Tuesday, November 10, 2009

Running a Raid in 4e, Part 1

Gamefiend threw out the question of whether or not you could run a WoW style raid in 4e, and I've been finding myself chewing on it. I'm pretty confident the answer is "Yes", though I admit I think you'd be kind of crazy to try it, since anything that involves juggling that many players is crazy enough to really qualify as a stunt. It's something you could do at a convention, but it's probably better suited to online play (heck, maybe it's a good use of google wave). This is the a window into my thinking on this, and while it's super, super raw, I figured I'd share it as I work through the details on how I'd make it happen.

Now, you could do it as a straight-up fight, just by making a staged villain[1], probably a triple solo to make sure he has enough hit points to last, but that could quickly end in madness as everyone takes their turn to move and arrange and track their hit points and so on. Also, you'd need to add more area attacks or the things just going to get torn apart by sheer numbers.

So instead, I would probably set it up like an open ended skill challenge with an abstract map. The trick is to capture the two elements of a good raid: distinct roles, and a necessity to change up tactics.

The simplest way to do the "map" is to set up areas of engagement: Near, middle and far. Near means adjacent to the boss (where the main tank and melee damage dealers are going to be), far means at a distance (where most of the healers, buffers and ranged damage-dealers hang out) and middle is the space between, most occupied by skirmishers, off tanks, and specialized roles.

The boss, whatever he is, has a level. That level pretty much determines exactly what sort of threat he is, and also is the basis of most of his capabilities. His hit points are abstracted into the number of "Damage successes" it will take to take him down. I'm calling the ballpark on that his level (which I'll just call [L]) times 5, but this is totally unplaytested, so that might be off base, especially since it would probably be smart to scale him with the number of players. I'd endorse representing these with tokens, like poker chips, but it could just be that I'm nuts for tokens.

In a straight up fight, things go like this: On your turn you can move one "space" (or two spaces if you use any kind of movement power) or use a power (whether to attack or not). You might also do something else, but that's much more situational. Moving is hopefully self explanatory, but the actual fighting is something else.

When a player attempts to attack the boss, he can use any of his attacks, with the following limitation. Melee/close attacks require the character be near the boss, and ranged attacks (anything that would invite an attack of opportunity) need to be from the middle or far distance. If a character uses an ability to heal or buff an ally, that works normally.

An attack is always assumed to hit successfully, but most of the time the only thing that matters is the damage dealt - Raid Bosses are immune to all manner of special effects and statuses (except when they are not, see below), but benefits that help allies can still be triggered. If the damage dealt meets or exceeds a certain threshold (probably based off [L]), then it counts as a success, and the boss takes one point of "damage" - accruing damage successes is what ultimately takes the boss down.[2] However, any failure gives the Boss one "Menace Point".

This matters a lot because, after everyone's taken a turn, the boss gets to go. He has a number of menace points equal to his level, plus any he's gained from player failures. He uses them to build his attacks - he should have a list of abilities that have menace point costs, but in the absence of that it works something like this.

For 1 menace point he can make an attack against a single target in the "close" area. It is an attack against AC, where the boss effectively rolls 15+[L] and does damage equal to the low normal damage on the table of the gods (aka page 42 of the DMG)[3]. For each additional menace point spent he can enhance the attack by doing one of the following:
  • Affect an additional target
  • Affect everyone in the area (costs 3 menace points)
  • Target the middle distance instead
  • Target the long distance instead (Costs 2 menace points)
  • Increase damage one step to the right (low normal becomes Medium normal, high normal becomes low limited and so on, costs 2 points)
  • Increase the effective "attack roll" by +2
  • Change the attack to a different defense (Reflex, fortitude or Will)[4]
The boss can make as many attacks as he has menace to pay for, and certain bosses will have special abilities that they can spend menace to trigger that do more than just damage.

This proceeds, round robin, until either all the PCs are dead or the monster has taken enough damage to go down. Simple as that.

Now, this is very basic, and very mechanistic (which is, arguably, very apt for a raid) with very little in the way of tactics. It would be intensely boring because most player will simply do the same thing every round. However, this lays down the baseline for the next step, adding in important things like aggro, roles, abilities and events. And that comes next.

1 - A staged villain is one who, when reduced to zero hit points, changes rather than dies. In most cases this is a physical transformation (like clay statues that have snakes burst out of their chests when they're beaten) but it's also a useful way to simulate the changes in tactics that are familiar to video game players. Normally, each "stage" has normal hit points for its level, though you can just as easily make one stage elite or solo (but I normally wouldn't - the point of this trick is to offset the long dull endgame of solo fights). Easy to budget for it too, as each stage is just treated as its own critter, which is a little bit kind to the players, but I think it comes out in the wash.

2 - As an optional rule, you might allow additional thresholds to speed things along, and to make strikers feel more valuable. So if a creature's threshold is 11 and a hit deals 23 points of damage, you might decide it removes 2 points of damage and so on. Alternately, that might just be situational, but that's something for tomorrow's post.

3 - I am tempted to have damage be measured in healing surges rather than hit points, but I need to think about the impact this has on healing, and whether it means tanks (sorry, defenders) end up insufficiently tough. If damage is in hit points, it's the one thing that can't just be handled with different colored poker chips, but the onus of bookkeeping is on players, so it is distributed. That said, tokens or cards definitely have certain logistical advantages, especially for the large combats. Similar thinking could be applied to damage (just say that at-wills, Encounters and Dailies do 1, 2 and 3 respectively) and while that simplifies things, it's actually a bad idea because it makes things REALLY boring for the player, and removes the tactical choice-making I hope to introduce next.

4 - This is, by the way, insanely abusable, which is why the DM shoud not actually use this on the fly unless he's willing to show some restraint. It's a decent yardstick for pricing boss powers though, and I'll use it later when I craft up some demo raid bosses.


  1. I get the need to do abstraction here, but y'know, part of me wonders if the abstraction shouldn't be done along different vectors.

    Consider what happens if you pull a Hard Boiled Armies gig on this and instead treat each group of characters that fill a particular role as one character for purposes of taking action and moving around the map, etc.

    Or define the problem a bit more tightly. Sure, it'd be insanity to do a 40-man or even 20-man raid straight up using the rules as written, but I'm betting 4e could easily sustain a table of 10. I'd want to add in a chess clock, giving players a tight amount of time to complete their moves; part of what makes a WoW raid really sing is the high pressure timing of everything.

  2. I totally agree the HBA approach is much better if we want to talk about running a raid with more characters than players - I think you could do that very elegantly with. But part of this exercise is based on the assumption of 1 character: 1 player (which is part of why I really consider it a stunt).

    That said, you're right about the timeliness issue - there needs to be a sense of pressure, and every player needs to quickly make a decision (and at the same time, there needs to be more to that decision than 'just keep doing the thing I'm doing'). I was thinking about that in terms of giving a very short window to tell the DM what you're doing when it comes to you (like, 5-10 seconds) but the clock might be even better for the group as a whole. Hmm.

    -Rob D.

  3. My disagreement is more abstract. If MMOs evolved from D&D, and 4E evolved from MMOs, then returning to MMOs again for the concept of the single-boss raid seems to me like an evolutionary cul-de-sac. Raid is one of the furthest things from regular D&D play, and it sacrifices a lot of believability for gameplay. I worry about the same here.

    To be fair, you've nailed the essential elements worth converting, though.

    Also, I think Agon handles most of these issues pretty well, IIRC.

  4. I think D&D 4E simply isn't suited to online play like this. It's like trying to make a real-life board game out of Tetris. I guess you can, but why?

    D&D 4E assumes certain facets of play -- a bunch of people at a table gathered around a single battle mat, lots of interactions among players, in-depth knowledge of ally powers -- that just won't translate well to this sort of scenario.

    How can 40 people all see and interact with the same battle mat? How many of my allies will know that I can push an enemy 1 square with one of my at-wills?

  5. nice! To add some of my own thoughts, as I've been chewing as well:

    a) I definitely agree with you on the zones. That should be a definite part of the setup.

    b) Menace Levels --love this! You'd have to tinker with the maths to tune it, but the basic concept is sound. This makes the whole thing work. You can also do the same thing with "Adds", minions or lieutenants the minion can generate. For the accumulation of points though, I think I'd have the monster on a steady gain. Each turn he gains 1d6 menace. This generates the "clock" that you and fred bring up, placing pressure on players to get rid of that SOB as soon as humanly possible. I would even do that out in the open (with a huge six-sided die!) to put even more pressure on the players. This is in addition to physical factors.

    c) fred, I was thinking 10 man was the limit. Any higher than that and it's something else entirely.

    d) Environment. I think that one thing a boss should be able to do is trigger environment shakeups that conjure mini-skill challenges in and of themselves. The Boss smashes the floor, tearing it apart in such a way that there is a skill challenge to navigate and get to the safe parts of the zone. Stuff like this.

    e)semioticity, I can definitely see where you hesitation lies. For me, it's more of a bounds-push. What can we do with this ruleset? How can we have different play experiences? that's what I'm interested in. I don't think I'd be playing that on any regular basis, but I would love to try it out a few times and make the rules serve it if possible. It would never be a mainstream way to play 4e I don't think.

  6. @brent This is, I concede, a little crazy. 40 people really means google wave, or a fast larp in the same vein as a game of Werewolf.

    That said, this shaves off a lot of the specifics, quite intentionally, by making it damage only. The additional special effects simply don't work directly, so you don't need to worry about tracking that one square push and so on. But that doesn't mean it gets rid of them entirely - I have something specific in mind to keep the powers differentiated.

    @gamefiend yes, definitely, but let me get through tomorrow first - we're already on the same page for where I'm going with this.

  7. The solution that occurs to me is to keep the players fragmented into smaller, more manageable groups, each with its own DM. The Delve that WotC has occasionally run at Gen Con is a good model.

    You'll need to come up with a plotline that justifies keeping the groups a bit apart from each other, but for a special event that seems feasible. One idea: a sort of massive Lovecraftian creature is breaking through, slowly spreading in multiple directions through the dungeon. The PCs need to head it off in multiple directions at once, so they have to split up. Ideally there would be a reason for one party's actions to impact another party to emphasize the group aspect. It could be as a simple as the dungeon being interconnected enough that a party doing well might send one or two people off to help a group doing poorly. You could also build in effects that require teams to work in conjunction.

    Ultimately, as Rob says up front, this is really a stunt. But stunts can be fun.

  8. You could definitely do this "straight up" in a con setting. I think you could have multiple tables with one big ass battlemat in between. Each table gets one DM and a group of players. As PCs get bumped off, you can coalesce to fewer tables. I could see DMs shouting out: "I'm using an action point." And the other DMs check it off. "We just got tagged with radiant vulnerability 10." "Bloodied!" Etc. Some brief record-keeping at the end of each round, e.g., probably deal with boss's total hp loss then.

    It would require playtesting to get the balance right, i.e., attrition of # of PCs to boss's hps. I think it would be a blast. Hmmm, I call dibs for Gen Con next year.

  9. Don't know how I missed your post, Alan, but we were thinking along the same lines obviously.

  10. The first thought I had while reading this is that it'd be very interesting if the boss had the ability to cut off certain movement, like destroying part of the environment so characters could not easily move from close to mid range, leaving some people stranded in the melee and some stranded too far away. Easily this could be done as a mini-skill challenge during the whole thing or just cause some different decisions / tactics from the players. I really like the base idea though, can't wait to see how it plays!


Note: Only a member of this blog may post a comment.