Banner and Crosspost

December 11, 2019

Brutal+ Analysis

Brutal+ Analysis

Brutal+ is a series of additional difficulties introduced in the recent patch 4.11.0. I want to first look at how it works, how it compares to weekly mutations and at other interesting data. After that, I'll add my own commentary and suggestions. For detailed information on individual mutators check my mutator compendium.

How it works

Each mutator has a difficulty score (e.g., Barrier mutator has a score of 2), and each Brutal+ difficulty has a difficulty score range (e.g., Brutal+1 has 8–10 range). The sum of difficulty scores of all mutators has to fit into the difficulty score range. For example, Fear (3) + Purifier Beam (2) + Outbreak (3) adds up to the total score of 8. And that fits into Brutal+1 with its 8–10 range.

Another restriction is on the number of mutators, but it's the same for all Brutal+ difficulties. You can get between two and four mutators, no more, no less.

These values have been changed in patch 4.11.3. I will post a new table and mutator frequencies soonTM.

Difficulty +XP% Max XP Min Points Max Points Mutators
Brutal     +100% 49.5k
Brutal+1 +175%
66.0k
8
10
2 – 4
Brutal+2 +200%
71.5k
11
12
2 – 4
Brutal+3 +225%
77.0k
13
14
2 – 4
Brutal+4 +250%
82.5k
15
16
2 – 4
Brutal+5 +275%
88.0k
17
18
2 – 4
Brutal+6 +300%
93.5k
19
20
2 – 4
(*Max XP calculated with a random map bonus and bonus objectives completed. No warchest, holiday or stimpack bonus.)


Now for individual mutator difficulty scores:
1
Concussive Attacks Photon Overload
Eminent Domain Shortsighted
Evasive Maneuvers Temporal Field
Life Leech Time Warp
Orbital Strike
2
Alien Incubation Mineral Shields
Barrier Moment of Silence
Darkness Purifier Beam
Hardened Will Scorched Earth
Inspiration Speed Freaks
Laser Drill Twister
Long Range Walking Infested
3
Aggressive Deployment Lava Burst
Diffusion Missile Command
Double Edged Outbreak
Fatal Attraction Self Destruction
Fear We Move Unseen
Going Nuclear
4
Blizzard Magnificent
5
Avenger
Power Overwhelming
Micro Transactions Slim Pickings
Mutually Assured Destruction Void Reanimators
6
Kill Bots Minesweeper
7
Black Death Polarity
Just Die! Transmutation
8
Propagators
10
Heroes from the Storm Void Rifts


So a result can be something like this:
  • Brutal+1 (8): Diffusion (3) + Purifier Beam (2) + Shortsighted (1) + Twister (2)
  • Brutal+6 (19): Heroes from the Storm (10) + Avenger (5) + Inspiration (2) + Barrier (2)

Frequencies

Now let's look at how likely you are to see individual mutators at each difficulty, and how likely you are to see a certain number of mutators.

For this I generated 200 000 mutations in each difficulty. There is still some error – as it can be seen from the variation between mutators with the same difficulty score (they should be equal). Red-blue histograms were fully generated in-game. The rest of figures were created in Python with Matplotlib.

Low-tier mutators (on the left) are represented equally in Brutal+1. Higher tier mutators are less likely to show up. This is because the game tries to make a game with four mutators. There are weights specifying that 55% games should have four mutators, 40% three mutators and 5% two mutators. Lower tier mutators are more likely to fit nicely into four mutators and 8–10 difficulty score range. However, as the difficulty increase, we will see this shift from lower tier mutators to higher tiers – as the game has to fit more difficulty points into the same number of mutators.

The distribution shifted to higher tier mutators. In Brutal+6, the chance to see Heroes from the Storm, or Void Rifts, climbed up to 25% (compared to 0% on Brutal+1). The lowest tier of mutators dropped from 8.5% on Brutal+1 down to 3.1% on Brutal+6.

Chances for the number of mutators were affected as well. The chance of four mutators increased from 55% to 87% between Brutal+1 and Brutal+6, and the chance for two mutators dropped from 5% down to 0.07% with only one possible mutator combination – Void Rifts + Heroes from the Storm.

Brutal+ vs weekly mutations

It might be interesting to see how this stacks up against weekly mutations. This is a similar figure that shows how often these same mutators were used in weekly mutations. Repeated and holiday mutations were excluded.

Barrier, Speed Freaks, Self Destruction and Just Die! were the most used mutators (11 occurrences). But overall the distribution looks roughly constant across mutator tiers – favoring neither easy nor hard mutators. There are five mutations with only one mutator, but all of them are actually Wheel of Misfortune. They could be technically counted as three-mutator mutations as well.

Now let's assign Brutal+ difficulty scores to weekly mutations, and see how they are distributed.

And divide weekly mutations into Brutal+ difficulties.

One third of weekly mutations have simply too low difficulty score to be even in Brutal+ (~34.1%). Another third would fit into Brutal+1 (~34.1%), and the rest belongs into higher Brutal+ difficulties. Only a single mutation would be in Brutal+6"Cold is the Void" with Void Rifts, Void Reanimators and Blizzard. "Memorable Boss" mutation (Heroes from the Storm, Moment of Silence on Void Launch) didn't make it to Brutal+6. But neither of these mutations can be played in their original state due game patches.

If we look at the number of mutators in weekly mutations, roughly one third have two mutators and two thirds have three mutators. For individual difficulties this is very similar.


Let's compare Brutal+ difficulty score for weekly mutations to public voting (collected by CtG).


It correlates pretty well. The main outliers are "Burning Legion" (one of the first difficult mutations with Void Rifts, Avenger and Scorched Earth), "Thunder Dome" (Heroes from the Storm, but on a fast and easy mission). "Night Drive" and "Graveyard Shift" are two mutations with Afraid of the Dark mutator on Cradle of Death and Dead of Night respectively.

Commentary on Brutal+1

A lot of it comes down to two important questions:

1. Is the current difficulty correct for Brutal+1?

It's a difficult question. You can never be sure what the intended difficulty is. But we can lean on few things:

  • It's made to be played in the matchmaking queue with a random partner (and without the retry function).
  • It's the next difficulty after Brutal. It's described as having "little extra challenge". There are also five additional difficulties beyond this one.

As for the first point, there are various sources of randomness (ally, map, enemy compositions and mutators). If you make the difficulty too hard, there isn't enough room for improvisation in sub-optimal situations. For example, getting Kill Bots in already difficult mutation with commanders that can't deal with them effectively means that you will lose no matter what. That's not very fun. However, if the mutation is easier, there might be room for improvisation. Lowering difficulty reduces randomness that's affecting the outcome of the game, but you still get all the extra variety from random elements.

Regarding the second point, currently there is a big gap between Brutal and Brutal+. Many players are left in the void where Brutal feels too easy and Brutal+ too hard. Considering there are whole five additional difficulties beyond Brutal+1, it would make sense to reduce this gap. This would cover the playerbase better and provide clearer way for players to improve and challenge themselves.


2. Is it better to target 3–4 or fewer mutators per game?

This question is separate and still valid no matter what difficulty you prefer.

Weekly mutations target 2–3 mutators which is a great balance between variety and complexity (~ how many additional rules players have to keep track of). Complexity is usually seen in game design as a cost to a feature, a cost that should be kept low. You don't want to overwhelm players with numbers and rules. Instead, you want a deep system that's simple on the surface.

Getting four mutators seems counterproductive in Brutal+1. You are overloading players with additional complexity. Having fewer but more powerful mutators leads to a better result than more mutators that are less memorable.

Let's consider two mutations:

  • Shortsighted (1) + Mineral Shields (2) + Fear (3) + Blizzard (4)
  • Just Die! (7) + Outbreak (3)

Which one of these options seems more interesting and memorable? Can you easily imagine how the game will play out, and what you have to do in the first mutation? Both have the same difficulty score (10).

Improving Brutal+1

It understandable that on higher Brutal+ difficulties we will see more mutators. All that difficulty score has to fit into few mutators somehow. But on Brutal+1 the complexity can be decreased. We can reduce the number of mutators without making any mutators over-represented.

In Brutal+1, the game is currently favoring low-cost mutators in order to get 3–4 mutators.

I would reduce the difficulty range of Brutal+1 from 8–10 to 6–7, and reduce the number of mutators to 1–3 (weights: 10%, 50%, 40%). Even on this difficulty, you can still get fairly challenging combinations (e.g., Avenger + Speed Freaks; 7 score). It's possible to get 1-mutator games (e.g., Polarity), but majority of games will have 2–3 mutators like weekly mutations.

There is a spike in the figure for Blizzard and Mag-nificent mutators as their 4-point cost fits well into games with 2–3 mutators. This could be avoided by tweaking numbers (e.g., loosening target difficulty range), but ~0.9% increase isn't a significant problem.

Rewards

Ideally, you could get some other rewards for beating higher levels of Brutal+. Bonus experience is good, but nobody is playing Brutal+ for experience. And the experience alone doesn't show that you have overcame any challenge. Playing on casual can get you the same ascension level just as well. Having some rewards would also encourage players to form parties and engage with the mode more.



Concept for co-op portrait border


Achievements are probably not a good option. Completionist players would go crazy if they couldn't get their 100%. However, other rewards are even better – portraits, portraits borders, banners, commander loading screen images, etc.
Alternative commander loading images?

Mutation Design

This mode suffers a bit because many mutators were designed for weekly mutations. Unlike Brutal+, weekly mutations work more like a "puzzle" where you look at the current map and mutators, and choose the best commander and masteries for it. In that situation, mutators can work almost in any way:

  • extremely difficult for certain commanders (Kill Bots, Black Death)
  • generally annoying (Concussive Attacks, Blizzard, Afraid of the Dark)
  • go against the very things that make StarCraft fun (Micro Transactions)
  • etc.

The only exception is Vertigo mutator. It turns out making people physically sick isn't good even if it's a weekly mutation.

Even if some mutators are better than others, they can work pretty well in weekly mutations. However, if you get them at random and push difficulty to the limit, you will see issues crop up. It's a similar situation as with Wheel of Misfortune. I wrote about that as well, and suggested making it easier to reduce outcome randomness.

~ ~ ~

Brutal+1 could be made easier which alone would help. But as for other difficulties, some mutators could be tweaked more to provide better experience (e.g., Fatal Attraction), while others could be made to appear less often (e.g., Micro Transactions).

Ideally, you could select commanders and masteries in-game after the mutators are known. This would combat most of the negative outcome randomness. However, that would require at the very least reskin of Blizzard's commander select screen currently used for testing. I have created my own reskin, but it's still not up to Blizzard's quality.

Public Reception

Brutal+ mode has received some criticism after its release. In my view, it's mostly because how it's framed. If it was called something like "random mutations", there would much fewer complaints. With "Brutal+" name, everyone already had their own idea how to increase the difficulty. Personally, I enjoy it, even though Brutal+1 could be much better with just few tweaks to its parameters.

Design-wise, using mutators isn't inherently bad. There are few ways how to increase difficulty, and all of them have their own significant issues. Using mutators happens to be the easiest to implement and the most scalable solution, and that's likely why we got it. It's also worth noting that randomness and imbalance will inevitably appear if you increase difficulty in Co-op in any way.

I will post about my Survival Mode for Co-op this week. It takes another route to increase the difficulty. And I hope I will eventually finish my asymmetric 2v1 and write about it, which is yet another way to increase difficulty in Co-op.

Other notes

  • Retry function is great.
  • Having matchmaking queue only for Brutal+ is a good decision in my view.
  • Check out another well written Brutal+ Analysis by SwankyPants
  • It would be good to have a different loading screen overlay indicating it's Brutal+ (similar to weekly mutations). The same thing should be done for your match history as there is currently no indication of Brutal+ mode.

Get new posts through email:

Recent posts

Recent posts

Copyright

© Blizzard Entertainment, Inc. All rights reserved. Wings of Liberty, Heart of the Swarm, Legacy of the Void, and StarCraft are trademarks or registered trademarks of Blizzard Entertainment, Inc. in the U.S. and/or other countries.
Powered by Blogger