, and has little to do with APM. Never said otherwise. I think you missed the part where I said "at the same level of skill," literally. It's right there. So, to repeat, I was talking about necessary action. The amount of action necessary, I'd argue, can definitely determine difficulty. "Skill" is such an obtuse concept I wouldn't even try to define it. | ||
NonY
8751 Posts
On July 30 2009 15:54 FakeSteve[TPR] wrote: This doesn't take into account things like maps and player skill. imo that doesn't need to be done | ||
|
Tom Phoenix
1114 Posts
On July 30 2009 17:39 OneOther wrote: lol how long will the protoss 1a2a3a joke live on, especially with protoss getting raped left and right? It will never die, beacuse stereotypes defy all logic. Nevermind the fact that the idea that Protoss are easier to play would mean that Protoss progamers would have an easier time then others winning titles (which they do not). Nevermind the fact that the Protoss have never had a bonjwa of their own (and likely never will in BW) and that the only potential bonjwa they have is heavily disputed. Nevermind the fact that Brood War is famous for being well balanced and thus not giving any race an advantage. No, people will rather warp the very fabric of reality and convince themselves that the race they are facing is overpowered rather then admit that it is the faults in their own skill which prevent them from beating that race. Sadly, that is the nature of a game with asymmetrical races, regardless of how well balanced those are. | ||
|
Avidkeystamper
United States8556 Posts
| ||
|
sudo.era
United States300 Posts
On July 30 2009 23:37 Tom Phoenix wrote: It will never die, beacuse stereotypes defy all logic. Nevermind the fact that the idea that Protoss are easier to play would mean that Protoss progamers would have an easier time then others winning titles (which they do not). Nevermind the fact that the Protoss have never had a bonjwa of their own (and likely never will in BW) and that the only potential bonjwa they have is heavily disputed. Nevermind the fact that Brood War is famous for being well balanced and thus not giving any race an advantage. No, people will rather warp the very fabric of reality and convince themselves that the race they are facing is overpowered rather then admit that it is the faults in their own skill which prevent them from beating that race. Sadly, that is the nature of a game with asymmetrical races, regardless of how well balanced those are. I doubt many people will read my ridiculously long post, so I'll surmise how it applies in this reply. level of necessary control != ease of winning Where "skill" fits in is relative to your perspective. | ||
|
Zato-1
Chile4253 Posts
On July 30 2009 23:05 sudo.era wrote: I want to start off by saying that there's a major flaw in your chart, and that's that mirror matchups can't be included. I don't know how in the world they fit in, because regardless of who wins in a PvP, it will be a protoss. Overall, the percentage of protoss that win against protoss is 100%. That's one way of looking at it. Another way would be to say, 'Protoss always loses in the mirror'- no matter who wins, Protoss always loses. It makes just as much sense as your own argument. How do you make sense of it? Look at how it's done in the other matchups. If Protoss player A has played against T 100 times and won 55 games, you say his winrate in PvT is 55%. If he's played 100 games vZ and won 40 of those, you say his winrate in PvZ is 40%. If he's played 100 games vP and won 52 of those, you say his winrate PvP is 52%. If you're only considering the top 50 players, I'd expect the average PvP winrate will be somewhat over 50% because protosses in the top 50 should have a better than even win record vs Protosses below the top 50. | ||
|
sudo.era
United States300 Posts
On July 30 2009 23:57 Zato-1 wrote: That's one way of looking at it. Another way would be to say, 'Protoss always loses in the mirror'- no matter who wins, Protoss always loses. It makes just as much sense as your own argument. How do you make sense of it? Look at how it's done in the other matchups. If Protoss player A has played against T 100 times and won 55 games, you say his winrate in PvT is 55%. If he's played 100 games vZ and won 40 of those, you say his winrate in PvZ is 40%. If he's played 100 games vP and won 52 of those, you say his winrate PvP is 52%. If you're only considering the top 50 players, I'd expect the average PvP winrate will be somewhat over 50% because protosses in the top 50 should have a better than even win record vs Protosses below the top 50. Well yes, of course there's always a winner and loser. That's the point. We're not looking at one specific player and how he has done in every matchup. We're getting an average. Say we get an average of 2 protoss players in their PvP records. They've only played eachother, so each win is included in the average, as is each loss. Now in reality, each protoss player doesn't play against another exclusively. But, when you're taking an average of every single pvp, it's the same exact thing. You're averaging in every win and every loss of every game. When player B wins, player Z loses. For every win, there is a loss. It is exactly 50% wins and 50% losses. | ||
|
Tom Phoenix
1114 Posts
On July 30 2009 23:46 sudo.era wrote: I doubt many people will read my ridiculously long post, so I'll surmise how it applies in this reply. level of necessary control != ease of winning Where "skill" fits in is relative to your perspective. No offence, but level of necessary control means that whoever requires less of it has an advantage, which then translates into who has an easier time of winning. This means that "level of necessary control" and "ease of winning" are closely connected. | ||
|
sudo.era
United States300 Posts
On July 31 2009 00:04 Tom Phoenix wrote: No offence, but level of necessary control means that whoever requires less of it has an advantage, which then translates into who has an easier time of winning. This means that "level of necessary control" and "ease of winning" are closely connected. Nope, strategy. Your gameplan and how it fits into your opponent's is an enormous factor, if the not the majority determining factor. For example, something we see everyday, even in low-level iccup games -- control will not help you if your base is being harassed while your units are outside your base. There's also decisionmaking ability, which is part strategy but also has much to do with unit distrobution and expansion/upgrade timing. If you chose a poor army distrobution, control will not help. If you're at a section of the game where you need just one more minute before you have the necessary units/upgrades/abilities to push out, and then you get attacked and overwhelmed, control may not save the game. Winning in a specific situation can also have much to do with chance - out of each player's control. To end it, I want to point out that as long as you can provide the necessary control for your race/build/strategy, it isn't an issue. Pros who need more control to do what they do -- do it, and get themselves on equal footing. | ||
|
EchOne
United States2906 Posts
On July 30 2009 23:37 Avidkeystamper wrote: Well, you chose a period where most protoss are slumping. Try seeing what you get in the Golden Age or Arena MSL time. I don't think encouraging him to focus his research on a period of anomaly will provide more accurate or helpful results. | ||
|
EchOne
United States2906 Posts
On July 31 2009 00:07 sudo.era wrote: Nope, strategy. Your gameplan and how it fits into your opponent's is an enormous factor, if the not the majority determining factor. Talking about something we see everyday, even in low-level iccup games, control will not help you if your base is being harassed while your units are outside your base. You deflect attention from the connection by mentioning a section that fails to actually disprove. Regardless of strategy, when the level of necessary control is lower, the player has more liberty to exercise control beyond the necessary. Pushing the envelope this way is not always relevant to victory since an opponent's strategy can mitigate this advantage, but this mitigation is variable and it is impossible for strategy to totally negate the advantage in all cases. In the aggregate, greater liberty to multitask or over-micro must bestow greater chances of winning. | ||
|
sudo.era
United States300 Posts
On July 31 2009 00:18 EchOne wrote: You deflect attention from the connection by mentioning a section that fails to actually disprove. Regardless of strategy, when the level of necessary control is lower, the player has more liberty to exercise control beyond the necessary. Pushing the envelope this way is not always relevant to victory since an opponent's strategy can mitigate this advantage, but this mitigation is variable and it is impossible for strategy to totally negate the advantage in all cases. In the aggregate, greater liberty to multitask or over-micro must bestow greater chances of winning. Not if 'over-micro' provides no advantage. And I'd argue it doesn't. If you need to move unit A to position B, you do it. You don't need to move it halfway, back to A, then to B. The 'amount of necessary control' isn't a baseline. It's the relative ceiling before spam. It's what the player does that he wants done - that should be done. Anything more is excessive and accomplishes nothing. Anyway, the point is null. I already pointed out that, at the pro level, you can see that the average protoss player has less EAPM (even APM) than the average Zerg player. This is just an example matchup. If you were correct, the protoss players would simply 'over-micro' and catch up in actions. But they don't, because they don't need to -- unless they're executing a specific strategy that requires it. (Before somebody reads the above paragraph and interprets it as a comment on skill, fuck off and read further up in the convo. Thanks.) | ||
|
Foucault
Sweden2826 Posts
On July 30 2009 15:54 FakeSteve[TPR] wrote: This doesn't take into account things like maps and player skill. You can't determine game balance by looking at statistics so simply. My thoughts exactly. This is a "classic" scientifical dumbed-down analysis of something that lacks parameters like how good the actual players are, regardless of race. Also maps play in as we all know. This is almost as dumb as people who only look at map statistics and go: "wow 4-2 PvT, this map favours protoss over terran". Your stats might as well imply that Terrans have generally been the better players, regardless of race. It's just way more complicated than you think. Also, the reason protoss never got more dominant overall is PvZ. You maybe didn't hear it here first, but it's the goddamned truth. | ||
|
rextyrann
Germany41 Posts
On July 30 2009 15:54 FakeSteve[TPR] wrote: This doesn't take into account things like maps and player skill. You can't determine game balance by looking at statistics so simply. i dont fully agree with this. if we take an average/all maps+players into considertation it should even out (assuming that the maps dont "ALL" favor a specific race). but i do agree that the determination or conclusion which can be made by this form/amount of data is really really insufficient. there are some points which i would take into consideration before making any form of analysis: 1. wheighting of datas: how many players actually play T/Z/P ? you cant just take them all as average since higher amounts of a certain race thin out the actual performance of skill/ELO (its not chess or tennis where everyone has the "same" race) 2. how many of the top50 are T/P/Z ? it will influence the overall performance if wed assume for instance like 25 T players 15 Z and only 10 Ps since the Ts should naturally do better than the other 2 races in the top50 if there are more of them (well else they wouldnt be there right)? 3. it would be more intresting to see how the vs balance changes from lowranks ( D-C) to midranks (C+ -B+) and "kor" ranks (A-olympic). if you'd wheighted out the players oer race/ELO you can probably see/determine in what kind of "vs mu" -> actual strength weaknesses of races develop according to the ranklvl of iccup. meaning: then you would actually see which matchups change significantly and therefor are easier/harder to play at a certain skilllvl-> metagame changes etc. those are some basic ideas which came into my mind when i read your post. i guess a real mathematics student could even give you a deeper analysis and tool usage into that matter. | ||
|
EchOne
United States2906 Posts
On July 31 2009 00:25 sudo.era wrote: Not if 'over-micro' provides no advantage. And I'd argue it doesn't. If you need to move unit A to position B, you do it. You don't need to move it halfway, back to A, then to B. The 'amount of necessary control' isn't a baseline. It's the ceiling. It's what the player does that he wants done - that should be done. Anything more is excessive and accomplishes nothing. Anyway, the point is null. I already pointed out that, at the pro level, you can see that the average protoss player has less EAPM (even APM) than the average Zerg player. This is just an example matchup. If you were correct, the protoss players would simply 'over-micro' and catch up in actions. But they don't, because they don't need to -- unless they're executing a specific strategy that requires it. (Before somebody reads the above paragraph and interprets it as a comment on skill, fuck off and read further up in the convo.) Sorry about my misunderstanding of necessary control. I was thinking that from your premise of an average Protoss having ~100 less EAPM than an average Zerg while winning PvZ, whatever control he won with becomes what is necessary to perform the winning strategy. However, once his EAPM equals the Zerg's, he can leverage it into what I should really call a strategic change here, executing play like Bisu's that actually demands multitask and promises returns. While we're poking at each others' semantics, I'll question you again. Ease here refers to liberty from difficulty. If we take control to be difficult, and more control to be increasingly difficult, a lower level of necessary control (tied to a winning strategy as your understanding goes) should imply greater ease of winning. Obviously they're not equal, but how do you conclude that they're not at all related? To clarify my assumption on necessary control being related to a winning strategy: your understanding indicates that the control is a function of what the player should do, which essentially makes it the arm of decision making, which is what comprises a strategy. If anything greater than necessary control is always excessive and unproductive, necessary control at least ideally must refer to a strategy that will win. If a strategy fails, clearly a different strategy could have succeeded, with different necessary control requirements. Since strategy and control must vary in tandem to reach victory, the effect of one does not obviate deficiencies in the other. | ||
| ||
