Raid Management: The Magic Number

Managing the roster of raiders is one of the most difficult parts of running a mythic raiding guild in World of Warcraft. As long as a guild has enough happy raiders showing up to raid every week, the guild will live on and the group will, in all likelihood, defeat more and more bosses. If a raiding guild fails to sustain its roster, it will fade out of existence. This happens enough in Warcraft that it is colloquially referred to as a guild dying to “the roster boss.” There have been many posts written about how to find a guild or recruit. But how many raiders should the guild shoot for?

The Problem: Why Numbers Matter

In Warcraft currently, a group must bring exactly 20 raiders to the highest raiding difficulty, Mythic. For lower difficulty levels, a group can bring 10-30 raiders, and the encounter will adjust to match the size of the group. For those difficulties, a raiding roster of anywhere between 15-35 (or even lower/higher) can reliably raid every night it plans to without benching many (or any) raiders.

For Mythic, a roster of fewer than 20 will likely cause the guild to fail, as few prefer to raid in a group that is gimping itself by not bringing 20 guild raiders to the encounters. A roster of too much more than 20 will also not work well for the guild, as players bored of being benched too much will leave.

So, the key question is how many more than 20 is “too much more?” What’s the right number for most raid teams?

Advantages of Small and Large Rosters

Smaller rosters give a raiding group more consistency and allow the group as a whole to learn quicker. Fewer bench slots means more play time for everyone, giving everyone more experience on each boss. Small groups require fewer player swaps, so group composition is similar from night to night, requiring fewer strategy changes due to class/player composition changes. Smaller rosters also allow raid gear to be split among fewer players, minimizing the gear that is “wasted” on progression because it is equipped by a player benched for a progression fight.

Larger rosters give a raiding group more durability and flexibility. A guild with 24 members can weather the storm of 3 people absent on the same night, while a guild of 22 might have to take a night off. Additionally, large rosters tolerate players leaving/quitting the game better – a roster of 24 can carry on for a week or two if 2 players leave, while looking to recruit 2 replacements. A roster of 22 is suddenly in a very shaky position if 1 person leaves for another guild and 1 person quits the game. Also, larger roster sizes give a bit more flexibility in raid composition. If an encounter is easier with more of a specific class, or more ranged, or more melee, a larger guild is more likely to be able to shift its composition to take advantage of the encounter’s weakness.

Ultimately, I think most mythic guilds want as small a roster as possible to guarantee that they survive and have enough raiders on almost all raid nights. The advantages of a smaller roster are significant, and there are little disadvantages outside of “maybe having to skip raid due to absences.” Large rosters are an administrative headache, and can cause players to look for other guilds if they are spending too much raid time on the bench. Outside of world top 50 raiding, a guild essentially wants more than 20 players only to allow it to raid if people are absent and ensure long-term sustainability

A Mathematics Based Approach

So then, exactly how many players do we need? That number depends on your tolerance for risk (let’s quantify that as “how sure do we want to be that we have 20 to raid each night?”) and how high attendance you demand of your raiders.

I take raiding Mythic fairly seriously, as I imagine most people raiding Mythic do. If you wanted to just spend time with friends, and not be bothered if a certain number of players fail to show up, heroic is much easier and much less administrative headache than mythic. I therefore suggest that you want to have more than 20 players show up at least 95% of the time and ideally higher unless the guild has planned not to raid due to a holiday. If you can’t raid due to not enough people showing up on a non-holiday, you want it to be an extreme fluke, the kind of thing that happens once a  year.

As an officer of several guilds, I also suggest that an average attendance of 95% is realistic for mythic raiding guilds. Some players will never ever miss raid, and they will be balanced by players who miss about a night a month, which is close to 90% attendance for 2 or 3 night raid teams. I think an average 95% is reasonable, as I would be hesitant to kick an otherwise excellent raider if they missed one week every two months. Much more than that, and I would talk to them about their attendance and see if it was going to be improved going forwards.

I created this spreadsheet for you to estimate how often your guild will have 20+ people to raid, given two variables: the size of the raid roster and the average attendance of the raiders. Feel free to play around with it. Some takeaways:

Raid Roster Size Average Attendance 20+ Raiders how Often?
23 95% 97.4%
24 95% 99.4%
25 90% 96.6%

With 95% attendance, 24 raiders will allow the guild to raid over 99% of the time. 24 seems like an ideal number then, right?

Reality Trumps Math

There are three kinds of lies: lies, damned lies and statistics.

Based on the above math, 24 seems like the right number to shoot for. In practice, I think that will be a little low. There are a few factors that suggest one or two more players is prudent.

First, there is attrition. Guilds that I have been a part of, even excellent ones that mostly lost players to quitting the game and almost never to other guilds, lost 1-2 players a month. Also, recruitment is easier between tiers than during a progression push, especially if a guild temporarily “falls behind” its progression target. Given this, the guild needs space on the roster for recruits and should have enough players that after the guild loses a member the raid roster is still in a good spot.

Second, I don’t think having 20 people online to raid is the right number to shoot for. Having 22 online for raid is about ideal. This allows the raid to shift its composition slightly as required (more or fewer healers, for instance). Having an extra member or two online also allows for someone to swap in if someone is having a bad night, whether due to real life factors affecting their play or internet connection issues. To get 22 consistently with a 95% attendance rate, you need one or two more than 24 raiders.

Thus, I think a raid roster size of 25-26, including all trials, is the sweet spot. This will minimize players’ time on the bench while maximizing the guild’s stability.


Now, any guild size 20 and above can work. I have been a part of very successful mythic raiding teams with 21-22 on the roster (including my current guild, <Trap Run>. By the way, thanks to the GM of Trap Run for giving me his blessing to publish this article and for discussing it with me). I cleared Mythic Archimonde in <Cake Plantation> in Tier 18 at around US 167, and helped manage a roster of around 30.

22 raiders feels a bit low and “dangerous;” one raider disconnecting can force the guild to stop raiding if a few others are absent. 30 raiders allowed for a weekend guild with a rather low average attendance rate and was a logistical feat to accommodate fairly; most guilds would do well to have a smaller roster size and find players who can hit higher attendance numbers. The guilds I have been in with around 25-26 raiding members felt like they were both nimble and stable, minimizing players’ time on the bench while maximizing the guild’s durability.

How much attendance can you demand from your raiders? How much do you value short-term consistency versus long-term stability? What is your magic number?

To Consume or Not, That Is the Question

I have been enjoying my Repurposed Fel Focuser, a reusable item that costs 7500 Nethershards and provides a buff equal to a bit less than half a flask. I was running several easy Mythic+ with guild-members the other day, and one asked why I wasn’t using better consumables. I could have been using the full flask (and using Potions of Prolonged Power on cooldown).

I have over one million gold, so I could afford the 500 per hour for flask or 1000 gold per hour for potions of prolonged power – but why spend 1500 per hour when we are consistently 3-chesting the Mythic+ dungeons without me doing so? The idea is that it makes the dungeons go faster, and as a DPS, your job is to make the dungeons go faster.

For any difficult or progression content, I always use the best consumables I can get my hands on. For easy farm, the “why” eluded me a bit. The more I thought about it, the more I thought my guild-member might be right, though. At current wowtoken prices, the consumables would have cost an additional 25 cents per hour, and are at least partially offset by disenchanting the gear from the runs. Even if using better consumables shaves off a minute of time per hour, it’s probably worth doing.

Saving such little gold is a false economy if it makes the dungeons even slightly slower, especially with the prices being what they are relative to my gold and the price of a wowtoken. If the potions were 20,000g per hour that might be a closer call. But as is, why not use them, even on easy content?

My Fel Focuser is probably best used for world quests only, or even not at all in favor of more potent flasks. When prices are low, consume away!

Feasts Are Back in Patch 7.2

Per the 7.2 patch notes, the best Legion feast (Lavish Suramar Feast) grants 500 main stat (up from 200) in 7.2. This means that for most classes, feasts will be better than all other food options. Even the lesser feast (Hearty Feast) will grant 400 main stat.

If you cook your own food and don’t have rank 3 of the feasts, you might want to head back to Nomi with some materials.

Make sure your raid group is ready to feast on some feasts!

What Boss Order Should We Follow in Nighthold After Mythic Trilliax?

Now that my guild has cleared seven bosses in Mythic Nighthold, I have some perspective on what boss order you might want to pursue. What boss should you target now if you are just getting to 3/10 Mythic bosses down (with patch 7.2 right around the corner)?

The options for what boss to do fourth are Spellblade Aluriel, High Botanist Tel’arn, Krosus, and Tichondrius. From easiest to hardest, I suggest Spellblade < Krosus = Botanist < Tichondrius. Why?

Spellblade was heavily nerfed just this past week. The boss is now definitely the easiest path forward. Follow this video guide by Kelani or something similar. Our raid killed Spellblade 7th, and with the nerf, Spellblade is now significantly easier than the other three options.

Krosus has a DPS check that remains tight. However, the mechanics are fairly straightforward (check out the RaidAdvice guide), and if you lack the DPS check on Krosus, you are going to have a hard time on Tichondrius. There’s an argument for doing Botanist if your guild is missing the Krosus DPS check, but overall I think Krosus is a bit easier. It also rewards Tier gear.

Botanist is mechanically demanding and can be a frustrating fight. The hard part is getting down phase three, which comes after 6-8 fairly easy minutes of phase one and phase two. DPS is not a big issues on Botanist, and you may want to call for raiders to stop or slow DPS at certain points to ensure a smooth transition between the phases.

Miniguide on Botanist: We did the  Naturalist > Arcanist > Solarist kill order. In phase three, we soaked the first set of Call of Night debuffs, then had the 2nd and 3rd set suicide by running into Toxic Spores as soon as they got the debuff. Every ranged should be prepared to run alongside someone who gets the 2nd or 3rd debuff set to minimize ticking damage on the raid while they suicide, and healers should save big cooldowns in case ticking damage gets through. It’s key to avoid deaths to silly mistakes in phase one and phase two so you can resurrect key players who suicide in phase three.

Tichondrius is now the hardest of the bunch, in my opinion. It’s close, and you could argue for doing Tichondrius after Spellblade. But while Krosus has a hard DPS check and Botanist requires mechanical precision, Tichondrius demands a bit of both from your raid.

Blizzard did a great job with this balance of these bosses! Until the Spellblade nerfs, any of these four bosses was a viable fourth option. Giving meaningful choice to the raid in terms of boss order is cool. It allows guilds to pick a boss that might be more suited to their strengths (for example, Botanist for mechanically-oriented guilds or Krosus for guilds with top-notch DPS).

Our raid killed Krosus > Tichondrius > Botanist > Spellblade. Since we knew we had the DPS check for Krosus, having killed it fourth like many other guilds did, we decided to do Tichondrius over Botanist since the mechanics on Botanist are a bit trickier. Ultimately, I think that Tichondrius ended up being harder to kill. It was a close decision on whether to do Botanist or Spellblade next (before the Spellblade nerf).

Usually you can follow wowprogress kill totals and just do the boss most people have killed and do fine. The kill counts right now line up with my recommended order, except that Spellblade still has fewer than Krosus, probably because the Spellblade nerf was so recent.

Check out this reddit thread that discusses this topic.

The “correct” order might change based on subsequent patches – this is the order I recommend right at this moment. I intentionally do not address the question of “Are there issues in my guild if we are just now getting to 3/10M, and what do we do about them?” I am only addressing boss order in this post.

Good luck with your progression!

Mythic Plus Keystones Next Week in 7.2

Patch 7.2 is coming this week (worry not, the new raid is still months out). If you routinely do a 15 keystone each week, you may want to do an 18 this week. Why? Starting with 7.2, each level will get more difficult, and a 10+ generates the maximum reward (of a 905 in your weekly cache).

More significantly to this post, for this week only, if you do an 18 you will get a 10 in your bags next week.

See this blue post for more information:

Key you get next week based on highest key completed this week:

Mythic 2-8: level 2
Keystone Mythic 9-10: level 3
Mythic 11: level 4
Mythic 12: level 5
Mythic 13-14: level 6
Mythic 15: level 7 (each difficulty level above 15 increases next week’s Keystone level by 1)

Or, if you want to run lots of mythic plus next week for AP, consider doing only a level 10 this week, as that will generate the (maximum) 905 gear reward next week and you will start with a 3-level key you can run many times:

To put it more directly: completing a Keystone dungeon of difficulty 10 or higher this week will result in a reward with a minimum item level of 905 in your next weekly chest, which is the current maximum reward in Patch 7.2.

A tidbit for next week: Do the intro quest in the Broken Isles first thing, it leads to getting Artifact Knowledge 26 instantly, which increases AP gains by something like 400% over level 25.

Only after doing that should you open your weekly cache (which rewards Artifact Power now) (EDIT 2017-3-27: the weekly cache will not reward AP the first week of 7.2), do world quests, dungeons, etc.

Do you know of any more things like this to do this weekend to be prepared for 7.2?