NeoGame-i maps/balance statistics - Page 3
| Forum Index > BW General |
|
Hot_Bid
Braavos36402 Posts
| ||
|
tfeign
United States2980 Posts
On July 19 2004 15:18 STIMEY d okgm fish wrote: i'd like to see what the stats are for people who are 1300 and higher, or whatever. ahigh number so that there are less games between two players who arent very evenly matched (as u approach 1000 u get more and more mismatches due to new accounts and players who havent played enough games yet). STIMEY Look at the top 100 of ng-i. Count the number of zergs and then count the number of tossses...yeah..there are about 5 times as many z's than p's. And so logically, if you take into account players with 1300+, you will realize that the zvp %gap will be even worse. ZvP is imbalanced. Yes, either race can win. But with 2 equally skilled players Z will have a greater chance of beating P than vice versa. And to sum up my other posts I made in the other topic: One small mistake in pvz can many times cost you the game. P has to play almost perfect to win vs a decent zerg. That's why in the professional level p may be = z, because pro protosses can play at almost-perfect games but in any other kind of semi-pro or amateur level, Z >>> P Examples: P successfully corsair-rushed Z: p kills off a few ovies, get to watch Z's tech. Game is still up for grabs. Z successfully muta-rushed P: Game over. A successful muta rush would be gg 90% of the time. P successfully high templar drop Z's mineral line and killed off a bunch of drones. Z is slowed down momentarily while he simultaneously makes another bunch of drones from his hatches. Game still up for grabs. Z successfully lurker dropped P's mineral line and killed off a bunch of probes (note that since toss don't have many expansions, MANY more probes are gathered in the mineral lines). Lots of probes die and protoss cannot replenish his workers anywhere near as fast as zerg. No way off toss coming back into the game now. P successfully killed off one of Z's expasion. Z is slowed down a bit. Z successfully killed off one of P's expansion. Game over. Toss cannot afford to lose any exp vs zerg unless he is way ahead in the game. P catches Z offguard with fast DT. Zerg just has to pull back and wait until his ovies come along with his army and attack again. Z catches P offguard with fast lurkers. Now protoss will be contained in his base while Z expands and MASS. In pvz, a contained toss will lose most of the time. Etc... There are probably less than .0001% of the Starcraft population who can play at an almost perfect level. Should Blizzard balance the game for the .0001% of the Starcraft population or for the 99.999% of the Starcraft population? Definitely not saying that pvz is impossible, but it is not balanced for 99.999% of the players. Surely there is an ENORMOUS amount of zerg players compared to toss players in the top 100 at ng-i. Is this something you would want? | ||
|
Bill307
Canada9103 Posts
. They have a huge advantage over professional game stats in that the sample size is tens of thousands of times larger. Unfortunately, they also come with the following disadvantages:a) they include a lot of newbie games b) with the possible exception of LT (since this info can eventually be learned simply through experience), most players do not spend a great deal of time analysing and practicing a map and learning how to maximize the effects of its imbalances What MPXMX and RugBug have noticed is an extreme case of (b), where players don't know the map at all. To really learn how balanced or imbalanced a map is, one must look at games played by people who are very serious about trying to win or lose and devote quite a lot of time and effort into learning the maps and learning to exploit them -- AND how to counter possible exploitations that may be used against them. I agree with Stimey's suggestion that we should look at stats collected with a restriction based on rating. I think that the current sample scope is so large that the stats become meaningless. They reflect the average map balance, but this is NOT the same as map statistics from actual average players. By the intermediate value theorem (assuming the change in map statistics in continuous as you move from bad to good players), for each map, there must be some level of skill where that map's statistics match the actual statistics of games played at that level. But we cannot know where this accurate protrayl takes place. Consequently, these map statistics are not reliably useful to anyone. Ideally, we could get statistics from various rating ranges (high players, medium-range players, and maybe even sub-1000 players). Thus, players who fall neatly into one of these categories would be much more likely to find these statistics to be accurate. Even then, because of points (a) and (b) above, they probably won't be very good, but at least they would be better. Heck, it might even turn out that the statistics remain virtually identical for all of those sub-ranges, which would be an interesting find. But for all we know, there could be vast differences. Ultimately, though, I think that (b) will always prevent any nGi map statistics -- except for possibly LT -- from being truly useful. | ||
|
Bill307
Canada9103 Posts
Edit: for the idiots like tfeign, this last sentence refers to his suggestion above to change PvZ to accomodate the majority of gamers verus progamers. | ||
|
Bill307
Canada9103 Posts
| ||
|
tfeign
United States2980 Posts
On July 19 2004 15:45 Bill307 wrote: But changing SC's balance would be a completely selfish and stupid thing to do. WOW how unselfish! Changing SC's balance to make each race having the same chances of winning is so unselfish and stupid. Who would've thought. | ||
|
Bill307
Canada9103 Posts
On July 19 2004 15:52 tfeign wrote: WOW how unselfish! Changing SC's balance to make each race having the same chances of winning is so unselfish and stupid. Who would've thought. Wow, it must take a looooot of skill and intelligence to be able to isolate a single sentence, take it out of context, and then criticise it! While ignoring the rest of my post because you know full well that I'm right and you can't argue against it. | ||
|
tfeign
United States2980 Posts
On July 19 2004 15:48 Bill307 wrote: tfeign: another thing you fail to take into account is that ALMOST ALL nGi games are on LT (in fact, I would conjecture that the concentration of LT games is even higher near the top of the ladder than near the bottom). So instead of changing the game's balance, perhaps you should strive to fix PvZ on LT. Ok: Namja Iyagi: Z>P 63.2% Martian Cross: Z>P 61.5% Nostalgia: Z>P 54.2% 90% of games are played on LT. Should the game be balanced on the map that 90% of the people play on, or the map where less 5% of the people play on? | ||
|
Bill307
Canada9103 Posts
As for WCG, they should rethink their map selection. | ||
|
Famouzze
971 Posts
| ||
|
tfeign
United States2980 Posts
On July 19 2004 15:58 Bill307 wrote: Then maybe toss players shouldn't play on maps that favour zerg so much? Players can sometimes be coerced into playing on LT because it's so widespread, but with the exception of the beacon of ignorance that is WCG, I don't see why anyone would have to PvZ on a map where that matchup is horribly imbalanced. In fact, they could probably just get the game played on LT instead. As for WCG, they should rethink their map selection. Yes, WCG maps suck. I dont know who chose the maps, but the maps picked were horrible. I am willing to bet my life that no toss player will win WCG this year. And making people stop LT is not going to work. People have tried doing that for years, especially in pro gaming where no LT games are played. But that isnt going to make people stop playing LT. Tell me, what is easier to change? Tweak up game balance or making people stop playing LT? | ||
|
8882
2737 Posts
top 100players at ngi are nearly all koreans maybe blizzard should balance it? IMO if they played protoss they would still beat your ass and you would write that terran/zerg is the imbalanced race (btw no, Im not gloryfying koreans or sth, they just play a lot more and are usually better, so they win. they used to play zerg a lot and they own protoss players (with lower number of games) easily. Koreans always thought that protoss is the worst race and not many of them played it to reach a high level now (it's one of their weird theories, like the one that pressing f4=faster download). Now when nal_ra introduced the new strategies, protoss became more popular among koreans, but still most of koreans switched to terran to be like nada/oov and they own protoss players). maybe the ngi staff could also add a chart about the %of each starcraft race played by each country (and the %of each race at point levels) then we could be sure, that koreans dislike protoss and nonkoreans like to live for aiur | ||
|
STIMEY d okgm fish
Canada6140 Posts
therefore, i think trusting the 200k or so LT games played on neogamei would be at least as fair as trusting progamer stats. of course, the more lower ranks you rule out, the more bullshit you rule out, but you also decrease the sample size. so should we look at 1100 and up LT stats, 1200+, 1300+, 1400+? i would start with 1300+ probably since that's a fairly significant amount of games in one direction to move up but this is quite a subjective estimation on my part D: but again, the positions on LT can be so different! is a map really fair just because protoss wins 70% of the time from 10% of the possible starting locations and then does 40% of the time from the other positions? i think naught. as for balance changes, i have recently pointed out some initial ideas of how you can effect some matchups without effecting others (significantly): you can adjust psi or supply requirements of various relevant units (effects tvp more than other matchups), you can effect templars (effects pvz more than other matchups), you can effect zealot hp and shield balance (effects pvz more than other matchups), you can also make dual changes to two races such that their matchup stays balanced while effecting a different matchup. it's possible to favorably improve games in a specific matchup with changes even though sc is already good how it is. it's possible to improve the game however slightly, however unneeded, :| | ||
|
Liquid`Jinro
Sweden33723 Posts
As I said LT isn't balanced at the highest levels of play. However, it only affects the highest levels hence, you can stfu. | ||
|
Bill307
Canada9103 Posts
On July 19 2004 16:19 STIMEY d okgm fish wrote: but again, the positions on LT can be so different! is a map really fair just because protoss wins 70% of the time from 10% of the possible starting locations and then does 40% of the time from the other positions? i think naught. I really like this point. It effectively means that the near-50% LT statistics are meaningless because there could still be very significant differences depending on the positions. I know that replays contain the starting location info (thx to bwchart), and if nGi were able to obtain that info as well and produce statistics from it then that would give us a much more accurate picture of how LT is balanced. Also, I think that I should start keeping track of positional statistics for the iTV maps as well, especially LT. | ||
|
Bill307
Canada9103 Posts
On July 19 2004 16:07 tfeign wrote: And making people stop LT is not going to work. People have tried doing that for years, especially in pro gaming where no LT games are played. But that isnt going to make people stop playing LT. Tell me, what is easier to change? Tweak up game balance or making people stop playing LT? I don't see how the ease of this decision can be more important than my original point: that by tweaking game balance you could be make-or-break-ing progamer careers. How is the game balance more important for the masses when none of us make any money off the game anyway? (exceptions exist near the top, but the masses on the whole essentially do not and even at the top level of non-pro the money doesn't compare to progamer-level money anyway) Fair competition may be important to the top non-pros of nGi, but for professionals it is many, many times more important. Thus, I suggested re-engineering LT (which you managed to rephrase as "getting players to play on non-LT maps" ) so that PvZ is fair, as an alternative to screwing up the game balance for pros. | ||
|
STIMEY d okgm fish
Canada6140 Posts
0 red. 9 on most lt's 1 blue. 6 on most lt's 2 teal. 12 on most lt's. 3 purp. 3 on most lt's. such stats would be very fucking valuable, the sooner they start the better. | ||
|
sv14
Russian Federation66 Posts
Enjoy! rating scope = (900,1000) total matches = 158915 Z>T 51.0 P>T 51.4 Z>P 53.8 rating scope = (1000,1100) total matches = 148730 Z>T 50.5 P>T 51.9 Z>P 54.9 rating scope = (1100,1200) total matches = 75779 T>Z 50.5 P>T 50.5 Z>P 55.8 rating scope = (1200,1300) total matches = 40319 T>Z 54.5 P>T 54.2 P>Z 51.1 rating scope = (1300,1400) total matches = 19421 T>Z 51.2 P>T 52.3 Z>P 55.3 rating scope = (1400,1500) total matches = 10671 T>Z 53.6 P>T 53.6 Z>P 61.8 rating scope = (1500,1600) total matches = 4924 Z>T 51.7 P>T 62.9 P>Z 58.4 (!!!) rating scope = (1600,1700) total matches = 2304 Z>T 55.8 T>P 57.5 (!!!) Z>P 60.3 | ||
|
LordOfDabu
United States394 Posts
But the change in Protoss between 1500-1600 and 1600-1700 boggles my mind. | ||
|
Hot_Bid
Braavos36402 Posts
| ||
| ||
. They have a huge advantage over professional game stats in that the sample size is tens of thousands of times larger. Unfortunately, they also come with the following disadvantages: