Depth vs. Complexity in Raiding

a.k.a “Why 10 Player Formats Are Better”

Many pages have been devoted by fans of Warcraft’s raiding scene to whether one raid format is harder than the others. This conversation started way back in the era of Wrath of the Lich King, when there genuinely was a built-in difficulty difference between smaller and larger raids. 25 player raids were slightly harder, and rewarded slightly better loot as a result. When we moved to Cataclysm (and, later, Mists of Pandaria), the difficulty levels of each raid size were explicitly stated to be equal. Naturally, not all players agreed and the “25man players are the only real raiders” attitude became both less easy to argue and more infuriating to hear.

Now, we’re in the age of Flexible raiding and 20-man (“one size fits nobody”) Mythic raiding. I sincerely and desparately miss 10-player raiding, and I’m going to go out on a limb and tell you that it was better. Not “harder”, but better.

Cue the QQ in 3… 2… 1…

While I realize that the whole concept of “better” is inherently subjective, and therefore any personal opinion on the differences between formats isn’t wrong per se, I do want to put forward what I think is a powerful argument for 10-player raiding over higher raid sizes. In my experience, this reason alone made 10-player raiding significantly better for me as a player.

The whole discussion is rooted in the concept of elegance of design, something which I think is probably one of the most important things when discussing a videogame. What makes game design elegant? For the purposes of this discussion, I’d recommend that you watch this video from Extra Credits about Depth vs. Complexity – I’ll be basing my discussion on the general ideas presented there.

Depth/Complexity = Elegance

Let’s summarize the discussion in the video I linked, first. Effectively, a well-designed game gets the maximum amount of depth from the minimum amount of complexity. Good games, we can all agree, are deep – they have a large number of meaningful decisions for you to think around as a player. However, there is a nontrivial relationship between depth and complexity – you can’t get a deep game without introducing some complexity, but since the actual depth of a game is dictated by the degree to which you can actually think around it, too much complexity effectively cuts off the depth of the game.


Possibly the single most important part of Extra Credits’ discussion on the topic of depth. I think that the latter half is probably the most important.

There are different forms of complexity, which are discussed in the video I linked, so let’s reduce the discussion a bit. We’re going to assume that we are playing each raid size with the same UI – whatever that may be – with the absolute minimal changes between raid sizes (e.g. you now have 30 players on Grid instead of 25). We’re also going to assume that when one brings players on board to go up to higher raid sizes, the average raid skill is kept constant.

So, the central statement of my argument is that raids are better when they are elegantly designed – the best raid size is the one which gets the player the most depth with the least complexity.

Now, in order to drive home why I think 10-man is a much more elegant raid size than 20-player Mythic, we’ll go through the reasons that I think more players makes the raiding experience less deep.


Let’s face it – larger raids need proportionally higher organization  to be successful. Organization is a task which drastically increases the complexity of a raid (especially when you factor in managing your raiders, roster, sub-teams, role leaders, and so on). This is simply a jump in complexity of raiding which results in no noticeably deeper experience for the player. It’s straight-up just a bad thing for the raid, since it’s a jump in the bad thing without a corresponding increase in the good thing.

You can (perfectly validly) argue that most players won’t be involved in the organization of a raid, and I’d agree to the extent that they’re not asked to dictate strategy or to manage the raid roster. On the other hand, raid management in larger raids tends to be stricter than in smaller raids due simply to the fact that more players can screw things up. Everyone is involved in raid organization at that level, and I think it doesn’t make the experience better for anyone involved.


I wonder if you can tell what this was meant to represent in the Extra Credits video. Looks… familiar…


With 10 regular players, one gets to know each one of them well and learns to predict and understand their actions without explicit communication. The level of personal interaction is both extensive and meaningful in low raid sizes – a good example of depth. However, as more players are added to the raid, you see proportionally less personal understanding between the raid. I don’t mean that there’s none of this in a large Mythic raid, but the importance and extent of this component of raiding is a lot lower than in smaller raids. This was my experience in Cataclysm, in Pandaria, and now in Warlords.

So, in larger raids the depth of the experience might increase marginally (each player might have marginally more decisions to make based on personal understanding), but these decisions are both less important and at the expense of significantly elevated complexity. The Depth-Complexity ratio from interpersonal understanding is better in lower raid sizes.

Central to this section is the fact that individual personal connections matter less than before, and I think there’s a more general truth here.

Individual Decisions Don’t Matter

This is, I think, the best argument for 10-player raiding being more elegant. The number of player decisions per raid is approximately the same between raid sizes, but in terms of how much they matter, each decision a raider makes in a 10-man raid is proportionally much more significant than in a 20-man raid. While the individual decisions do still matter somewhat, there is much lower impact from a missed heal or a raider standing in fire when there are simply more players.

To exacerbate the problem, each decision made by a player (and in particular healers) are significantly more complex in larger raid sizes. They’re more complex because there are many more players in the raid… but they aren’t actually deeper in any way. So, overall ones decisions in a raid are more complex and less significant without at any point becoming experientially deeper.

Of course, when each decision matters less and less, you start to get to the point where First Order Optimal Strategies become just Optimal Strategies. (More info: this other Extra Credits video).

AoE Spam

Now we’re going to talk exclusively from the point of view of a healer. We saw in the Siege of Orgrimmar the phenomenon of Smart Heals being definitively the Optimal Strategy. To put it another way: the best possible strategy in late SoO was to press the keys that did the most healing automatically (Healing Stream Totem, Chain Heal, Healing Rain) regardless of the situation or context, and fill in around that with whatever else you had.

Naturally, SoO was a particular outlier due to the extreme strength and proliferation of Smart Heals. Still, there is a general case to discuss here which ties in with the previous point that your decisions matter comparatively less. When there are more healers, there are more heals flying around and you tend to get into a race to get healing done before anyone else. In that case, a First Order Optimal strategy is to just spam whatever AoE spells you have. The problem is that as the raid size goes up, the power of this strategy increases dramatically! It is less deep play to spam AoE heals regardless of context, a strategy which is enabled by high numbers of healers, and so the fewer healers you have in the raid the better your depth to complexity ratio becomes.

Of course, there are other factors at play. In the current environment – early to mid tier WoD healing – spamming your AoE heals is not entirely a good idea due to the way damage patterns and mana regen coincide. On the other hand, some of my co-healers are already complaining of not being able to single-target heal fast enough to use up all their mana. Seriously.

We saw a very similar thing happen back in Cataclysm’s Tier 11 to Tier 12 transition, at which point large raids started to become all-you-can-spam festivals whereas smaller raids didn’t suffer as much.

Why Elegance Matters

To summarize the previous section; raiding with a larger number of players seriously reduces the ability for a player to make meaningful choices during gameplay. It adds complexity and reduces depth, and for that reason smaller raiding formats are significantly better experiences. My contention now is that elegant = better.

I think that the elegance in design of raiding matters quite a lot. Raiding is something that the player puts a lot of work into and expects to be rewarded for. Where elegance comes in is that if you have been playing a format with a high depth to complexity ratio, it feels like it was you making the decisions with your own agency.

That agency is important to me, because once you’ve completed a boss that sense of agency is what makes raiding really worth it. The more I feel like I’ve been the active part of the raid – someone that made important and worthwhile decisions – the better. I want to feel like the success or failure of my raid depends on my skill and decision making as a player. The relative elegance of the raiding format, as we’ve discussed here, dictates how much of that sense of agency I get.

So, at least for me, 10 player raid formats are significantly better than 20-30 player formats. This discussion has nothing at all to do with inherent difficulty of the different raid sizes, so I hope you can see why it’s relevant to today’s age of flexible raiding.

Of course, you might disagree. There are plenty of places where you might say that you feel your experience of larger raid formats differs from mine, or that you have a different source of enjoyment from your raids. You wouldn’t be wrong, either, because this kind of thing is quite subjective.

About stoove

A physicist, researcher, and gamesman. Likes to think about the mathematics and mechanics behind all sorts of different things, and writing up the thoughts for you to read. A competent programmer, enjoys public speaking and mechanical keyboards. Has opinions which might even change from time to time.
This entry was posted in Game Design, World of Warcraft and tagged , , , , , , , , , , . Bookmark the permalink.

3 Responses to Depth vs. Complexity in Raiding

  1. Talarian says:

    Excellent post, good read.

    Question for you, however, just to explore your hypothesis. Why not 5 players? By your logic, that should be even more elegant. Is there an inflection point somewhere where the loss of depth as you drop players outpaces the loss of complexity (assuming both variables scale with the number of players, just with different slops?). And if so, is there a way for us to measure this?

    • stoove says:

      I think you’re right. It did occur to me that this was the case while I was writing, but honestly I didn’t feel like I could fit a discussion in the main text. Effectively, I think that the complexity increase from 5 to 10 is comparatively small and affords a large depth increase to compensate. I actually picture complexity as scaling exponentially with player number and depth perhaps polynomially.

      Another factor here is that raiding is fundamentally limited to a minimum of 10 players at the moment. There are some interesting differences between 5-man content and 10+ content in terms of mechanics, and I think that some of the discrepancy is certainly accounted for in this framework. However, I’m not sure that this is a measurable type of discussion – objectivity is hard in this kind of thing.

      It’s funny that you talk about 5-player, actually, because I thought the bigger topic would be 15-player formats. From what littler experience I have with that kind of format, I think that the complexity increase from 10 to 15 is actually slightly outweighed by the depth increase, especially back in SoO when that afforded you an extra healer. 15 is probably the inflexion point in this graph, but we have so few samples from that region in comparison that I’d be reluctant to declare it de-facto better right now.

      Thanks for your comment and kind words! =)

  2. Pingback: Elite: Boring – Impact of always online covered by a single picture | El Dobablos

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s