Raid Scaling Busted?

[capri sunset]

There's a really simple explanation actually. They want to keep smaller groups in normal and heroic for longer so they get to spend less time complaining about not having any further content to do as 20 man is unfeasible for many guilds on less populous servers.

(I'm not being serious, it's probably *just* negligence.)

[Marrilaife]

It's been an issue since forever flex exists, I remember pugging gorefiend in HFC and it was always more people = easier kill. Most raids worked best with size 14-25, less than that and it's an issue of lack of hands to work on the bitch jobs, and more than that and raid starts having fps issues... Some bosses like Kil'jaeden received aggressive nerfs turning them from "unviable with 10 man" into "quite easier with 10 man than big group", but that was quite late into the tier.

[Craaazyyy]

Raid scaling has been busted since it was introduced in flex mode Siege of Orgrimmar and it doesn't look like it will ever be balanced.

And to think some people want mythic to scale .....

more like ever since they've made 10 and 25 heroics essentially the same difficulty.. most of the time 10m was harder

[kneehidude]

Heroic 10m Horridon in Throne of Thunder had all the same mechanics as Heroic 25m Horridon, and the adds had the same health. The devs didn't believe us until nearly 6 weeks into the expansion when we linked the progression rates for 10m vs 25m, and more 25m guilds had cleared Lei Shen (last non-secret boss), than 10m heroic raids who had cleared the second boss. It killed a LOT of guilds in the meantime.

Couple of notes:

1) Patch, not expansion (as ToT was the 2nd raid tier of MoP).

2) I remember Horridon had to be nerfed early on (but it was literally the first week of the patch, aka normal mode week). Furthermore, even a quick search on wowprogress shows that no, more guilds hadn't killed H Lei Shen (25) than had killed H Horridon (10). Even suggesting that is completely absurd.

[exochaft]

There are ways to universally adjust raids to scale better with raid size, but the problem comes down to the intangibles. For example, I remember there being huge debates over whether 10man or 25man content was the harder content in Cataclysm, and it honestly depended on the fights (still the same with flex raiding). If there are a static number of mechanics that pop out (especially soaking mechanics where immunities work), a larger raid can deal with it more readily. If there are mechanics that require decent spread of your raid, smaller raids generally can deal with it better. In an ideal word, 10man would be harder due each individual having to carry more weight/responsibility compared to a 30man raid, and you could potentially make the boss difficulty (HP/damage/etc.) scale exponentially with each added person to simulate the same level of responsibility experienced by smaller raids. However, as I mentioned, the difficulty of an encounter can be harder with a small raid or a large raid depending upon the mechanics, so no universal fix can really be made.

[capri sunset]

There are ways to universally adjust raids to scale better with raid size, but the problem comes down to the intangibles. For example, I remember there being huge debates over whether 10man or 25man content was the harder content in Cataclysm, and it honestly depended on the fights (still the same with flex raiding). If there are a static number of mechanics that pop out (especially soaking mechanics where immunities work), a larger raid can deal with it more readily. If there are mechanics that require decent spread of your raid, smaller raids generally can deal with it better. In an ideal word, 10man would be harder due each individual having to carry more weight/responsibility compared to a 30man raid, and you could potentially make the boss difficulty (HP/damage/etc.) scale exponentially with each added person to simulate the same level of responsibility experienced by smaller raids. However, as I mentioned, the difficulty of an encounter can be harder with a small raid or a large raid depending upon the mechanics, so no universal fix can really be made.

You're right there's no universal fix, there's the option of paying attention to the individual flex difficulties but that takes us right back to their justification for removing 10/25. I think it's safe to say given that they've failed to deliver on the rest of the talking points of 20 man, the plan was simply to reduce the workload by more or less ignoring smaller flex groups, which of course they've done ever since. Just a bit of a sad situation really.

[Niwes]

It's not hard. It's basically maths a chimp could do. They just don't care enough.

i am sw dev since 20 years. sign that.

[XDurionX]

This is a gross oversimplification of what balance is. If they design a fight in 25 man for 3 tanks, 6 healers, 16 dps, how do you scale that appropriately to 10 man?

That's just pure numbers, if you take mechanics into account it gets even crazier. 10 Man Mekkatorque is proof of this right now. 3 people get shrunk, 3 people have bombs, that leaves 4 people to actually do anything, compared to on 30 man, there are 24 people that can do whatever they want. If you reduce the number of bombs, the fight becomes trivial, if you reduce the number of robots, the fight becomes trivial, the fight is just as hard on 30 man because you have significantly more people that need to learn the fight, however the ACTUAL difficulty of the fight is harder on 10.

Other bosses that have this issue on 10 man are Opulence and Blockade (Basically 4 heal fights for the average group), which strains the dps because they can't be bad.

It's not a simple equation, every boss has to be balanced for both difficulties, and it was just too much work to not even get it right.
In MoP paragons was easier on 10, while siegecrafter was easier on 25.



The hotfixes at the end of mop dealt with this in a smart way as well, instead of it being 14/5 = 2 it was 14/5 = 2 with an 80% chance of a third.

Most of what you said is true, HOWEVER you do only have 2 bombs on smaller raids.

[Demeia]

Everything works with numbers. At some points numbers become so small that it's either too much or too little. Like the hex that you mention, having only 1 would be trivial and 2 seems like a lot for that amount of people/healers. In pure mathematic balancing the correct number would probably be 1.5 or some shit, which obviously can't happen in reality, so yeah, flex raiding has its issues, it's just the nature of the beast, the ony people we can blame are those who asked for it, not the devs, they do what they can with what they have.

Yeah, I'm sorry, I agree with you that this is not a trival problem; but I also agree they are not trying very hard to solve it. In the above example, for instance, you could alternate the number, sometimes giving 1 sometimes giving two, averaging out to 1.5. A smaller raid could possibly deal with that. They just seem not to care. I think their internal argument is: "this raid should just PuG some more players". Which I might myself do, but my raid has never been willing to take such chances, and honestly if you're advertising flex, you really should not. I get that it might be hard at first to do 10m but the fact that they sometimes never make adjustments is extremely hard to justify.

They just need to work harder on this.

[Meiffert]

I also agree they are not trying very hard to solve it. In the above example, for instance, you could alternate the number, sometimes giving 1 sometimes giving two, averaging out to 1.5. A smaller raid could possibly deal with that.

How would you deal it with that any better if you don't have 2 dispels available?
Now you randomly die to the hex half the time instead of every time?

If you have 2 healers on each side (so that you can get both Amethyst and Sapphire buffs on each side), you can dispel both hexes easily.
If you only have 1 healer on a side, you should pair them with a another class able to dispel magic (such as a priest or a warlock).

Admittedly this fight doesn't scale very well, not just because of the hexes, but also because of gems.
I agree that the flex scaling could be better on some encounters, but this example alone shows that it's not as easy as many players think.
At the end of the day I'd rather see a few awkwardly scaled bosses in extreme cases (very small or very large groups) rather than limit encounter design and miss out on interesting mechanics and fights.

[BigToast]

It's not hard. It's basically maths a chimp could do. They just don't care enough.

How do you handle scaling in your mmo?

- - - Updated - - -

If you only have 1 healer on a side, you should pair them with a another class able to dispel magic (such as a priest or a warlock).

Oh yea, players other than healers can dispel also!

[Freshouttajail]

Yeah, I'm sorry, I agree with you that this is not a trival problem; but I also agree they are not trying very hard to solve it. In the above example, for instance, you could alternate the number, sometimes giving 1 sometimes giving two, averaging out to 1.5. A smaller raid could possibly deal with that. They just seem not to care. I think their internal argument is: "this raid should just PuG some more players". Which I might myself do, but my raid has never been willing to take such chances, and honestly if you're advertising flex, you really should not. I get that it might be hard at first to do 10m but the fact that they sometimes never make adjustments is extremely hard to justify.

They just need to work harder on this.

To be fair they don't really NEED to work harder on this. They're not losing players over this, or if they are, it's not enough to justify putting much time/ressources into fixing this, it's harder than you think. "Randomizing" things like that isn't the way to go, people will just complain about the RNG aspect of it or they won't expect it so it could cause wipes that wouldn't happen if you got the "easy" number.

I think the best way to fix issues like that is to not let the issue happen in the first place. These issues happen because of flex raiding, not because they don't try to balance flex. Flex should be more like a tiny flexibility. What I mean by that is like, instead of allowing 10 to 30 man raiding, it should be more like 18 to 22. I really believe the game should have a stricter model where guilds are expected to have a 20 man roster (or whatever number really, just something to base difficulty on) and have a slight flexibility to allow for missing people or extra online.

People act like it's a curse to be a 10 man raiding guild. Just recruit, like wtf lol. Or merge with another 10man raid. There's plenty of options. Of course there's the "but I don't like it when there's too many people", well guess what, people who don't like racing games just don't play racing games.

[Osmeric]

The reason for sublinear scaling in flexible modes is to keep raid groups from kicking their worst members. If the scaling were exactly proportional to raid size then the encounter would get easier, at least numerically, if your dropped anyone who was below the raid average.

Of course, this means small raids are handicapped. Small F&F guilds will face harder content, which is probably not something that the designers should want.