Class tuning

To the World of Warcraft Development Team,
I’m writing as a Brewmaster and Fury player in Midnight Season 1 with feedback on class balance and how tuning decisions are being made.

On the philosophy of balancing:
The core problem is that there’s no clear target driving tuning decisions. Buffs and nerfs happen patch to patch without an obvious benchmark, and it shows.

My suggestion: use Raidbots/SimulationCraft as the foundation for balance targets. Sim DPS output is the closest thing we have to objective truth for what a spec should be doing at a given gear level. Set a target number, sim every spec against it, and close the gaps. There are obstacles to making that work perfectly — fight style variance, skill expression, real-world versus sim conditions — but those are just problems to solve on the way to the goal. Without a defined goal, you have no balance. You’re just guessing.

On item level and the player population:
Balance should not be tuned around max item level characters. The vast majority of players will never reach the seasonal ilvl ceiling. The players who actually care about tuning — the ones pushing Mythic plus and progression raids, the ones who feel these changes — are mid-gearing characters working their way up.
A reasonable and realistic balance benchmark is item level two fifty. Players who are engaged enough to care about balance reach this threshold quickly. A player brand new to Midnight can hit two fifty five within just a few days of starting. That makes two fifty a natural breaking point — it’s achievable, it reflects the engaged playerbase, and it’s a concrete number you can actually sim around. Balance the specs there. Everything above that is a bonus, not the target.

On transparency and player insight:
When you make tuning changes, explain why. Show us the target, show us the reasoning, and show us how each spec compares to that target. Right now there’s no visibility into your decision-making process, which makes it impossible for players to understand the direction or trust that there’s a coherent strategy behind the changes.
More importantly, ask players for insight. We’re running the content every day. We feel what works and what doesn’t in ways that sims and metrics alone can’t capture. Players have ground-truth feedback that could inform much better tuning decisions. Open the conversation instead of closing it with opaque patch notes.

On current Monk tuning:
Windwalker has been passed over in multiple tuning passes while Warriors received a ten percent damage buff and Vengeance Demon Hunters continue to benefit from unaddressed bugs inflating their output. Fists of Fury remains undertuned relative to comparable rotational abilities on other specs. Windwalker’s rotation is also more punishing than most, meaning the gap between sim ceiling and real-world performance is wider than it should be — which compounds the problem.
Brewmaster’s Master of Harmony tree took meaningful nerfs without clear justification, effectively making Shado-Pan the only competitive hero talent choice and removing real decision-making from the spec.
The ask is simple: define a target ilvl, sim every spec against it, tune toward parity, and talk to us about why you’re making the decisions you make. Right now there’s no visible goal — and without one, balance will always feel arbitrary.
— Mike

oooof I’d like to tune a few line breaks into this post.

6 Likes

Blizzard already has telemetry data from every person playing the game. This is a much better balancing metric to use than running a bunch of sims. The data includes players of all skill levels, all content types, all key levels and raid difficulties and encounters, which are elements that should be accounted for.

Sims are a robot in a vacuum. Much like the AI that wrote this post.

Sorry Brew got nerfed, I guess?

3 Likes

Player skill level shouldn’t be a factor when balancing damage between specs and classes. The goal is balance and balance means that any spec, in the same gear, in the same scenario, should be putting out the same damage as any other spec. A Mage and a Warrior in identical gear should be doing the same damage in the same fight. That is currently not the case, which tells us the current balancing method isn’t working.

Factoring skill level into balance decisions actually works against that goal. It pushes players toward whatever’s meta and away from anything that isn’t, because the “balance” gets calibrated around what top players squeeze out of certain specs rather than what the spec is actually capable of in equal conditions. A proper balancing strategy needs to account for all skill levels, not optimize around the top end.

To be clear collecting data across every player in the game is great, and I’m not arguing against that. The issue isn’t the data itself, it’s how it’s being used. The current approach clearly isn’t translating that data into actual class/spec parity.

We also need to be honest about the dungeon vs. raid problem. You can’t balance damage for one environment without considering the other, or you fix M+ and break raid (or vice versa). Any balancing framework has to weigh both scenarios together.

So let’s actually define the problem we’re trying to solve. If the goal is “every spec performs equivalently in equal gear and equal conditions,” then the current method isn’t getting us there. Sims as a balancing input is a proposal not necessarily the best solution, but I’m putting it out there to make the bigger point: what currently exists isn’t working, and we need to talk about what would.

Incorrect. Specs have strengths, weaknesses, and preference. The goal is not to homogenise every damage pattern and encounter type, it’s to ensure everyone has the opportunity to feel powerful doing something while making sure others don’t pull too far away elsewhere.

Also ofc specs are balanced around what people who can play them can do Vs what people who can’t.

It absolutely should be, because you’re playing a game designed to be played by humans. What kind of nonsense is this?

There are so many variables factors that go into damage throughput: Target count, damage profile, available buffs, cooldown timings, uptime availability/optimization, as well as the moment-to-moment gameplay baked into many specs such as proc chance.

Part of what makes the game enjoyable is steering around and mastering these variables.

Wanting such homogenization is counter-productive to making a game that is fun to play.

This is what got Unholy DK massively nerfed and they had to revert it in less than 48 hours. lol

Perfect example why you do not tune based on sims.

realistically they should start there internally and go up/down accordingly based on desired strength and weakness, imo
at least, i feel like that’d be easy as a starting point if I were doing it
but im not a game dev so there’s probably variables im not considering

1 Like

If the goal is to make sure every player “feels powerful,” that’s also a failed framing.

Players at the bottom of the DPS meter are never going to feel powerful regardless, because the WoW gameplay loop is built around gear progression gear is what makes you feel powerful. So the reward for grinding gear should be feeling stronger than the player who hasn’t done that grind, not feeling stronger because you happened to pick the spec Blizzard tuned up this patch.

The “every spec should have a scenario where it feels powerful” design philosophy sounds good on paper, but it’s not achievable. There are too many specs to carve out a unique power fantasy scenario for each one. Saying “Mages are good at range, Melee just have to struggle with uptime” isn’t balance it’s papering over inherited tuning weaknesses with situational excuses.

The realistic, achievable goal is simpler: balance every spec to put out the same output in equal gear and equal conditions. Then let the gear grind be the thing that differentiates players. If I’m out-DPSing someone, it should be because I put in the work to get better gear not because I rolled the spec that got buffed last reset.

That actually opens the game up to a more diverse player base, because suddenly every spec is viable in every scenario. People can play what they enjoy instead of what the meta forces them into, and the gear chase the actual core loop of the game becomes the thing that makes you feel powerful.

Except if you want specs to play differently that’s impossible. Some well favour different fight lengths, different target counts, some are more mobile, some have hard casts that pay off if they have the opportunity to plant.

Homogenisation is not the direction to go if you want an actually enjoyable and varied game.

We don’t actually know why or how Blizzard chose to nerf Unholy DK in the first place, and they also didn’t share any insight into why they reverted the change after applying it. That alone makes it pretty clear they don’t have a firm grip on what they’re doing here.

So when you say “perfect example of why you don’t tune based on sims,” you’re assuming you’re Blizzard that you know how they were gathering their data and why they pulled the change back. You don’t. None of us do, because they didn’t tell us.

Blizzard has demonstrated repeatedly that they don’t have balance figured out. So pointing at one of their reversals and using it as evidence against sim-based tuning falls apart pretty quickly when you actually think it through. You’re using a Blizzard mistake as proof that an alternative approach is wrong, without any visibility into what method they were even using or why it failed.

A few minutes of logic against that argument and it doesn’t hold up.

And you think you do?

:rofl:

Yes it’s a good starting point. Think of it as a factory reset.

Let’s at least start here and rebuild from scratch, with a clear methodology for how we tune DPS specs specifically. Tuning has gotten so complicated that whatever Blizzard is currently doing clearly isn’t working. Reset, start fresh, and slowly and intelligently navigate the progression of the tuning philosophy from there.

And to be clear this should be a dynamic, adaptive philosophy. Not a rigid framework set in stone. It can and should evolve as we learn what works.

I’m just starting the conversation here, but consider this: we already all rely on Raidbots to tell us which piece of gear does more damage. We can plug two specs into the same scenario with the same gear and immediately see that one is doing way more or way less damage than the other. That’s not balanced. The data is right there, and it’s been right there.

So let’s start here. Not end here start. From this baseline we can fold in all the other opinions and approaches the community has to offer and actually move the conversation forward.

prot war gets nerfed cause revenge damage over it being spent on ignore pain

No, I don’t claim to. But I do see evidence that Blizzard doesn’t either and I play the game, and it’s not balanced.

There’s an entire community built around tier lists, and most of us have just accepted that as the way it has to be. I’d argue it doesn’t. That acceptance is part of the problem.

I’m proposing new solutions. I’m not saying I have the answer I’m saying let’s open our minds to other possibilities instead of defending a system that clearly isn’t working.

Which are bad and results in a boring game.

It’s a bit more complex than just doing sims.

Some specs scale wildly with certain secondary stat values.. so if they balance things with everyone having X stats and X ilvl, then things won’t be balanced evenly throughout the gearing process… and you could have some specs that don’t even function until they’re basically max level gear… which makes those specs really bad for 90% of the people who play them.

The other component of this, is that what you can do on a target dummy and what you do in actual content are two very different things.

I think they need to work on making sure that classes work are fun to play in all content – and some of that involves damage profiles.

If you can do 500 million in one big burst, once every 2 minutes, but you’re useless outside that window – your solo experience is going to be absolutely miserable… and your dungeon experience will be miserable for 90% of the time. Even if it averages out on overall damage – you’re probably going to be bored most of the dungeon… then if you mess up your 2 minute CDs, it’s going to feel even more punishing.

I don’t think Blizzard actually has any good metrics for playtesting these kind of “feel” checks… or the people they’re taking feedback from are really dropping the ball in what they’re giving feedback on / testing… and honestly at this point I’d rather they prioritize making things feel good first before worrying about making sure they can all do exactly the same damage on a target dummy over the span of 5 minutes.

Either way though – whatever they’re currently doing is pretty flawed too.

Post invalidated from the rip. Post from your supposed main Brewmaster or Fury warr

Yeah, but barely. ‘Tis but a flesh wound.

lol like that matters damn trolls