The Mechanics of DMing

[Everynone]

So I'm a somewhat new player of D&D 3.x, but I would like to (if possible) DM a simple campaign at my house with my younger brother. This is, quite obviously, a challenge.

I know the ideas behind DMing almost as well as everyone else, and I understand all their responsibilities. The only thing I dont understand well is the mechanics of DMing. For example, generating an encounter (challenging and fun, but not dangerous), and generating the loot & XP that comes with the slain monsters.

Theres another side to this: coming up with a simple yet engaging campaign setting (I feel like sourcebook settings might be too complex for the kind of informal campaign I hope we'll be doing). Im sure this can be an issue for every DM, so Id like some tips.

[limejuicepowder]

Being a good DM based only on the advice of others is kind of like watching a tutorial on youtube then suddenly becoming an expert: it ein't gonna happen. The only way to be a good DM is to DM, most likely for many hours. I'm not saying this to rain on your parade, or that there is no advice to be given, I'm just putting the disclaimer up front.

With that said, here's some things I've found useful in my DM experiences -

1) Characters are everything. More than descriptions, more than plot points, the NPCs is what makes the game come to life. If the players feel like they are speaking to real people, immersion in the game is increased exponentially. Accents, speaking patterns, personality traits; anything you can think of to differentiate one NPC from the next is good.

2) Don't plan too far in advance. Railroading is one of the easiest ways to DM, and one of the easiest ways to make the game boring. Avoid this by only coming up with major events and doing the rest on the fly. I won't understate the difficulty; this is not easy to do, especially in the beginning. However, you can give yourself a big leg up by following advice item 1: if YOU feel like the NPC is a real person, with real motivations, it becomes much easier to imagine what that person would do in a given situation. Thus you can respond to the player's actions on the fly.

3) Giving your group appropriate challenges depends entirely on system mastery. The more experienced you become with the system, the more accurately you will be able to predict what a group can handle. More than the others, this will be a trial by fire. Use the monster manual CR as a guideline and see what happens. As the game progresses, you will see what is a good fight and what is too easy/hard. With that said, you can't always tell: luck is built in to the system, and sometimes things just go south (for the PC or the NPC). Roll with it and keep going.

[CaladanMoonblad]

I've found Donjon (http://donjon.bin.sh/) to be an exceptional tool (populated with SRD items). If you play around, you can find the Random Generator for things like communities, magic tomes, and even a random fantasy inn in addition to random encounters, random loot, etc.

The website is based from servers on an old British protectorate island in the south Atlantic ocean, so sometimes it takes a bit to load.


As for story construction, I used to use the old 3 Act play composition for my stories (Intro, Rising Action, Conclusion) before I went into nonlinear stories based solely on geography and countdown timers for three or four major stories progressing at the same time (but this requires a lot of work that players may not actually encounter due to their decisions).

[Hendel]

Keep it simple and fun. I am not sure how old your "little brother" is but if you want to keep him focused, keep it simple.

I am assuming you are starting at 1st level. Will you have a character or is he going solo or are there other friends that will join?

Either way, start out with a short dungeon that you draw on graph paper or use the Dungeon board game idea or a simple pre-drawn dungeon floor.

Put some rats, a giant spider, and some goblins in it. Give the rats some random treasure like 2d6 gold pieces (there are treasure guidelines here (http://paizo.com/pathfinderRPG/prd/gamemastering.html)), but just keep it small and simple at first. Maybe the spider has a gem worth 100gp in its web and a scrollcase with a scroll of Magic Missile or something. Just keep making stuff like that up and write it down on paper or the graph paper.

Voila, you are on the way to making your first dungeon and it should be enough fun to keep everyone's interest and to allow him to practice playing and you to practice DMing.

[Urpriest]

Make sure you are as confident as possible about the rules for monsters. They're a foundational part of the game, but many players don't end up learning how they work. I've got a handbook in my sig about it.

More generally...it depends on what you're going for. You may want to run something more rules-light first, if the goal is to get your brother interested in RPGs. You can move on to running him through a D&D game after you're feeling more confident about the game yourself.

[Everynone]

Being a good DM based only on the advice of others is kind of like watching a tutorial on youtube then suddenly becoming an expert: it ein't gonna happen. The only way to be a good DM is to DM, most likely for many hours. I'm not saying this to rain on your parade, or that there is no advice to be given, I'm just putting the disclaimer up front.

Noted. I know practice and experience are most important, but I figured getting advice ahead of time would be better than not.


1) Characters are everything.

I noticed this alittle bit in my first campaign, but my more recent one has been very PC heavy, so I dont get to see it in action as much. Thanx for reminding me; I'll try to build interesting and RPing NPCs thatll hold his attention and respond appropriately.


2) Don't plan too far in advance.

This, I did not know. Like, at all. Does this mean I should attempt to do an open world quest system, or something along those lines?


3) Giving your group appropriate challenges depends entirely on system mastery.

This I suspected, but didnt quite wana believe it. Practice, practice, and experience galore...I thought about giving myself a hypothetical dungeon situation, the trying to generate an encounter and loot, just to familiarize myself with the rolling and calculating. Is this a good idea?


I've found Donjon (http://donjon.bin.sh/) to be an exceptional tool (populated with SRD items). If you play around, you can find the Random Generator for things like communities, magic tomes, and even a random fantasy inn in addition to random encounters, random loot, etc.

At first glance, this appears to be an invaluble resource. I'll havta look over this more in-depth for sure.


As for story construction, I used to use the old 3 Act play composition for my stories (Intro, Rising Action, Conclusion) before I went into nonlinear stories based solely on geography and countdown timers for three or four major stories progressing at the same time (but this requires a lot of work that players may not actually encounter due to their decisions).

True...is there any way to circumvent that? Id like to do a non-linear storyline if possible, but Im not sure how Id do that without preping for multiple questlines.

[CaladanMoonblad]

True...is there any way to circumvent that? Id like to do a non-linear storyline if possible, but Im not sure how Id do that without preping for multiple questlines.

So... let me explain how I'm doing the current "zone" for my players.

My players have essentially entered a Frostburn campaign setting in their quest to "rescue/investigate" a previous PC who was retired (the player no longer is playing with us, but their character didn't die, just went back to his barbarian people).

I created a 300 mile x 300 mile map zone from our campaign world map (it was originally a random generated fractal from Campaign Cartographer). I populated it with communities (mostly taken directly from Donjon's random community generator (http://donjon.bin.sh/fantasy/random/#town) and changed as appropriate (including the Random Inn (http://donjon.bin.sh/fantasy/inn/) complete with Rumors and sample Menus). Based on the communities, I made decisions about the local economy; one village mines iron, another village has craftsfolk, another village farms exclusively, etc., and they all trade together...

So this leads to bandit quests... and clearing out monsters from local resources (mines, fields, etc.)... even politics (some community leaders are tyrants, some are elected, etc.)... and war (some communities want to use the PCs in their long standing rivalry with their neighbors)... etc. Anyway, I also made sure to put clues about the missing NPC in each community, because the heroes are learning what happened to their friend (a Chaotic-Neutral Barbarian) and these clues include leaving behind bastards with local women, trying to start a rebellion, and generally causing trouble that the PCs are now tasked with cleaning up after their friend came through the area.

I also populated the area with dungeons (don't use donjon for this; while the dungeon maps are great for caverns, the themed dungeons really are random and ignore thoughts about dungeon ecology, etc.)

So... I start with a main storyline (investigate their once friend), and populated it with side storylines based on geography (clearing out bandits, rescuing the common law wives of the errant PC, being used as pawns for local tyrants, etc.). I also have storylines that have their own countdown timer (ie, by the third week, news of the heroes' adventuring in the area catches the eye of so and so, who wants them to do A or B or C).

Sure, my players can just plow through the area and try to skip to finding their once ally, but each piece of information about what really happened to their friend becomes more dire, more evil, and more dangerous for the eventual encounter. Each local encounter gives them another piece of the puzzle and is preparing them to the full horror of what their friend is facing/has become.

One way to prod my heroes, is a bulletin board of quests (they belong to an organization called The Adventurer's Guild, which provides jobs in a variety of locales). My heroes have to do a short day by day journal on the campaign loot page document to help me keep track of the 3 or 4 main quest countdown timers.

[Kol Korran]

some of the others touched on other subjects, i'll try to address your specific questions:


For example, generating an encounter (challenging and fun, but not dangerous)
An encounter doesn't stand on it's own usually. sure, you can string a few challenges for different rooms of a dungeon (in computer game style), but there are a few things to consider:
1) The PCs are supposed to have their resources taxed (hp, spells, per-day uses of abilities, magical items) by encounters. the DMG assumes a party should face 4 CR equivalent encounters before resting, each consuming about 20% of their resources. there are quite a few problesm with this assumption though: first, the CR system is not very good, and at many times either underestimate or over estimate encounters. secondly the experience of both DM and players can alter this perception greatly. new players for example might do ingenious ideas/ newbie mistakes, which might make an encounter far easier or more difficult than it's CR. and lastly, who wants to fight 4 similar difficulty challenges? you want some things to be quite easy, and some more challenging, more difficult. these are a bit harder to gauge.

but the general concept that you should consider a set of encounters as using the party's resources is a solid. you can go with the "explore dungeon till running on empty, go rest, repeat" style of video games, but most good adventures don't have the dungeon/ adventure site just sit and wait for the PCs to rest- delays have consequences (they should if the players want to feel their characters live in a "real" world).

what to do about this? I can't give you an accurate mechanic. I'd suggest to DO use the basic CR concept above, for a session or two, but design the encounters with 1 or 2 of them a bit more difficult, and some less. I suggest to design with a good stopping point avialable after about 3 encounters, or to just design the adventure lag with 3 encounters, with one extra that might be thrown in. as you growinto DMng, and your players into gaming, you will find your balance.

All that said, I suggest avoiding to design too many specific encounters, instead i suggest to:
2) design situations: i'll use an example. instead of having an old abandoned fort filled with 30 orcs, their beasts and such, each detailed in their own little rooms and groups, I'd suggest you desing how the fort works daily- it's routine, it's patrols, the power structure, defenses and such. prepare 2-3 exemplary "encounter forces" that you might throw at parties here and there, according to the situation, develop some strengths and weaknesses for the fort, and DESIGN NO OUTRIGHT SOLUTION/S on how to succeed in the situation. instead let the party come up with ideas, strategies and so on, and respond to them, basing your reactions on your familiarity with the fort, the way it works, and the personalities of the major characters inside.

This is a bit more difficult, but feels far less static, more "real", and leads to much better adventures. I highly suggest making the "slaughter them all!" option quite difficult and dangerous, but enabling some more down scaled battles here and there (like fighting a patrol, or a hunting party, and so on). let the party face greater odds, but be smart about it. it makes for far better adventures.

try to put many leverages points for the party to might try and use. makes things possible, some easier, some more difficult, but most times possible, try not to block creative ideas, the game is about the PCs being heroes after all. (if something is stupid though, try to give a fair warning). for example- the fort might have an easy breach over looking a cliff, that the orcs put just a small guard upon, so scaling the cliff might be an easy way in. or there might be a fey in the forest that might help the party in exchange for... something, or the orc lieutenant believes in "honorable combat" kind of stuff, and might agree to face one of the PCs in a one on one duel, if they wish, or there might be a secret tunnel to the fort's chapel, guarded by... something. and so on.

3) scaling encounters up and down: this is a bit difficult to do, especially with a new DM and a new player. first thing to remember is that it's ok if some encounters are a cake walk, while others turn into life and death. sometimes it's not even the encounters you planned! that's ok, it happens- PC decisions and the luck of the dice can change many things... it's not ok if nearly all of your encounters are a cake walk or life and death. (for most groups at least, some like one of the two extremes). a few tricks I found might help when learning to gauge the power of the group:
- more/ less people: the party stumbled upon a patrol of the orcs... too easy? hey! suddenly another patrol is nearing, having heard the sound of battle. too tough? the patrol leader calls to 2-3 of the orcs: "call for reinforcements! there might be more hiding!" and some take off.
- "what did you just say to me?": sometime one character just has all the bad luck. this is a good time for the monster/s to divert their attention to another character, ESPECIALLY if they are trying to get it's attention. i'm a bit hesitant about this advice- don't do it unless it's quite believable, or players might feel you're babying their characters.
- "capture them alive!": instead of killing the party (or killing all of it in a very bad situation) intelligent monsters might decide to capture the party for later interrogation and so on. this enables you to change the adventure a bit (usually not much of a problem if you're designing situations, instead of just encounters). have some (or all) of the party's equipment taken. most players for some reason take a much greater offense to this than having their characters killed. so they will feel loss, but the game can move on none the less.
- "killing you is not my goal": this works usually well in important battle, and mostly in defensive scenarios: the PCs are an obstacle to the enemy, but their goal is different than killing them. they wish to take some item, kidnap someone, kill someone else pass through the magic portal and so on. once the PCs fail to stop the villain due to bad lack, tactics or whatever, the villain ignores them (or set their minions to occupy them) while they accomplish their goal. I suggest very strongly to avoid situations where the PCs MUST succeed (except for an end of campaign situation), and allow failue to have other consequences than stopping the campaign. This often makes for a more memorable campaign- again, failure means the game is more "real", and can often build motivation against a major villain and so on.


and generating the loot & XP that comes with the slain monsters.
there are many different approaches here. My general approach is this: decide how much XP the party is assumed to have at the end of this adventure lag (if we use the example above- rescuing the prisoner from the orc fort), using the tables in the DMG work out how much treasure they will be assumed to have gathered. write down that number. then, i'll assign about 25-40% of that amount as a reward for mission accomplished, the rest I will use to put treasure in the adventure, about 40% of it in the hands of NPCs/ characters that can use it. the rest as treasure of ruined temples, treasure chest and all that jazz.

I make a list of the treasure locations, and value of each. when the PCs get some treasure I mark it. If at the end of the adventure they missed some treasure due to being clever or resourceful (negotiating with the orcs for the prisoners, or sneaking in and rescuing them, or creating a diversion for such a rescue) i add the missing treasure to the reward ("you did a fine job! here's some extra stuff!") if they just overlooked something stupidly then I usually not add it to this adventure, but perhaps to the next.

Sometime I add special challenges in an adventure, a sort of "bonus challenge", that the party may or may not tackle. If they do they usually find an extra treasure. (usually minor, but with flavor and useful qualities) this does not get added to the reward if they miss it.


Theres another side to this: coming up with a simple yet engaging campaign setting (I feel like sourcebook settings might be too complex for the kind of informal campaign I hope we'll be doing). Im sure this can be an issue for every DM, so Id like some tips.
new groups don't have to have tons of new stuff, intrigues, twists and so on. you can start with the basics. I suggest to start small- a simple region, with 1-4 settlements, only one of them big ({small city should do it), and some places for adventure hooks. make a short history of the region, and perhaps a few notes about things beyond it. as you play, and ideas pour in, you can expand on the region, and possibly on what is beyond it. in a few more levels the party might travel abroad. and you expand as you go along.

as to published settings- you can easily take what you want from them, and forget the rest. take what you need, refurnish it, shape it. I quite often like maps. If you can find the "Red Hand of Doom" map somewhere (there are a few guides and logs of this adventure on the site) you might have a nice map of a small region, with about 3-4 settlements, and quite a few areas to occupy the party. and just make it what you want!

I hope this helps, good luck with DMing, each finds their own way of doing it, finding your own way is half the fun! :smallsmile:

[Beldar]

General Principles:

#1 - You're playing to have fun. So don't bog down on things like rules questions. My default for 30 years of dm'ing has been this: In any confusing rules situation, make up something that sounds reasonable to all & go with that, then after the session, look up the actual rules for use in later sessions (if there even are any rules on that particular question - sometimes there are not). This works, keeps all reasonably happy, & keeps the game going.

#2 - Keep it simple. Just like you don't want to bog down in rules debates, you also don't want to bog down in too much plot complexity or other things. Extra-complex plots & so forth can waith until you and your players are experts (by which time that level of complexity will seem fairly simple).

#3 Don't be afraid to 'adjust' things as needed to keep it moving and fun. The simplest example of this is rolling your dm dice behind a screen so you have the option of ignoring the critical hit the monster just got on the party's best fighter, which would drop him and result in the party all dying. This is generally frowned upon later on (the players like to know that their achievements really are valid, rather than created as needed), but when you are just starting out & haven't yet got a feel for game balance, it is necessary sometimes. Don't do it to extremes, but a little bit of judicious fugding will really help you get started.
That could also include things like providing access to an important NPC that fills a need, or dropping some treasure that is tailored to the party's needs, rather than randomly rolled.

#4 A wise game designer once said "a game is a series of interesting decisions". This is why railroading is unpopular (since they ultimately have no real decisions to make - they just go down the tracks to their pre-determined destiny). So set things up to where they have a number of options, large or small. That can be things like
"you can go back to town now to make it there for Market Day, or you can try to stick it out in place one more day until the last goblin locked in the forted-up room runs out of food and comes out". Make each option real, with real advantages to it. Life is like that - you never have enough time, or money to do everything that might be nice, so you have to choose what sounds best to you.
Other kinds of interesting choices can come from giving them varied loot, thereby giving them weapons options ("My sword has X advantage but my bow has Y advantage, which would be best in this situation..."), and choices about whether or not to take the time to drink a potion or not. They don't get to make that decision if they don't have a potion.
I don't mean load them down with treasure - just that a few odd scrolls or whatever can give them more options and it is generally interesting to have options.
I have had some fun times dropping some really weird treasure on players to see what use they would make of it.

Story:

It's been said that there is "nothing new under the sun". This is true: all stories have been told, in one variation or another. We just randomize the names & details to keep it more-or-less fresh.
Realizing this, the easiest way to start DM'ing is to take a simple story and change some details.
For example, when starting a new, simple game for my young nieces and nephews, I began with the story of Hansel and Gretel.
I thought about the basic setup: an evil witch living deep in the woods. 2 kids in the nearby town go missing. The witch has them and plans to eat them.
Add in the party: They show up in a small town & are obviously adventurers. The mayor, in grief and desperation, comes to them & offers to hire them to rescue his missing kids (might as well make Hansel and Gretel his kids - also I changed their names).
Throw in some random encounters (the woods are dangerous & contain monsters, and she lives far enough in that it will take some time, and tracking rolls, to get there).
Come up with an excuse to give the party some time (the witch is feeding the kids to fatten them up).
And we have a basic adventure.
You can then throw in complications to taste (maybe a bad snowstorm hits and delays them & spoils the tracks, so they have to detour to the old oracle lady to find out where to go. Or maybe a rival witch steals the kids from the 1st witch).

The way I keep it more open to whatever the players decide, & not a railroad, is to decide what is happening in the background story. The background story moves on, incorporating whatever the players do. For example, in the above simple adventure, the players can choose to ignore the Mayor. In that case he becomes beligerent & has them arrested on a trumped-up charge. The village, which would have been friendly to them, is now unfriendly. They will still get the option to rescue the kids (to clear themselves of the charge, and probably with some kind of enforcement, such as a Mark of Justice), but they don't have to take it. We could end up with the village opposing the players & branding them as outlaws. Then the adventure turns into something logical from that standpoint, such as "where do we get food & do our shopping, since the village is no longer an option" or "survive a long journey to a new location, so you can start fresh" or whetever. They could go a robin-hood route, or even join forces with the wicth if they are inclined that way.
Or, if the players stumble in a way that gives the bad guy (the witch above) warning that they are coming, think through what would the bad guy do with that knowledge? They may flee to a new location, or may set traps, hire mercenaries, spy on the party with magic or other means, etc.
Making reactive bad guys keeps the players thinking - in a situation where they have interesting decisions to make.

There are all sorts of stories out there to take as your starting spot: movies, video games, books & on & on.
Maybe the village was a mining town until kobolds moved in & took the mine. They need you to clear out the kobolds quick so they can mine some ore needed to help arm the kingdom for an impending war (Idea stolen from the starting human area in World of Warcraft)
Just change enough details so it isn't obvious what idea you've started from, or it will seem cheesy/campy/other derogatory term.

If you know what is happening in the background, then you can have specific details prepared one or two sessions in advance. You don't want to prepare too many specifics in advance, since you don't know what the partry will do that may change things.

Sometimes you will have to make up things completely on the fly, since the players did something totally unexpected for which you are not prepared. In that case it helps to have 'cheat sheets'. For example, they take an unexpected turn and go a a new city you have not prepared at all. The most helpful thing you can do is to quick grab an example that it is like. So, you could for example, decide that it is like Stormwind in World of Warcraft. Change plenty of details so they don't catch you at it, but now you suddenly have answers to all sorts of questions like "do they have a bakery" and "can I get in and see the king". You also now have its general description - lots of white stone construction with plenty of large statues. This is tremendously helpful.
If you need just a bit more time to cope with something the party just did, throwing in a battle can be very helpful. While they are fighting the 'random' encounter, you have a while to think up what happens next. A non-battle event (such as "the bridge is out & the river is swollen, how do you get across") can also give you time to think.

With all this, I have found it indispensable to keep notes.
For me, that is basically 4 docs:

Scratchpad
Next Time doc
DM Ideas doc
Status 3-26-2013 (or whatever date you played on)

The DM Ideas doc has the background story, plus plenty of story hooks for later use. It is a general dump doc for what could happen sometime. It has details on villains, villages, who is doing what, etc.

The Next Time doc has my specifics for the next session or two. I have a summary list of events/challenges at the start, then details on them further down.
Generally what is in it at any moment is enough to last 3 sessions, in case things don't take as long as I thought, or if they skip some things.
I also note what date & time in the adventure it is (not necessary, but helpful if they have countdown timers going).

The scratchpad doc I have open while we play, on my laptop. I update it with quick notes & rulings as we go, so I know where we've been & what we've been doing.
This includes stuff from current hp of monsters they are fighting, down to what the weather is (if that came up), or that the party failed to notice something that will come back later in some form. These are usually terse and incomplete since I am writing them as we go. But they serve as a reminder.

The status doc I write up later, one for each game session, as a reminder to myself of what happned. It is basically the scratchpad notes, made more coherent, plus my memories of what transpired. It probably isn't purely necessary, but it has come in tremendously helpful to me many times. I can answer all sorts of questions, like "what town were we in where that guy sold that thing" or whatever.
If you feel the need, you can also record exactly what treasure and xp were awarded. That has been helpful at times too (like when they get around to identifying magic items).

There are all sorts of additional tricks for advanced DM'ing. But you asked for a quick starting guide, so that's what I tried to focus on.

One last thought - many assume (and it is more or less built into the game) that you shouldn't fight things you can't take on successfully. This rule can be broken, but if you do then you have to telegraph it clearly to the party that this encounter is different.
For example, once when I introduced the Super-monster they needed to run from (an interesting new challenge - effective fleeing), I first introduced a competing party, racing with the players towards the goal & so sometimes fighting alongside them. The party could see that these guys were just as tough as they were. Then when the super-monster came on the scene, the competing party got there first & got wiped out spectacularly, with the players in range to see it all. That whole sequence was mainly to clearly indicate to the players that they could not fight this monster & must flee, build up strength while hiding, and then take it on later, preferably in non-standard ways.
But that kind of thing is for later, once you are more comfortable DM'ing.

Also, I agree with the previous poster about desiging how the fort works, patrols etc, with no pre-cut solutions. Let the players figure out how best to take it on.
In a way, you are the pitcher - throw the ball but leave it up to them how they hit it, whether they bunt or go for the walk, whether they try for one more base than they safely can get etc.

I also agree about scaling encounters on the fly as needed. Know the rules for non-lethal damage, so if you need to, you can just knock out the party & have them wake up in a jail cell/chains etc. That just alters the adventure rather than ending it.

[Beldar]

You also mentioned you needed help generating encounters and the XP, and loot resulting from them.

Here is a handy short list I keep in my DM ideas doc:
Page 38 DMG is xp awards
Page 53 DMG is treasure values per encounter level. (p.55 is gems & art)
Page 216 DMG Random Magic Item generation (more tables on p 222-223 for weapons, 227 specific weapons, 230 for potions, 231 for Rings, 234 for rods, scrolls on 238-243, staves on 243, wands on 246, minor Wondrous Items on 247, medium Wondrous Items on 249, Major Wondrous Items on 251.)

Page 31 DMG for example Skill DC's

As for encounters:
In the back of your Monster Manual, it lists monsters by CR, or Challenge Rating.
Theoretically, a party of 4 players, each at level 3, should be matched against a CR 3 monster.

You can have multiple monsters of a lower CR (every doubling of monsters adds 2 to the CR).

if your party is fewer than 4 (or more than 4) then add or take away a few monsters to make it harder or easier as appropriate.

And, as always, adjust to taste.

[Bucky]

If you're worried about botching the XP curve, you might arrange with your players to give levels for quests instead of levelling with XP.



As for encounters:
In the back of your Monster Manual, it lists monsters by CR, or Challenge Rating.
Theoretically, a party of 4 players, each at level 3, should be matched against a CR 3 monster.

You can have multiple monsters of a lower CR (every doubling of monsters adds 2 to the CR).

if your party is fewer than 4 (or more than 4) then add or take away a few monsters to make it harder or easier as appropriate.

And, as always, adjust to taste.

First off, a party of 4 level 5s facing a CR 5 monster is an 'easy' encounter; they are expected to win, but the play comes from trying to do so while burning as few resources as possible.

A CR 6 or 7 encounter is 'medium' or 'hard'; they're fine to hit the party with on a routine basis, but don't spring a CR6 encounter on them when they are low on daily resources. Also, CR7 is about the point where the encounter is a serious threat to individual PCs; don't put them at a tactical disadvantage on top of the higher CR AND make sure they have near-full resources.

A CR 9 encounter is a 'fair' fight - that is, the monsters should be able to beat the party about half the time. This is boss fight difficulty, and should be arranged so that the players either have a large tactical advantage or don't die if they lose.

Having 3 PCs instead of 4 lowers their CR by 1. Having 6 raises it by 1. 5 PCs are at an odd half-CR point, which you can compensate for by adding an extra CR2 minion if you need to.

Finally, the encounters are all CR6:
A CR 6 Monster
2 CR 4 Monsters
3 CR 3 Monsters
4 CR 2 Monsters
2 CR 2 Monsters and a CR 4 Monster
1 CR 5 Monster and a CR 3 Monster
1 CR 5 Monster and 3 CR 1 Monsters
6 CR 1 Monsters

You can use these stock encounters as templates and level them up with the party, substituting in stuff of the correct CR.

[Hyde]

Let's see, without needlessly rehashing anything that anyone has said, except for one minor point:

I recommend party xp. Basically, everyone progresses at the same rate, and there's no individual variance in xp. This means raised characters come back at the level they are (or are behind as normal, but then are leveled current with the party the next time the party levels up), and you ignore any xp expenditure for stuff.

I also recommend using Pathfinder instead of 3.5- It has it's own problems, but has a lower learning curve than 3.5, at least at entry level (cross-class skills, half ranks? no thanks) I might also recommend 4e for starters, if only because it treats everything with kid gloves and will get you used to the more abstract parts of DMing. That said, doing that runs the risk for system confusion when you switch to a real system (frankly, we still use some of the rules we found in 4th, like diagonal movement and 1 hp "minions")

Probably the most important thing is to remember that it's not "the DM vs. the Players". You can't "lose" the game, even if some fluke happens and your Big Bad gets killed unceremoniously- don't let it bother you, and don't worry about trying to "get" the PCs. If everyone's not having fun, something is wrong.

A tip on worldbuilding- Asking "why?" will help fill in a lot of story holes you might have. "Okay, I want my weird wizard hermit on top of superdoom Mountain. Why does he live there instead of Pleasant Hill?" The answer could be something like "Superdoom Mountain is the only source of powdered facemelt, which he likes to put in his sandwiches." But by asking why or how something would have to be as you designed it, you can avoid a lot of "a wizard did it" moments.

Also, don't be afraid of running a module or something to kick things off. They're usually not terribly in depth, and can be flavored to your setting. Definitely a good way to get a feel for things might go.

Just don't use "The world's largest dungeon" (dull) or "Tomb of Horrors" (eep).

[Za'hynie Laya]

TEN WAYS FOR A GAME MASTER TO RUIN A CAMPAIGN

1. Never listen to your players. You're in charge - what do they know?
2. Change the rules constantly. Who needs a consistent experience? Certainly not you! Why should your players?
3. Give the players tons to read. It's more than a pastime - it's class time!
4. Give the players none to read. What's to know? Point `em at the bad guys and roll some dice.
5. Spotlight, Schmotlight. They're lucky they even get to have a character in this story. You'll focus on your pet, thank you very much!
6. Put the NPC in charge. "Get my luggage, will you?"
7. Let nothing ever change. Your world is preciously perfect; you dare not let the players do anything to it!
8. Cancel. Lots. Without warning when possible. So what if they planned their calendars around this; you've got a raid tonight on WoW.
9. Let no argument go. You're authority has been challenged; be certain to waste at least an hour on this to be sure they understand who's running this show.
10. Promise one thing, deliver another. "Remember all those cool combat abilities I let you have? Welcome to Dream Fairy Land, where none of them work..."

The above is a quote from another forum but I cannot remember who to credit. I played under a DM like this, twice. Shame on me. Years ago, Dragon Magazine ran some wonderful articles called Dungeoncraft. I hope this list will help.

The Rules of Dungeoncraft:
1. Never force yourself to create more than you must.
2. Whenever you fill in a major piece of the campaign world, always devise at least one secret related to that piece.
3. Whenever you must decide the chance of something happening and you have no idea what that chance should be, it's always 50%.
4. Good adventures always challenge the players and challenge their characters.
5. What's done is done.
6. Simple, easily identifiable characteristics are the best tools for portraying NPCs.
7. Running a good campaign is about building a world, not building a story.
8. An Important encounter should give EVERY player something to do.