However, I'm running into the problem of character/NPC/monster levels. In this type of campaign, the same NPC may be present for a long time, meaning the PCs will level up several times. How do you keep an NPC adversary relevant and not overpowering while keeping the suspension of disbelief intact during those conditions?
For example, let's say the head of the Assassin's Guild is one such NPC. Sometimes he's an ally of the PCs, sometimes their enemy. At first he should be a dangerous opponent in a fight, but not invincible, and he should stay a relevant threat at least until the PCs are well into paragon levels. Likewise, he should be a useful ally in and out of combat for those levels. Letting him level up* at approximately the same speed as the PCs is one solution, but could make the players frustrated - "We can never beat this guy!" (This type of frustration can be good, but if it seems the DM is cheating to keep his pet NPC alive it can quickly turn sour and break suspension of disbelief.)
Likewise, the city guards should be a relevant threat for most of the campaign. Here leveling them up at (nearly) the same pace seems weird: a single NPC becoming more powerful is less of a problem than the whole city guard going through the same process.
There are some options:
- Leveling up enemies as mentioned above, though that mostly only works for single NPCs rather than whole groups. Letting everything keep pace will put a big strain on suspension of disbelief and may make the players feel frustrated that their increase in power doesn't mean anything.
- Just ignoring the issue - let the PCs grow and the NPCs stay static. The problem with this is that it means I will have to define in advance which NPCs the players should be credible threats and suitable opponents at which level in advance. This doesn't fit my DM style very well, and definitely not for this type of open campaign where the players should be allowed to choose their own goals and paths as much as possible.
- Reducing the speed of level gain, or, more extremely, removing leveling completely. This will somewhat mitigate the problem or in the extreme case eliminate it, but can again cause player frustration: leveling up is fun and an integral part of the game.
- Switch systems. This isn't really an option in this case: I love DMing 4e and our group likes playing it.
Have you run this type of campaign? How did you handle this issue? Or are you just smart enough to see the obvious solution I missed?
Any advice appreciated!
* Which doesn't mean he's necessarily constructed using the rules for NPCs - he could be made as a monster that's increased in level, occasionally changing powers to stay appropriate.
If the NPC is supposed to be the same level as the players but you want to show advancement, have him advance at the same rate as the PCs or slightly faster (Kinda like the best friend-turned-evil thing).
As for the guards, have different factions of the guard or keep the guards at various levels rather than one static level. Guards have ranks too, so vary the levels according to the rank they hold. Nothing wrong with mixing it up a bit.
Just a few suggestions...
Having been on the receiving end of an 'unkillable' NPC as a player I understand the potential for player frustration in the face of an impossible situation--after all, the players are supposed to be the action heroes, right?--but as DM sometimes the players are supposed to be outclassed. It is not nice, but it shows a realism that will improve the game even if the players do not appreciate it at face value. Another solution is to keep the players and NPC out of combat together. He is more skilled than they are (he is the head of the Assassin's Guild, after all!), so he chooses the environment and situation to show himself, putting them at such a disadvantage the players would be suicidal to attack him or he can escape immediately at the first sign of trouble. I guarantee that even if he is a pushover when the players finally catch him unawares they will feel a sense of accomplishment if they have to work hard confront him on their terms.
Regarding the guards, in the real world different levels of training is the norm for police forces. At the bottom you have civilian community watch, then there are the beat cops you see and the investigative detectives you do not. If the players begin throwing their weight around they could face the medieval equivalent of SWAT, and if they cause a real problem and SWAT cannot handle the situation maybe the military will march in to save the city.
Any number of things can mitigate the "enemies become relatively weaker" problem.
1) This can be OK as eventually the enemy should be defeated and character's level progression provides a logic for why they can win in a rematch.
2) Why should the super villain BE defeatable at all early on? This goes with 1. Eventually the PCs grow in power, the NPC doesn't, they beat him.
3) Enemies CAN grow in power at a fairly slow rate without serious difficulty. If the arch enemy is 6 levels higher than the PCs early on, he may be equal level by the time they hit a point where they can beat him.
4) Beaten enemies aren't necessarily defeated. They can come back or continue to operate in the background. They just may not be a combat threat anymore.
5) An enemy can be encountered in all sorts of situations. Just because a party defeated him earlier doesn't mean they can do it again. He may get stronger allies, a better lair, etc.
6) There can always be story justification for a radical repower of an enemy. This can be handled by say adding a new template to the enemy. Maybe he becomes undead (Uh, they PCs did kill him before, well, they should have burned the body...).
7) Just introduce a hierarchy of bigger bad guys. This is the standard approach. Maybe the PCs know there is an uber badguy, but the ones that actually appear on stage are just sub-bosses.
8) Put the PCs into the clutches of the bad guy. They can defeat him now that they gained levels in a FAIR fight, but how about if they're locked in a cage with no equipment being lowered into a pool of laser armed sharks?
As for "background" NPCs. I would keep them static levels, except maybe you can introduce a new type once in a while when needed. So ordinary City Watch are level 2 say, but if there's trouble level 3 sergeant shows up with reinforcements, and if there's BIG trouble a level 5 captain shows up with a 4th level wizard in tow.
As for other NPCs, if you don't come into conflict with them then who cares what level they are? They just operate however is needed for the story to work. If some ability they have becomes a plot point then it can be explained that they cast a ritual, hired a wizard, or unbeknownst to all they are actually a pretty good sword fighter.
One possible solution to the city guard problem could be to increase their level but reduce their type. At first they could be normal monsters, with elites for the, well, elites, but later the standard guards will be reduced to minions or swarms (like the Zombie Throng in Open Grave). That way they can stay a credible threat (sort of): for a lvl 10 party a mob of lvl 10 minions makes more sense game-wise than a mob of lvl 2 normal monsters.
Actually, perhaps the same thing could work for major potential enemy NPCs: one might start out as a lvl 6 solo (very dangerous even if the whole party fights him alone) but gradually be reduced in relative level and type, so that when the PCs have gained five levels he may only have gained two and been reduced to an elite, making this sample NPC a lvl 8 elite for a lvl 6 party (still a worthy opponent if he has appropriate support but easy to take down for the whole party together), and then down to a lvl 10 normal monster by the time the PC hit paragon.
Hm. I'll have to think more about this.
Of course, I could just ignore keeping track of NPC levels and only stat them up when combat becomes a real risk.
I guess part of my problem stems from the conflict between "create a world and let the players loose in it" and "make things up as they become relevant".
Anyway, thanks again for the input!
Nothing really exists until the PCs see it
That sounds like a good rule of thumb.
I have done similar things before, just not in a level-based system. In these situations I tend to base my description of the area around the NPCs which usually entails giving them at least approximate stats, if for no other reason than to remember more clearly the role I had envisioned for them. I'll just have to fight that instinct I guess.
For example, in the entry for Thyatis (a military empire, not unlike the roman one) I wrote for the guards:
- City Guard, town, grunt: lvl 3-5
- City Guard, town, officer: lvl 3-10
- City Guard, imperial city, grunt: level 6-10
- City Guard, imperial city, officer: level 11-15
That way I know what services are available, and what are credible threats to a village or city, and what the guard can actually handle. The players will also notice, when they are below, on par, or stronger than NPCs, which tells them, when they are on their own and when there is a chance of getting help.
Note, that players don't know NPC levels, they can only guess by what they experience. It is OK, if certain "name" NPCs level up as the players do, while most should be static.
You can use a recurrent villain whenever you like, you don't have to tell the players his level. When running my 1st Edition campaigns in Greyhawk I never told the players the level of the NPCs (they could guess that a cleric providing Raise Dead must be 9th or higher, but not much more), and nobody ever noticed that the villain they met at level 10 was about the same threat at level 18 - he was always Lord Ravenwood to them, and never "a Fighter 12" or an "antipaladin 17").
There is also one piece of advice I wish to force on you
Never (and I mean never) count on players choosing a (sensible) goal of their own, and have always an adventure prepared with a bright red pointer towards it. I found that players like a bit of railroading, if done with tact, and that most are not happy with picking a goal for themselves.
Never (and I mean never) count on players choosing a (sensible) goal of their own, and have always an adventure prepared with a bright red pointer towards it. I found that players like a bit of railroading, if done with tact, and that most are not happy with picking a goal for themselves.
That is something I am very much aware of. However, in my experience picking your own goal can work if it's done at the time of character creation. If the group of players get to choose at least a direction they want to go in and can design their characters with that in mind it's not necessarily a bad thing.
I have no problem with providing a plot if they're not going anywhere by themselves, though.
But again, in my experience if you provide the players with what seems to you to be a clear choice between taking the road to the left or the road to the right they will at best barge out into the forest and more commonly stop at the crossing and dig a hole.- Till the need for mechanics come up give him none. Simply free-hand RP him, he can do whatever and act appropriately because it fits his character. The PCs and Players can't see his mechanics so they don't know he doesn't have stats.
- Build different versions of him for different situations that require mechanics. A person isn't limited to a specific group of skills, they can grow, change, etc. So in one case he shows his expertise in say parkour in a Skill Challenge, the next in diplomacy, the next in combat. Don't try to fit them all into one build but instead make different builds for whatever scene he would be in.