A common discussion in the Dungeons & Dragons community concerns the traditional experience point system versus milestone/session-based leveling. Fans of milestones like a greater focus on narrative, but its detractors think they’re just too lazy or stupid to do arithmetic. I like the ease of session-based leveling, but am going to back that up with meticulous math!
As mentioned in one of my first blog posts, How I Design D&D Sessions, I usually plan for each session to have the following scenes:
1 hard-deadly combat encounter
2 medium-hard combat encounters
2-4 noncombat scenes, traps, or easy-medium combat encounters
1 optional easy wandering monster encounter
This provides the backbone of what an average session looks like for me, and gives me something to base my calculations on. By figuring out how many experience points each of these encounter types gives a player on average, we can get a rough estimate of how much they should get per session.
Anyone who writes their own D&D adventures is probably familiar with page 82 of the Dungeon Master’s Guide, which has the XP Thresholds by Character Level chart. This gives guidelines for what sort of encounter will be easy, medium, hard, or deadly for a character of any given level. Obviously this chart can not be trusted 100%, but I’ve found that the guidelines are actually pretty accurate most of the time.
There are some great tools for generating encounters of different difficulty levels, such as Kobold Fight Club. I can’t do anything the easy way though, so I have a spreadsheet with 879 encounters of either 1-10 monsters of one type, or a group composed of one “leader” and 1-9 lower challenge monsters. Here’s a snapshot of what it looks like:
I then take this data to make additional sheets for each character level 1-20 with easy/medium/hard encounters that can be easily used for 3, 4, or 5+ players. Here’s level 4 for reference.
Tangent: Why do I design my encounters this way? For one reason, I sometimes don’t know how many players I will have on a given night, so it’s nice to put in my notes “5-7 zombies” and then use whatever is needed for whoever shows up that night.
It also helps prevent me from getting tempted into using too many low-level monsters, or one single very high level monster. I find that putting in more than twice as many monsters as there are player characters can easily turn into a slog, and monsters whose CR is much higher than the party's level turn the game into rocket tag. Either the players overwhelm the boss by turn two, or it hits them for over half their hit points at a time.
Back to experience points, you can see that I put in a formula to calculate the total experience of each encounter at each level, and then average them all together. This gives me a rough average experience point value for each type of encounter (easy, medium, hard) for me to plug in to my session template.
1 hard encounter (220) + 2 medium encounters (176 each) + 3 1/2 easy encounters (108 each) = 950
Can you guess how many experience points level four characters need to advance to level five? 3,800. Do you know what 950 x 4 is? Exactly 3,800. Pretty crazy right? You’d almost think Wizards of the Coast paid people to figure this stuff out ahead of time, although I don't think they use exactly my methodology.
So now that I know that, I can say with confidence that if I don’t stray too far from my session template, a group of level four characters will get to the next level in four sessions and I can plan accordingly without having to track how many goblins they actually killed or decide how much to award for every trap and NPC interaction. Here’s how this breaks down for each level, in case you’re curious:
Level 1 - 1 session
Level 2 - 2 sessions
Level 3 - 3 sessions
Levels 4 to 11 - 4 sessions
Levels 12 to 19 - 3 sessions
The Dungeon Master’s Guide recommends on page 261 to do one session for levels one and two, then two to three for subsequent levels, but they’re also explicitly assuming four hour sessions, while I shoot for two to three.
Some of you may be screaming, “But what if they skip over encounters!” So what? If they hustle through the dungeon and never end up getting ambushed by that wandering monster, that’s due to their choice not to take short rests after every combat, and they deserve a reward for that. If they bypass a more significant encounter without fighting, it probably involved an even more significant effort. Or maybe they actually ending up getting into some fights I didn’t expect. It all come out in the wash.
That’s all my math, but I haven’t even begun to address narrative structure. It’s a lot simpler, and doesn’t require any spreadsheets.
I’m big into “fractal” narrative structure. I think that every session should end with some sort of resolution, every session where the party levels up with a more significant resolution, and then every four or so levels with the resolution of a major story arc. I can do this because I have a very good idea ahead of time as to when these plot points will occur.
This is how I plan my campaigns with a lot of narrative focus, while still appeasing the D&D-is-still-a-game-and-should-have-hard-rules-based-in-numbers side of me. I may have no idea what the adventurers are going to do on their path to defeating Lord Garfield, but I do know that when they do, it will coincide with obtaining level five. And this isn’t just because I said so, it’s backed up with math. And spreadsheets. So many spreadsheets.