|
I thought it might be fruitful to look at maps where Protoss has been doing well vs Zerg. Maybe we can identify some of the relevant map features and use them more frequently in the future.
Here's a selection of maps from eloboard.com where Protoss has a win rate above 50% versus Zerg. I left out maps which are rather non-standard, e.g. Monty Hall or Minstrel. There actually aren't that many maps that fit this description, so I included three more maps with WRs very close to 50% (Apocalypse, Pole Star, Neo Sylphid).
maps that nerf the 973 build
KnockOut+ Show Spoiler +
Tempest+ Show Spoiler +
Revolver+ Show Spoiler +
Each of these three maps somehow nerf the 973 hydra builds. I'd really like to see both the low-ground stripe of KnockOut used more often in the future, as well as the particular double entrance layout on Revolver.
other
Butter+ Show Spoiler +
Citadel+ Show Spoiler +
Invader+ Show Spoiler +
PoleStar+ Show Spoiler +
Apocalypse+ Show Spoiler +
Neo Sylphid+ Show Spoiler +
Here it's somewhat harder to pinpoint a specific feature. To me it seems like on some of these maps its particularly hard to play the midgame hydra/lurker zone control as Zerg. As Zerg, you'd like to have a single, big Lurker field which can be constantly reinforced and Protoss is forced to attack into it. Some of these maps either have a layout where the expansions are positioned in such a way that it's awkward to set up this defensive lurker field (Citadel, Blitz, Invader); or the map layout is in a way such that Protoss can attack into the lurker field from an advantageous high ground position (Butter, Pole Star).
A somewhat recurring feature might also be the presence of easily defensible, single-ramp highground expansions.
---
Last, there are some specific features that can be added to basically any map. One is the possibility to cannon rush at the natural, and another is the presence of tight spots at expansions where Zealots can be positioned so that only a single Zergling can attack them. Anything else?
Please add if anything is missing here and why you think the maps above are good for Protoss.
What other features could be used on maps that would help Protoss in this matchup? I think of particular interest are things that don't otherwise break or change the game too much, while also not having too big of an influence on the other matchups...
|
On October 28 2025 10:41 Kraekkling wrote:I thought it might be fruitful to look at maps where Protoss has been doing well vs Zerg. Maybe we can identify some of the relevant map features and use them more frequently in the future. Here's a selection of maps from eloboard where Protoss has a WR above 50%. I left out maps which are rather non-standard, e.g. Monty Hall or Minstrel. There actually aren't that many maps that fit this description, so I included three more maps with WRs very close to 50% (Apocalypse, Pole Star, Neo Sylphid). maps that nerf the 973 buildKnockOut + Show Spoiler +Tempest + Show Spoiler +Revolver + Show Spoiler +Each of these three maps somehow nerf the 973 hydra builds. I'd really like to see both the low-ground stripe of KnockOut used more often in the future, as well as the particular double entrance layout on Revolver. otherButter + Show Spoiler +Citadel + Show Spoiler +Invader + Show Spoiler +PoleStar + Show Spoiler +Apocalypse + Show Spoiler +Neo Sylphid + Show Spoiler +Here it's somewhat harder to pinpoint a specific feature. To me it seems like on some of these maps its particularly hard to play the midgame hydra/lurker zone control as Zerg. As Zerg, you'd like to have a single, big Lurker field which can be constantly reinforced and Protoss is forced to attack into it. Some of these maps either have a layout where the expansions are positioned in such a way that it's awkward to set up this defensive lurker field (Citadel, Blitz, Invader); or the map layout is in a way such that Protoss can attack into the lurker field from an advantageous high ground position (Butter, Pole Star). A somewhat recurring feature might also be the presence of easily defensible, single-ramp highground expansions. --- Last, there are some specific features that can be added to basically any map. One is the possibility to cannon rush at the natural, and another is the presence of tight spots at expansions where Zealots can be positioned so that only a single Zergling can attack them. Anything else? Please add if anything is missing here and why you think the maps above are good for Protoss. What other features could be used on maps that would help Protoss in this matchup? I think of particular interest are things that don't otherwise break or change the game too much, while also not having too big of an influence on the other matchups...
You might make some good points, but we basically check out when someone says "I".
I = me = myself. We do things together always.
Can't really do anything with people who think they are doing things by themself.
Not a great vibe as the millenials would say 🤷♂️🤷♀️
|
On October 28 2025 11:44 perfectspheres wrote:Show nested quote +On October 28 2025 10:41 Kraekkling wrote:I thought it might be fruitful to look at maps where Protoss has been doing well vs Zerg. Maybe we can identify some of the relevant map features and use them more frequently in the future. Here's a selection of maps from eloboard where Protoss has a WR above 50%. I left out maps which are rather non-standard, e.g. Monty Hall or Minstrel. There actually aren't that many maps that fit this description, so I included three more maps with WRs very close to 50% (Apocalypse, Pole Star, Neo Sylphid). maps that nerf the 973 buildKnockOut + Show Spoiler +Tempest + Show Spoiler +Revolver + Show Spoiler +Each of these three maps somehow nerf the 973 hydra builds. I'd really like to see both the low-ground stripe of KnockOut used more often in the future, as well as the particular double entrance layout on Revolver. otherButter + Show Spoiler +Citadel + Show Spoiler +Invader + Show Spoiler +PoleStar + Show Spoiler +Apocalypse + Show Spoiler +Neo Sylphid + Show Spoiler +Here it's somewhat harder to pinpoint a specific feature. To me it seems like on some of these maps its particularly hard to play the midgame hydra/lurker zone control as Zerg. As Zerg, you'd like to have a single, big Lurker field which can be constantly reinforced and Protoss is forced to attack into it. Some of these maps either have a layout where the expansions are positioned in such a way that it's awkward to set up this defensive lurker field (Citadel, Blitz, Invader); or the map layout is in a way such that Protoss can attack into the lurker field from an advantageous high ground position (Butter, Pole Star). A somewhat recurring feature might also be the presence of easily defensible, single-ramp highground expansions. --- Last, there are some specific features that can be added to basically any map. One is the possibility to cannon rush at the natural, and another is the presence of tight spots at expansions where Zealots can be positioned so that only a single Zergling can attack them. Anything else? Please add if anything is missing here and why you think the maps above are good for Protoss. What other features could be used on maps that would help Protoss in this matchup? I think of particular interest are things that don't otherwise break or change the game too much, while also not having too big of an influence on the other matchups... You might make some good points, but we basically check out when someone says "I". I = me = myself. We do things together always. Can't really do anything with people who think they are doing things by themself. Not a great vibe as the millenials would say 🤷♂️🤷♀️
why would he not use "i" when describing his own opinion? what a stupid post
|
|
|
sorry for the confusion, this was meant to be about maps where Protoss has a positive win rate against Zerg, so Neo Dark Origin doesn't fit the description - I've updated the original post to make this more clear.
|
I actually did something similar 2 weeks ago, compiling stats from eloboard.com and filter out which maps havve "acceptable" ZvP balance, according to the following criteria:
1. Must have more than 200 games played per matchup
2. All three matchups must have winrates within 5% deviation from 50%
Here are my ZvP results. Keep in mind these stats were from October 15th:
----------MAP------------------------------ZvP Winrate-------Absolute Deviation
1 Neo Sylphid (9,464)---------------------50.07%------------------0.07%
2 Apocalypse (6,728)---------------------50.78%------------------0.78%
3 Polestar (3,590)--------------------------50.93%------------------0.93%
4 Invaders (1,934)-------------------------48.87%------------------1.13%
5 KICK BACK (1,352)---------------------51.15%------------------1.15%
6 Butter (2,310)----------------------------48.67%-------------------1.33%
7 Tempest (1,764)-------------------------48.64%------------------1.36%
8 Radeon (10,818)-------------------------51.58%------------------1.58%
9 Deja Vu (4,418)---------------------------51.66%------------------1.66%
10 Nemesis (2,448)------------------------52.06%------------------2.06%
11 Heartbreak's Ridge (1,768)----------52.13%------------------2.13%
12 Ascension (2,784)----------------------52.18%------------------2.18%
13 Retro (12,454)---------------------------52.19%------------------2.19%
14 Vermeer (15,506)-----------------------52.22%------------------2.22%
15 Neo Dark Origin (5,852)-------------52.25%------------------2.25%
16 Citadel (3,122)--------------------------47.52%------------------2.48%
17 Metropolis (3,194)----------------------52.75%------------------2.75%
18 Polypoid (25,542)-----------------------46.87%------------------3.13%
19 Fighting Spirit (5,194)---------------53.66%------------------3.66%
20 Eclipse (10,180)------------------------53.70%------------------3.70%
21 Pantheon (1,738)-----------------------53.72%------------------3.72%
22 Dominator (5,044)----------------------54.38%------------------4.38%
Since this thread is only about ZvP, I will not bring up the lists from the other 2 matchups. However, I did make an attempt to list out which maps have "acceptable" balance within ALL THREE MATCHUPS (within 45-55% winrates), and I ranked them based on the cumulative deviation. Here are the results:
-------------MAP------------------------------Sum of Absolute Deviation across all 3 matchups
1-----Neo Dark Origin (5,852)-----------------------------------------5.19%
2-----Apocalypse (6,728)-----------------------------------------------6.00%
3-----Butter (2,310)------------------------------------------------------6.54%
4-----Fighting Spirit (5,194)--------------------------------------------7.29%
5-----Radeon (10,818)--------------------------------------------------7.38%
6-----Tempest (1,764)---------------------------------------------------8.07%
7-----Invaders (1,934)---------------------------------------------------8.10%
8-----Eclipse (10,180)---------------------------------------------------8.51%
I know some of these results will surprise many people, so I will only make one comment.
Apocalypse is a very well-respected map. It says as much on Namuwiki, being one of the only 2 maps that are deemed "concept maps", the other being Fighting Spirit - which I'm sure people on this forums will also have something to say, too.
It's unfortunate that Apocalypse was only played in 2 ASL seasons, and even in those seasons, they were constantly avoided in PVZ games.
Also @Soulforged has some nice suggestions for map features such as
1. 10 mineral stacks in main
2. Alternative entrance to main to allow mineralwalking (for PvZ scouting)
3. Open ground 3rd and 4th bases
4. High ground naturals.
|
On October 28 2025 21:34 Kraekkling wrote:sorry for the confusion, this was meant to be about maps where Protoss has a positive win rate against Zerg, so Neo Dark Origin doesn't fit the description - I've updated the original post to make this more clear.
If it's above 50%, it's not good either, is it?
|
On October 29 2025 23:44 iopq wrote:Show nested quote +On October 28 2025 21:34 Kraekkling wrote:sorry for the confusion, this was meant to be about maps where Protoss has a positive win rate against Zerg, so Neo Dark Origin doesn't fit the description - I've updated the original post to make this more clear. If it's above 50%, it's not good either, is it?
you're right in general of course. if the intent is to find the most balanced map, we should look for maps that are as close to 50% win rate in all matchups as possible. this is just as good of an intent, and also quite interesting. for example, just by looking at stats, we should conclude that Apocalypse is better balanced than FS for the PvZ mu. do we understand why?
I think matchup balance on specific maps can sometimes be evaluated through analysis of the strength of specific builds.
for example, think of the 973 build. this build will be present on any standard map. the timing and strength is mostly fixed by things that can't be changed, such as worker mining rates, the hydra costs and build time, etc.
some other parts of this build we can influence via map features, think rush distance; how many cannons protoss can fit at their natural; horizontal or vertical nat entrance; the high/low-ground layout of the natural, etc
+ Show Spoiler +this is somewhat off-topic, but this is the reason why I personally dislike maps with any sort of second-entrance-backdoor, like on Litmus. most of the times, it's just a very random buff to a set of a few specific builds
---
now, since historically most maps have a Z>P bias, I thought it might be interesting to take a closer look on maps with a P>Z bias.
my particular hope was that Protoss players (or high level players in general) might comment on what they feel makes a map good for PvZ. also, Zerg players comment on what makes maps worse for them in ZvP.
it's been said for ages that the game is being balanced by maps. since most maps have a Z>P bias, this balance-by-maps doesn't seem to work that well. thus, we should examine those other few maps, where the reverse is true, i.e. P>Z.
the funny thing is, even the maps listed here for their particular P>Z bias, are at the order of 53% win rate for Protoss. this is just around the average win rate for Zerg vs Protoss for like 80% of all the other maps.
thus if we can't manage to make maps that are 50:50, we would need to regularly include one of these P>Z maps to counteract the inherent Z>P bias which is present on most other maps.
|
On October 30 2025 00:48 Kraekkling wrote:
my particular hope was that Protoss players (or high level players in general) might comment on what they feel makes a map good for PvZ. also, Zerg players comment on what makes maps worse for them in ZvP.
So I would comment there are good ways to balance and bad ways to balance. For example, it's harder for Zerg on Radeon because you can't make a good wall before you build your hatchery. That's because the eggs are too far forward and you can't build your hydra den (which is the best to wall with because hatchery-den make a zealot tight but hydra passable wall) without completing the hatchery. This is a non-sense "balance" path where you're just making it awkward for one race to play (go full muta and THEN make a satisfactory wall? Make a creep colony? Make a shit wall?)
In terms of 973 horizontal walls used to be preferred like on Radeon, but these days Protosses are happy to tuck the forge in so vertical walls seem to be better! So 973 nerfed vertical wall map like Knockout is already pretty good for Protoss in PvZ. You can see that in Tempest as well, these maps are already quite balanced in ZvP
the issue is you want a map that's also good for ZvT and PvT
for ZvT I would want uphill single ramp third bases and safe overlord spots from marines on the way to scouting the natural with the first overlord. I don't want the scouting to take so long you can't get there like in Minstrel which is awful
But on Neo Sylphid there's an issue where you might take your third high ground base, but the tanks can siege it from low ground and even if you have swarm it might be awkward to defend. I suggest when you siege the Nexus or Hatchery the siege position should be close to the ramp, not off to the side to give the defender slightly better chance to get down the ramp to kill them.
So just make a three player Tempest or something
|
On October 30 2025 05:15 iopq wrote:Show nested quote +On October 30 2025 00:48 Kraekkling wrote:
my particular hope was that Protoss players (or high level players in general) might comment on what they feel makes a map good for PvZ. also, Zerg players comment on what makes maps worse for them in ZvP.
So I would comment there are good ways to balance and bad ways to balance. For example, it's harder for Zerg on Radeon because you can't make a good wall before you build your hatchery. That's because the eggs are too far forward and you can't build your hydra den (which is the best to wall with because hatchery-den make a zealot tight but hydra passable wall) without completing the hatchery. This is a non-sense "balance" path where you're just making it awkward for one race to play (go full muta and THEN make a satisfactory wall? Make a creep colony? Make a shit wall?) In terms of 973 horizontal walls used to be preferred like on Radeon, but these days Protosses are happy to tuck the forge in so vertical walls seem to be better! So 973 nerfed vertical wall map like Knockout is already pretty good for Protoss in PvZ. You can see that in Tempest as well, these maps are already quite balanced in ZvP the issue is you want a map that's also good for ZvT and PvT for ZvT I would want uphill single ramp third bases and safe overlord spots from marines on the way to scouting the natural with the first overlord. I don't want the scouting to take so long you can't get there like in Minstrel which is awful But on Neo Sylphid there's an issue where you might take your third high ground base, but the tanks can siege it from low ground and even if you have swarm it might be awkward to defend. I suggest when you siege the Nexus or Hatchery the siege position should be close to the ramp, not off to the side to give the defender slightly better chance to get down the ramp to kill them. So just make a three player Tempest or something
Good points. Are you a fan of KnockOut, Apocalypse, or even Fighting Spirit (despite the notorious spawn location advantages) in terms of ZvT?
There was also a map made by one of the veteran mapmakers, Jukado, who was very active here back in the days, called "Clay Fields". Have you heard of that map, if you have what do you think?
https://tl.net/forum/brood-war/533461-map-4clay-fields-summer-update
I thought Jukado put a lot of thoughts into that map to make sure it's as fair as possible for a bog-standard 4p map, and it was used in a couple BSL seasons, and was played by some Korean pros online for a short while.
|
On October 28 2025 11:44 perfectspheres wrote:
We do things together always.
Can't really do anything with people who think they are doing things by themself.
I am definitely not letting someone else in the bathroom with me. Definitely not something I'll be doing "together" with anyone. Nope! Ain't no frickin way.
|
On October 29 2025 23:44 iopq wrote:Show nested quote +On October 28 2025 21:34 Kraekkling wrote:sorry for the confusion, this was meant to be about maps where Protoss has a positive win rate against Zerg, so Neo Dark Origin doesn't fit the description - I've updated the original post to make this more clear. If it's above 50%, it's not good either, is it?
59% nope
51-53 kinda good
|
Netherlands5185 Posts
I think it's unwise to categorize 4 player, 3 player and 2 player maps in the same pool when looking at what features can make for good PvZ balance.
With 4 player maps you got X-spawn and close-spawn that can have different winrates but still average to a perfect 50% in theory. Different winrates in X-spawn and close-spawn is probably unavoidable, but how much skew would be acceptable there?
|
That might be, but I don't think it is balanced.
The map features highland expansions with unprotected hillside minerals. Protoss can psistorm there and break a stalemate like it happened on ASL 17 Ro24 between Jaedong and Snow. It wasn't a fair fight. HT sniping drones with no defensive capability by zerg is not the definition of balance, it is the same cheese, but in the opposite. Zerg cannot defend there. No other unit except guardians have the same range as HTs, however guardians are still high value targets that HTs would counter effectively.
|
On October 30 2025 13:27 Kanzzer wrote:Show nested quote +On October 30 2025 05:15 iopq wrote:On October 30 2025 00:48 Kraekkling wrote:
my particular hope was that Protoss players (or high level players in general) might comment on what they feel makes a map good for PvZ. also, Zerg players comment on what makes maps worse for them in ZvP.
So I would comment there are good ways to balance and bad ways to balance. For example, it's harder for Zerg on Radeon because you can't make a good wall before you build your hatchery. That's because the eggs are too far forward and you can't build your hydra den (which is the best to wall with because hatchery-den make a zealot tight but hydra passable wall) without completing the hatchery. This is a non-sense "balance" path where you're just making it awkward for one race to play (go full muta and THEN make a satisfactory wall? Make a creep colony? Make a shit wall?) In terms of 973 horizontal walls used to be preferred like on Radeon, but these days Protosses are happy to tuck the forge in so vertical walls seem to be better! So 973 nerfed vertical wall map like Knockout is already pretty good for Protoss in PvZ. You can see that in Tempest as well, these maps are already quite balanced in ZvP the issue is you want a map that's also good for ZvT and PvT for ZvT I would want uphill single ramp third bases and safe overlord spots from marines on the way to scouting the natural with the first overlord. I don't want the scouting to take so long you can't get there like in Minstrel which is awful But on Neo Sylphid there's an issue where you might take your third high ground base, but the tanks can siege it from low ground and even if you have swarm it might be awkward to defend. I suggest when you siege the Nexus or Hatchery the siege position should be close to the ramp, not off to the side to give the defender slightly better chance to get down the ramp to kill them. So just make a three player Tempest or something Good points. Are you a fan of KnockOut, Apocalypse, or even Fighting Spirit (despite the notorious spawn location advantages) in terms of ZvT? There was also a map made by one of the veteran mapmakers, Jukado, who was very active here back in the days, called "Clay Fields". Have you heard of that map, if you have what do you think? https://tl.net/forum/brood-war/533461-map-4clay-fields-summer-updateI thought Jukado put a lot of thoughts into that map to make sure it's as fair as possible for a bog-standard 4p map, and it was used in a couple BSL seasons, and was played by some Korean pros online for a short while.
Same with Radeon and KnockOut.
I think we are approaching it from the wrong perspective. Ratios might be equal, but uneven textures put the defensive player at a disadvantage. Why would this matter? Because distances are measures of cross spawn balance, yet these uneven texture terrains enable closer than birdseye distance areas where the defender is at a disadvantage.
Like, on KnockOut, we saw Soulkey beat Best in the final GF game in ASL19. Therefore map got changed, however it still put Jaedong at a disadvantage vs Mini in the next ASL20 group C 2nd game. It seems like the ditch didn't give advantage to the defending player, on the contrary it made Jaedong's position more open since it restricted mobility for the zerg player while it gave no advantage for him to defend since the ditch blocked sunkens being placed further out into the triangle. Mini could easily move his HTs and Jaedong literally had no place to defend and fall back to.
In Radeon, Flash easily pushes his tanks to the wall in front of the bridge to the expansion and once he is there, Cloud has no way to break Flash's artillery position. Artosis even joked that it might be a future meta to block units from engaging your siege tanks by starting constructions surrounding sieged tanks. It is that imbalanced to put a map feature that shortens the distance into a position accessible from the center of the map. Maps with backdoor mineral walk paths are more fair due to this reason. They don't have centerside exposed highground bases. Their bases are far, giving room for both good defence and the occasional mineral walking backdoor 4 pool surprise added in.
|
On November 03 2025 07:41 Peeano wrote:
I think it's unwise to categorize 4 player, 3 player and 2 player maps in the same pool when looking at what features can make for good PvZ balance.
With 4 player maps you got X-spawn and close-spawn that can have different winrates but still average to a perfect 50% in theory. Different winrates in X-spawn and close-spawn is probably unavoidable, but how much skew would be acceptable there?
Not acceptable, just make more three player maps.
Let's say you 12 Nex in PvT and if the Terran is next to you they can kill your Nexus and you have 45% win rate in this position, but if they are cross spawn you have 60% win rate because you just get ahead.
You have two close spawns at 45% and 1 cross at 60%, OVERALL it's 50% (150% / 3) but why should there be luck? Overall winrate might seem balanced, but that's not good map balance.
A three player map would solve this issue
|
On November 03 2025 06:42 XenOsky wrote:Show nested quote +On October 29 2025 23:44 iopq wrote:On October 28 2025 21:34 Kraekkling wrote:sorry for the confusion, this was meant to be about maps where Protoss has a positive win rate against Zerg, so Neo Dark Origin doesn't fit the description - I've updated the original post to make this more clear. If it's above 50%, it's not good either, is it? 59% nope 51-53 kinda good
Why should Protoss get 53% vs. Zerg? Why not make every map 50% vs. Zerg + wacky maps that could be Protoss or Zerg favored but we don't know (like Roaring Currents last season, what a good addition)
Just don't play the stupid unfair maps that are 60% Zerg favored and you won't need Protoss maps in the map pool.
|
On November 03 2025 07:41 Peeano wrote:
I think it's unwise to categorize 4 player, 3 player and 2 player maps in the same pool when looking at what features can make for good PvZ balance.
With 4 player maps you got X-spawn and close-spawn that can have different winrates but still average to a perfect 50% in theory. Different winrates in X-spawn and close-spawn is probably unavoidable, but how much skew would be acceptable there?
4 player maps and 2 player maps need different balance metrics. Roaring Currents worked out so well because they were far spawn locations. 4 player maps balance out close and cross spawn rush distance with scouting luck, still I cannot say I liked Barracks vs Mini Ro8 finale. Personally, I think it defies map mechanics when players can blind proxy, or cannon rush to win. You cannot do those in Roaring Currents, ever, for instance. Best I could find is that one time Queen cheesed Bisu on the KCM with relentless 9-pool ling rush, but it doesn't qualify as a dedicated rush.
|
|
|
|
|
|
EPT![[image loading]](https://liquipedia.net/commons/images/e/e4/KnockOut.jpg)
![[image loading]](https://liquipedia.net/commons/images/f/f8/Tempest_%28ASL%29.jpg)
![[image loading]](https://liquipedia.net/commons/images/f/f2/Revolver_se.jpg)
![[image loading]](https://liquipedia.net/commons/images/e/eb/Butter.jpg[/img[/spoiler]]<br>BlitzY[spoiler][img]https://liquipedia.net/commons/images/4/4b/Blitz_Y.jpg)
![[image loading]](https://liquipedia.net/commons/images/a/a2/Citadel.jpg)
![[image loading]](https://liquipedia.net/commons/images/5/5e/Invader_%28ASL%29.jpg)
![[image loading]](https://liquipedia.net/commons/images/9/9d/Pole_Star.jpg)
![[image loading]](https://liquipedia.net/commons/images/2/21/Apocalypse_%28ASL%29.jpg)
![[image loading]](https://liquipedia.net/commons/images/6/6c/Neo_Sylphid.jpg)
