|
On August 11 2010 00:25 kajeus wrote:Show nested quote +On August 11 2010 00:17 lt.dunbar wrote:
@kajeus:
You can't really use the popularity of zerg across all leagues to project the expected percent of higher level players that play zerg. This is because players motivations changes based on there skill level.
Terran might be more popular at lower levels becuase of familiarity, the campaign, or maybe its just more fun to play than zerg.
But at higher levels where simply winning is the priority you would expect players to play the race that gives them the best chance to win. If the game were balanced the total number of players across the races at high levels should relatively even.
By focusing on the global population you are getting data skewed by the lower level players (most players) that make it seem like zerg is over represented in diamond league.
For example, if hypothetically, Idra was the only zerg player left you could say "100% of zerg are in diamond league" and make it seem like zerg was overpowered, when really its just no one plays zerg and there would probably be a gameplay/balance related reason. http://www.teamliquid.net/forum/viewmessage.php?topic_id=141732¤tpage=15#2836.15% of zerg players are in diamond. If IdrA were the only zerg left, 100% would be in diamond. The expected value is 5%, given that 5% of players overall are in diamond league. Moreover, zergs would be a tiny tiny percentage of all diamond league players in that case -- obviously not balanced either. In reality, 6.15% of zerg players are in diamond (not 100%), and they are 24% of the total number of diamond players (not .00001%). Given that they are 20% of the total number of players everywhere, this is a completely reasonable outcome. "If the game were balanced the total number of players across the races at high levels should relatively even." -- That is simply not true. If 20% of all players were zerg and zergs were 33% of diamond league, this would be truly bizarre. Let me give you an example. If you have 50% red jelly beans in a jar, 30% yellow jelly beans, and 20% white jelly beans, then take a large random handful of jelly beans, it is completely unreasonable to expect an even 33% split across all colours. Do you disagree with that?
Pointing out how many zergs are in diamond doesn't prove anything. Now, how many of those zergs are doing well in diamond? Sure, 6.15% of them might be in diamond, but maybe 75% of those 6.15% are hanging out at the bottom of their division (I don't know, I'm not obsessed with tracking these statistics down). Throwing out that statistic like it's the be-all, end-all of the argument is pretty pointless.
|
On August 03 2010 03:32 LaLuSh wrote:Medivac Suggestion: Lower the heal rate. 13.5hp/sec is really high. Reasoning: There's a reason why Organic Carapace isn't in the game anymore (OP). Even when it was in the game, it only provided for 10hp/sec regeneration for unburrowed roaches as opposed to 13.5hp/sec for Medivacs. Right now burrowed roaches without tunneling claws heal 5hp/sec. With tunneling claws it's 10hp/sec. A medivac supported army is essentially an army with an organic carapace upgrade. That's not my main beef however. Banelings deal well with that (until terran gets tanks at least). My main beef is the largely undiscovered potential of medivac drop play. Players like Brat_OK are already seen abusing them heavily. Shutting down expansions left and right. Doing quick main raids where half the tech/upgrade buildings die. A good terran can manipulate and constantly keep your army out of position.
First of all... ridiculous comparison to organic carapace. Medivac can only heal one unit at once. If you're comparing 13.5hp/sec to the 10hp/sec for burrowed roaches, you have to keep in mind ALL your roaches are regenning 10/sec AT THE SAME TIME. To reach the same level of hp/sec (as a group), you'd need a ton of medivacs.
Also, about your drop complaint... Apparently you're forgetting about overlords and warp prisms.
|
On August 11 2010 00:35 RxN wrote:Show nested quote +On August 11 2010 00:25 kajeus wrote:On August 11 2010 00:17 lt.dunbar wrote:
@kajeus:
You can't really use the popularity of zerg across all leagues to project the expected percent of higher level players that play zerg. This is because players motivations changes based on there skill level.
Terran might be more popular at lower levels becuase of familiarity, the campaign, or maybe its just more fun to play than zerg.
But at higher levels where simply winning is the priority you would expect players to play the race that gives them the best chance to win. If the game were balanced the total number of players across the races at high levels should relatively even.
By focusing on the global population you are getting data skewed by the lower level players (most players) that make it seem like zerg is over represented in diamond league.
For example, if hypothetically, Idra was the only zerg player left you could say "100% of zerg are in diamond league" and make it seem like zerg was overpowered, when really its just no one plays zerg and there would probably be a gameplay/balance related reason. http://www.teamliquid.net/forum/viewmessage.php?topic_id=141732¤tpage=15#2836.15% of zerg players are in diamond. If IdrA were the only zerg left, 100% would be in diamond. The expected value is 5%, given that 5% of players overall are in diamond league. Moreover, zergs would be a tiny tiny percentage of all diamond league players in that case -- obviously not balanced either. In reality, 6.15% of zerg players are in diamond (not 100%), and they are 24% of the total number of diamond players (not .00001%). Given that they are 20% of the total number of players everywhere, this is a completely reasonable outcome. "If the game were balanced the total number of players across the races at high levels should relatively even." -- That is simply not true. If 20% of all players were zerg and zergs were 33% of diamond league, this would be truly bizarre. Let me give you an example. If you have 50% red jelly beans in a jar, 30% yellow jelly beans, and 20% white jelly beans, then take a large random handful of jelly beans, it is completely unreasonable to expect an even 33% split across all colours. Do you disagree with that? Pointing out how many zergs are in diamond doesn't prove anything. Now, how many of those zergs are doing well in diamond? Sure, 6.15% of them might be in diamond, but maybe 75% of those 6.15% are hanging out at the bottom of their division (I don't know, I'm not obsessed with tracking these statistics down). Throwing out that statistic like it's the be-all, end-all of the argument is pretty pointless.
No, the same percentages hold in the top 200, top 100 -- pick a number, man. 20-24% of any given top-rated chunk will be zerg. This is exactly the same proportion as are present in the general population.
|
On August 11 2010 00:25 kajeus wrote:
If you have 50% red jelly beans in a jar, 30% yellow jelly beans, and 20% white jelly beans, then take a large random handful of jelly beans, it is completely unreasonable to expect an even 33% split across all colours. Do you disagree with that?
Of course this is true.
My point is that if we're talking about high level players, a "random handful" is precisely what is not happening.Its not a random sample of players from across all of BNet.
By using statistics from the total population you are presenting a skewed set of data. You are using the unpopularity of zergs throughout non-diamond leagues (95% of players apparently) to make it seem like zerg is over represented in diamond and therefore the match-up is balanced when really these numbers tell very little about balance.
|
On August 11 2010 00:35 driftme wrote:
First of all... ridiculous comparison to organic carapace. Medivac can only heal one unit at once. If you're comparing 13.5hp/sec to the 10hp/sec for burrowed roaches, you have to keep in mind ALL your roaches are regenning 10/sec AT THE SAME TIME. To reach the same level of hp/sec (as a group), you'd need a ton of medivacs.
Also, about your drop complaint... Apparently you're forgetting about overlords and warp prisms.
You're also forgetting something... Medivacs heal while the units attack. Organic Carapace heals when the unit STOPS ATTACKING. MASSIVE difference between the two. Also, if the Terran uses a scan, your burrowed units might as well be fish in a barrel. You need to consider those factors and why Terran Drops are just so godly awesome.
Finally, Overlords are not drop points. Yes they have that ability, but no I am not using my Need-Two-Upgrades-To-Drop and Pop Cap units to try and drop in the enemy base.
Terran loses Medivacs, they're out only the unit. Zerg lose Overlords and they lose Pop that they have to replace if they didn't spam Overlords just for the express purpose of Drop. (Which of course, Medivacs can do soemthign other than Drop.)
We aren't forgetting about Overlords, but it's a very risky and expensive endeavour to use them. Nydus doesn't run as much risk, but can be difficult due to a number of factors.
|
i think all ur suggestions r great other then the maruder slow effect and medivac change, stalkers etc have insane movement speed while maruders dont, i think itd break the point of getting maruders for alot of situations which right now, they would be used. and for medivac, they cost alot.. comparing it to burrowed roach regen just doesnt make sense. id def love to see a mule change where it requires more use, and imo thor does need air range reduction, not sure if i agree about the viking though
|
I've been lurking these boards / this thread for quite a while and registered just for this post.
kajeus, you have done a good job analyzing this data and I am amazed by your patience while explaining basic statistics to the masses on this board. It is consistently interesting how people will form new (nonsensical) arguments to support beliefs that they had before looking at the data.
On the topic of TvZ, I think limitations by many zerg players cause them to focus on certain strategies. For example many Zerg enter a ZvT game thinking "if he goes this I will get this, but if he gets this I will get THIS!" when in reality a Zerg player will have a better chance to win if he controls the tempo of the game himself.
|
On August 03 2010 04:04 Zoltan wrote:
So this is a nerf terran thread? What about warpgates being redonkulous? or that zerg can make 14!!! drones at once off of 2 bases?
arent u the same zoltan who used to play zerg but jumped ship and started playing terran? as i recall u used to have many many threads about zerg being underpowered.
|
On August 11 2010 01:57 geoIOPS wrote:
I've been lurking these boards / this thread for quite a while and registered just for this post.
kajeus, you have done a good job analyzing this data and I am amazed by your patience while explaining basic statistics to the masses on this board. It is consistently interesting how people will form new (nonsensical) arguments to support beliefs that they had before looking at the data.
On the topic of TvZ, I think limitations by many zerg players cause them to focus on certain strategies. For example many Zerg enter a ZvT game thinking "if he goes this I will get this, but if he gets this I will get THIS!" when in reality a Zerg player will have a better chance to win if he controls the tempo of the game himself.
How the heck do you think a zerg can control the tempo of the game against a decent terran ?
I believe that none of us think there's a vast imbalance in the MU, however, there seem to be a couple things that would make the game fun for both of the players and stop a 50 apm T to destroy a 200 apm zerg.
You know, this is what is so wrong about this match up, it might be fair, it might be balanced in the end when you look at the win percentage, facts remain, it is NOT fun to play from the zerg POV.
You are behing harassed, dropped and timed push in the early, mid and late game. Even if you end up outmacroing and starving the T for the win, you feel like you're never in control of the game until the very end. You either loose in the first 5-10-15-40 minutes or you win at the 49th.
|
|
|
On August 11 2010 02:22 Fitz wrote:
You know, this is what is so wrong about this match up, it might be fair, it might be balanced in the end when you look at the win percentage, facts remain, it is NOT fun to play from the zerg POV.
You are behing harassed, dropped and timed push in the early, mid and late game. Even if you end up outmacroing and starving the T for the win, you feel like you're never in control of the game until the very end. You either loose in the first 5-10-15-40 minutes or you win at the 49th.
No one is trying to argue what is fun and what is not. In fact, what you are saying is apparent in the data when we see a low amount of Zerg players overall, if 'fun' dictates how many people play a race. When it comes to balance, however, the stats will show that not only is TvZ ~50% (according to Blizzard) but Zerg also has a strong showing at all areas of player skill.
Look, you can exaggerate the situation by compounding all the ways you've been beaten to make a victory sound impossible but the fact of the matter is Zerg players win in TvZ and Zerg players are present in the Diamond league.
|
On August 11 2010 05:52 geoIOPS wrote:Show nested quote +On August 11 2010 02:22 Fitz wrote:
You know, this is what is so wrong about this match up, it might be fair, it might be balanced in the end when you look at the win percentage, facts remain, it is NOT fun to play from the zerg POV.
You are behing harassed, dropped and timed push in the early, mid and late game. Even if you end up outmacroing and starving the T for the win, you feel like you're never in control of the game until the very end. You either loose in the first 5-10-15-40 minutes or you win at the 49th. No one is trying to argue what is fun and what is not. In fact, what you are saying is apparent in the data when we see a low amount of Zerg players overall, if 'fun' dictates how many people play a race. When it comes to balance, however, the stats will show that not only is TvZ ~50% (according to Blizzard) but Zerg also has a strong showing at all areas of player skill. Look, you can exaggerate the situation by compounding all the ways you've been beaten to make a victory sound impossible but the fact of the matter is Zerg players win in TvZ and Zerg players are present in the Diamond league.
50% win-rate = matchmaking working. Blizzard has not commented on the actual real win-rate statistics in quite some time (i.e. beta). If I am ignorant of some ground-breaking blizzard press release, please provide me with a link. Otherwise, all the other data showing comparable win rates between races in division X merely points to the fact that Blizzard's automated matchmaking system is working as intended.
|
On August 11 2010 01:57 geoIOPS wrote:
when in reality a Zerg player will have a better chance to win if he controls the tempo of the game himself.
If only we could...
|
On August 11 2010 02:22 Fitz wrote:Show nested quote +On August 11 2010 01:57 geoIOPS wrote:
I've been lurking these boards / this thread for quite a while and registered just for this post.
kajeus, you have done a good job analyzing this data and I am amazed by your patience while explaining basic statistics to the masses on this board. It is consistently interesting how people will form new (nonsensical) arguments to support beliefs that they had before looking at the data.
On the topic of TvZ, I think limitations by many zerg players cause them to focus on certain strategies. For example many Zerg enter a ZvT game thinking "if he goes this I will get this, but if he gets this I will get THIS!" when in reality a Zerg player will have a better chance to win if he controls the tempo of the game himself. How the heck do you think a zerg can control the tempo of the game against a decent terran ?
Yeah, please enlighten me as well. My diamond zerg skills could use some help with that.
P.S. The fact is given the amout of effort it takes for us Zergs to macro would makes us Gods if only we switched to Terrans. But I have self-respect so I will not.
|
On August 11 2010 06:08 DTown wrote:Show nested quote +On August 11 2010 05:52 geoIOPS wrote:On August 11 2010 02:22 Fitz wrote:
You know, this is what is so wrong about this match up, it might be fair, it might be balanced in the end when you look at the win percentage, facts remain, it is NOT fun to play from the zerg POV.
You are behing harassed, dropped and timed push in the early, mid and late game. Even if you end up outmacroing and starving the T for the win, you feel like you're never in control of the game until the very end. You either loose in the first 5-10-15-40 minutes or you win at the 49th. No one is trying to argue what is fun and what is not. In fact, what you are saying is apparent in the data when we see a low amount of Zerg players overall, if 'fun' dictates how many people play a race. When it comes to balance, however, the stats will show that not only is TvZ ~50% (according to Blizzard) but Zerg also has a strong showing at all areas of player skill. Look, you can exaggerate the situation by compounding all the ways you've been beaten to make a victory sound impossible but the fact of the matter is Zerg players win in TvZ and Zerg players are present in the Diamond league. 50% win-rate = matchmaking working. Blizzard has not commented on the actual real win-rate statistics in quite some time (i.e. beta). If I am ignorant of some ground-breaking blizzard press release, please provide me with a link. Otherwise, all the other data showing comparable win rates between races in division X merely points to the fact that Blizzard's automated matchmaking system is working as intended.
If zerg were underpowered, 50-50 WLRs would generally mean good zergs are getting paired with bad terrans (e.g.), and hence zergs would be underrepresented in the upper echelons of the various regional ladders.
Imagine a zerg player and a terran player of absolutely equal ability. If the matchmaking system works, then the terran player will reach 50% at a point higher in the rankings than the zerg player will reach 50%. Generalize this to all players, and zergs will be underrepresented in the upper echelons (diamond, top 200, whatever).
However, in reality, zergs are (e.g.) 24% of the top 200 players in NA, as well as exactly 24% of diamond. They are a mere 20% of all players worldwide. In other words, the matchmaking is not only 50/50, but zergs are getting placed exactly where they're supposed to be (if not HIGHER than they're supposed to be), based on performance. This is strong evidence that zergs are doing okay.
If you disagree with this, you have to think about what outcome you would expect if zerg were underpowered. The outcome you would NOT expect in such a case is for zergs to be represented in equal proportions in the general population, diamond league as a whole, and the top 200 (e.g.).
|
most of u keep getting stats from sc2 ranks which is pure bs
League Random Protoss Terran Zerg
Diamond 56.08% (223,037) 56.42% (925,858) 56.74% (746,866) 56.24% (629,948)
224% out of 100 when clearly terran is like 30% more than zerg and random is like 60% less but all have 56%
If the total are 2523 as a solid number
toss are 925 is like 36,66
terran are 29,56
Zerg 24,5
so its clearly lower , not to the point of been complete useless but to the point that there is something to be fixed
|
On August 11 2010 06:56 st3roids wrote:
most of u keep getting stats from sc2 ranks which is pure bs
League Random Protoss Terran Zerg
Diamond 56.08% (223,037) 56.42% (925,858) 56.74% (746,866) 56.24% (629,948)
224% out of 100 when clearly terran is like 30% more than zerg and random is like 60% less but all have 56%
Those are average win rates...
|
I believe the point of slush wasnt only about nerfing terrans slightly, but also working toward having terrans work for their wins. By that Id say that terrans shouldnt be able to deny zerg expos with easily and absolutly free bunkers along with economic damage through hellions/reapers while slowly queuing units in their impenetrable fortress of a base until they got enought, 1a and roll over the weakened zerg.
|
I don't really understand all fuss about the TvZ match-up. However some changes need to be made to the range on the Viking. The range is beyond insanity if you compare it to the Corrupter and Phoenix.
If you look at the 3 Anti-air units:
- Corrupter
- Increased Damage Ability casts only on single target.
- Phoenix
- Graviton Beam (useless against air)
- Can shoot while moving
- Viking
- Range 9
Essentially the Corrupter must get close enough to not only cast but to fire. In that time the Vikings can generally get off a full volley of missiles.
The Phoenix move while attacking is useful, however when going against a group of Vikings the attack while move is useless because the Vikings can just sit in place with the mass range while the Phoenixes attack while moving is pointless.
Conclusion: Leave everything as it is. Except change the range on the Viking.
|
United Kingdom12064 Posts
On August 11 2010 07:09 jombeeno wrote:
I don't really understand all fuss about the TvZ match-up. However some changes need to be made to the range on the Viking. The range is beyond insanity if you compare it to the Corrupter and Phoenix.
If you look at the 3 Anti-air units:
- Corrupter
- Increased Damage Ability casts only on single target.
- Phoenix
- Graviton Beam (useless against air)
- Can shoot while moving
- Viking
- Range 9
Essentially the Corrupter must get close enough to not only cast but to fire. In that time the Vikings can generally get off a full volley of missiles.
The Phoenix move while attacking is useful, however when going against a group of Vikings the attack while move is useless because the Vikings can just sit in place with the mass range while the Phoenixes attack while moving is pointless.
Conclusion: Leave everything as it is. Except change the range on the Viking.
Phoenixes and Corrupters can easily get right next to vikings while only taking one volley and then the vikings so slow that it'll never be able to get back out of range again and they die so fast.
Also, reduce viking range and Collosi with the range upgrade will become almost unkillable when you have a mass number of stalkers, most maps don't really let you flank collosi with air units so it'd make the collosus far too powerful.
|
|
|
|
|
|
EPT![[image loading]](http://wiki.teamliquid.net/starcraft/images2/thumb/9/9f/Medivac.png/180px-Medivac.png)
