Thanks for buying this game! You'll do this many years from now! (by Larry Elmore)
How long does it take to play Dungeons & Dragons?
Several m0nths ago, James Malizewski observed that the Mentzer Companion Set effectively codified the much-needed “endgame” for Dungeons & Dragons. I respect James, but it’s not much of an endgame if it never arrives.
I’m going to say that Level 12 effectively qualifies as hitting the endgame. Under the Moldvay/Marsh/Cook version of the Basic Game, most classes need about 600,000 points to reach Level 12.
How long does it take to get there? For a game that’s been in play, in one form or another, for about 35 years, there seems to be very little hard data, though Maldoor made an early attempt.
I’ll take a guess based on a semi-official pronouncement. The Holmes Basic Rulebook states that it should take about six to eight adventures to level up. I don’t have my copy with me; I’m going from memory. Holmes, unfortunately, doesn’t explain whether this means six to eight sessions of play, or six to eight completed dungeon-adventures. (In any event the most helpful measurement would be points acquired per hour of play.) But let’s assume Holmes is talking about sessions of play. In that case it would take the player 72 to 96 sessions to reach Level 12–somewhere around three to four years if you’re playing twice a month.
Frankly I think Holmes’s math wasn’t intended to apply beyond the first few levels, which after all were his main focus in the Basic Rulebook. As an example, a Magic-User needs 40,000 points to go from Level 6 to Level 7. To do so in eight sessions would require earning an average of 5,000 points per session (maybe 20-30 thousand for the party as whole). That’s not impossible, but it’s a very steep pace to maintain given the adversaries you’d have to fight. I don’t have any hard data to suggest another rate of advancement, but presumably it gets a lot slower around high-level play.
But even at lower levels, Holmes’s estimate of 6-8 adventures to level seems way off for our group of players. Eric’s Principalities of Glantri game has been playing for about 20 sessions, and nearly all of the participants are still Level 1 (this is in large part due to turnover, both of characters and players). Tavis’s White Sandbox game has had about 15 sessions with a more stable (but larger) cast, and maybe 50% of the regulars have gained one level. Also, we’re playing with what amounts to double-XP-for-treasure rules, and Tavis is using the 100 XP per hit die of the enemy, so our advancement is considerably quicker than it would be if we were playing by the B/X rules. And we’re also playing using published modules (B2 and Caverns of Thracia).
To me this suggests that Holmes’s estimate is too generous by a factor of 2 or 3 (or it could be that the “adventures” he’s using as his unit of advancement are maybe 2-3 sessions in length). So that would mean hitting Level 12 would take anywhere from 144 to 288 sessions. Playing twice a month, that’s anywhere from 6 to 12 years.
So in order to reach the endgame of a D&D campaign, we’re talking about a time commitment of at least 3 to maybe up to 12 years. For a casual social activity, competing for attention with one’s professional and familial obligations, as well as whatever other interests one might have, it approaches absurdity.
Now, I’m assuming (1) that we’re playing from the low-levels to my arbitrarily imposed cap of Level 12, and (2) we’re advancing at a rate more-or-less as the rules intended. Either assumption might be wrong.
But to the extent that you’re measuring your game against some idealized mode of play where folks go from Level 1 dopes to Level 12 super heroes, that is a long haul. Maintaining your own interest, to say nothing of your players’, is going to be a serious challenge.
Charlatan
Eric Minton
maldoor
Greengoat
David Wellington
James Nostack
tavisallison
The Mule Abides 
Wow 15 or 20 sessions and still having 50% or more of the PCs at 1st level is abysmally slow for low level play. Holmes mentions a rate of 6 to 12 adventures to gain a level, with this holding up into higher levels with 10-20% of adventures proving to be relatively profitless.
I played almost weekly campaign and we got 4 to 6 sessions a level going on hiatus with the party ranging from 10th to 14th level. It seemed pretty quick for me, possibly a little to quick for mostly weekly play but I was mixing adventure sources like a mad man and not being too picky with the way one version of the game provided exp compared to another.
My longest running campaign was topping out in the mid teens and it ran every two weeks for 7-8 years using the AD&D rules. We did spend about half that time at or near name level.
I participated in an on and off game that lasted until we were playing in the 20’s with armies marching back and forth against each other across the nations of the D&D known world . I don’t recall leveling up during the last year or so of that campaign as it had stopped being about that by that point in play. The longest running players certainly logged over 200 sessions in that campaign which had been in the end game for so long it was called “The Wars” by the regulars.
Hey JD, thanks VERY much for those data points (and for the correction on Holmes’s figures – I’ll make the necessary corrections later tonight).
For point of reference, Tavis’s White Sandbox game started out at Level 3, so that may account for our leveling time leaning toward the highest end of Holmes’s figures. It’s a very big group, usually 8-10 players, and the average lifespan of a character is fairly brief – I’d bet our highest-level guys have played in about 8-10 sessions. One player has leveled up TWICE, but that’s because he basically got very lucky in the sessions he attended, picking up the pieces from a couple of 90% completed delves. I dream that my character can steal as much credit and hog as much glory as “John Fighter.”
One thing to keep in mind is that if you’re playing in ~100 sessions, and your character has a 5% chance of dying in any given session, then your odds against surviving for 100 sessions are 168:1. I guess that’s not too bad with resurrection magic. But if, say, it takes you 30 sessions to crawl from Level 1 to (say) Level 5 – that is, a level high enough that someone might care enough to cast Raise Dead on you – then your odds against survival are 4:1 against.
This is an area I’m keenly interested in, but unfortunately don’t have much data to contribute–yet.
I’ve got a couple of PCs who survived 5 sessions in the Caves of Chaos and earned just under 4,000xp–not enough to level since they’re both Elves. In my Moldvay-ish Lost Mine, there’s a Dwarf who has survived 6 sessions and earned about 300xp. Although I suspect that if they return for another session, they’re going to double that amount, because they’ve thoroughly mapped nearly the whole dungeon, and have unluckily managed to avoid the two large caches of treasure that remain (out of three).
Tonight I’m running the first completely BtB Moldvay dungeon I’ve made. There’s probably about 12,000gp on the first level alone, and about 1500xp worth of monsters. This one should be a much better test of BtB advancement.
“How long does it take to play Dungeons & Dragons? … I respect James, but it’s not much of an endgame if it never arrives.”
The implication here is that you’re not really playing the game if you don’t reach the endgame.
“The Holmes Basic Rulebook states that it should take about six to eight adventures to level up. … But let’s assume Holmes is talking about sessions of play. In that case it would take the player 72 to 96 sessions to reach Level 12–somewhere around three to four years if you’re playing twice a month.”
I doubt long-term play was intended to occur so sparingly. You and I are working on an urban timescale that doesn’t match what I know of early play, which apparently involved multiple sessions each week.
For a related anecdote, several years ago I discussed the “One World by Night” cross-campaign Vampire live action group with one of its members. Apparently the group was experiencing problems because different chapters of the organization were playing at drastically different rates. Whereas many East Coast groups met once or twice per month and gave out XP accordingly, some of the small-town Midwestern chapters met two to three times per week. Long-running PCs in the latter groups had something like ten times as much XP as those in the former, providing an overwhelming statistical advantage when they crossed over to other games in the franchise.
Working from your estimate, a group that plays three times a week will hit level 12 in six to eight months.
“Eric’s Principalities of Glantri game has been playing for about 20 sessions, and nearly all of the participants are still Level 1 (this is in large part due to turnover, both of characters and players).”
And there’s a whole lot of turnover! I don’t think the Glantri game can be used as an effective benchmark on that basis.
In addition, I expect that early players accomplished a lot more per session than we do in the Glantri game. Our sessions are quite short, and we also have a more frivolous attitude toward play than I expect from a tradition that started among wargamers, especially if the examples of play in the early rulebooks are to be taken at face value. Where we putter around cracking jokes and making unrealistic plans, old school players appeared to push hard to get through as many encounters as possible in the time available to them.
“Now, I’m assuming (1) that we’re playing from the low-levels to my arbitrarily imposed cap of Level 12, and (2) we’re advancing at a rate more-or-less as the rules intended. Either assumption might be wrong.”
I believe you’re wrong on both points. The latter is an objective matter; the former is subjective, as noted below.
“But to the extent that you’re measuring your game against some idealized mode of play where folks go from Level 1 dopes to Level 12 super heroes, that is a long haul. Maintaining your own interest, to say nothing of your players’, is going to be a serious challenge.”
To the question of how to maintain your interest, I’ll turn that question back on you.
What interests you about Dungeons & Dragons?
Eric, my post could be summarized in the following for questions:
(1) Do you think that advancing from level 1 to the (allegedly) rules-defined endgame in Dungeons & Dragons takes a long time?
(2) If so, how many sessions, over how many hours of play, and over what span of real-time?
(3) What are the the social opportunity costs of doing something for that many hours?
and, in light of my previous blog post:
(4) How does the play of Old-Timey Dungeons & Dragons satisfy or exceed those opportunity costs?
My own answers are (1) yes, (2) apparently ~100 sessions, ~400 hours of play, and (for our group) many many years; (3) ~400 hours is 10 standard work weeks, or about 14 whole weekends (assuming I’m conscious for about 30 hours per weekend); (4) for me it can’t, and thus I personally feel playing from Level 1 to Level 12 would be immensely dissatisfying.
Obviously YMMV.
Oh, and what interests me about D&D is a board-gamey element, and a caper element. Neither could sustain my interest for 400 hours, and that’s not a black mark against D&D specifically: very few things can do so.
On a slightly side note, somewhat related to “ideal” session length. I recall that my D&D time from my formative years (junior high – highschool) was absolutely chock full of dithered play time. I recall a scheduled all-nighter with five of my friends and we literally spent more time thinking of a name for our party than actually adventuring. I never started having serious play time (like I read about in the DMG) until college when time became more limited.
The slower advancement that we are experiencing in the current two red/white campaigns is somewhat of a shock to realize. Like some type of catholic guilt, I used to think I was doing something wrong or not playing seriously enough in those golden years.
I hear you GG. Back in the day my players excoriated me for being such a stingy, hard-ass DM. Turns out I was playing in the spirit of the game (Moldvay, anyhow).
A very good point emerges from James’ post and the back-and-forth with Eric. If these early editions of the game are built with 3 games a week schedules, longer sessions and more goal-oriented players, then we ought to seriously consider modifying advancement to match up with the way we play now, in order to experience the endgame, the midgame, or hell: how about second level?
I hear clerics actually get a spell there. :)
Some historical data on rates of advancement as of 1975, using the pre-Supplement I 100 XP per HD rules I also use in the White Sandbox, comes from the [Europa fanzine]. Sandy Eisen notes that “after 2 terms – 4 months – of weekly excursions, the highest members of our party are 5th level” (this seems to be a campaign run at Cambridge University in the UK).
Gygax writes that “There is a campaign I know of where I am informed by a player that after eight month of constant adventuring there is an 8th level Magic-User as the highest level in the game, and that is tough but good” – the lack of definition of what “constant” adventuring means makes this less useful for analysis, but it’s invaluable as an endorsement of what “doing it right” would look like.
The data is tantalizingly incomplete, isn’t it? Four months of weekly excursions are 16 sessions… but how many players per session? And as you said, what does “constant adventuring” mean? My guess is that Gary wouldn’t consider anything less than his own standards for adventuring to be “constant”, so that might mean multiple games per week. Or it might mean daily games (a frightening thought).
There are so many moving parts to advancement: the treasure tables, the dungeon stocking tables, how many “specials” the DM puts on the map, etc. And then there’s the wildly different rewards for monsters between OD&D and B/X.
Still, all this analysis of the advancement rules might miss the point: even if we’re doing it BtB, it might still not be “right”. Based on my limited experience with B/X D&D, I strongly suspect that the only way to get to higher levels is for the DM to create dungeons and place treasures according to his own system, and not the one outlined in Moldvay.
I don’t think its the DM’s job to control the rate of advancement. It’s the DM’s job to ensure that there are enough interesting monsters with treasure at appropriate levels.
Rather than enforce any particular rate of advancement, what I have done is populate a large sandbox with a lot of monsters, and then followed a strict treasure placement algorithm: For each 1xp of monsters in the campaign, there is exactly 4xp value of treasure. The treasure is generally a spread of coin, gems, jewelry, and “trade goods” weighing on average 1cn per 1xp. It usually is hidden in vaults or treasure rooms, protected by traps, etc. In any micro-region there may be monsters without treasure or treasure without monsters but the overall ratio is fixed.
With this in place, the rate of advancement lies in the party’s hands. Skilled, exploratory play, with fewer NPCs, can yield more treasure in a shorter period of time and consequently faster advancement. Unskilled play, or more cautious play, with more NPCs taking a share, slows advancement down.
Using this model, we’ve run 1 session per week for 26 sessions, each session lasting 5-6 hours. There are 5 PCs with anywhere from 4 to 16 NPC hirelings. Turnover among PCs has been 80% (i.e. 4 out of 5 characters died and were replaced with new ones) but virtually all turnover was between levels 1-3. Since level 3 there has not been any permanent PC death, as they have Raised their fallen.
As of the end of session 26, we have: 1 7th level Thief; 1 7th level Cleric; 1 6th level Fighter; 1 6th level Magic-User; and 1 5th level Elf. This represents leveling at a rate of 1 level per 4.33 sessions.
So around 143 hours of play for about 40,000 points per player?
“I don’t think its the DM’s job to control the rate of advancement. It’s the DM’s job to ensure that there are enough interesting monsters with treasure at appropriate levels.”
I don’t really see much difference. Can you elaborate?
James, perhaps the difference only becomes apparent if you consider the difference between a directed story campaign and a sandbox campaign. In a directed story campaign, the campaign is “going somewhere”. When you evaluate a directed story campaign, you’re evaluating it for whether it had good pacing, plot, and climax. Leveling is necessary to the directed story, as the protagonists advance in power to confront the antagonist, but leveling must also be constrained, so the protagonists don’t over-advance in power too soon, lest they become bigger than the story allows. So the DM of a directed story campaign literally must control the rate of advancement. A story-driven campaign may have a volume of treasure and monsters that is entirely out of whack with the speed of leveling that “should” happen. Dragonlance is a great example of this. But this is the very definition of how the game ought not be played, at least insofar as one is attempting to play “old school”.
In contrast, in a sandbox campaign, the campaign just is. The answer to “How long does it take to get there?” is “As long as it takes the players to accumulate XXX,XXXXgp.” And how long that will be will entirely depend on what the players do, and how they do it.
Or, to put it another way, consider a 500-wide hex map, with each hex being 6 miles. At the far end of the map is a temple of elemental evil. How long will it take the party to get to the temple of elemental evil?
In a story-driven campaign, the answer the DM will give you is something like “8-10 sessions, assuming that they find the location from the Sword-Wraith after they are betrayed by the Ogre Mage in session 4”.
In a sandbox campaign, the answer will be something like “They can move 24 miles per day, or 4 hexes, in good terrain, but how much of their gameplay they’ll be devoting to travelling towards the temple, I couldn’t tell you. And they might have wandering monsters. Or go the wrong direction. But if they get a magic carpet, it could be faster. I don’t know. Depends on them, I guess.”
For comparison, here are the numbers from the Moldvay generated dungeon I’m currently running:
23 rooms, ~65HD of monsters*, 1,175xp, 11,036gp.
* approximate because some monsters have 1/2 HD, some have 1-1HD, etc.
My average party is 5 PCs and 4 retainers, which comes out to about 1,744xp for each PC.
Using Alexander’s system that would be 650xp from monsters and 2,600gp, assuming 10xp per HD as in Moldvay. That’s a grand total of 3,250xp, which comes out to 464xp for each PC.
Which leads me to think I’m misunderstanding Alexander’s system or he’s not playing Moldvay Basic. :)
It’ll be interesting to see how many sessions it takes my group(s) to clear it. My guess is about 12 three-hour sessions.
Alexander, I think we may have different ideas about what “story-driven” means, but I think I understand what you’re getting at.
We are using Moldvay/Cook. I haven’t altered the xp award from monsters, but I have altered the treasure distribution in that rather than use the random roll Treasure Types tables, I hand-place treasure at an overall fixed ratio. Note that in some cases this means the amount of treasure I award is actually less than the amount that a random roll in Moldvay/Cook would be, while sometimes its more.
All I’m really saying, I guess, is that in a sandbox campaign the rate of advancement is entirely dependent on (a) how much treasure you put in the setting, (b) how challenging you make it to get the treasure, and (c) how skilled the players are in meeting the challenge. The answer to “how long it takes to get to the endgame” is simply a result of your inputs to (a),(b), and (c). The particular ratio I chose for (a) and (b), with players of whatever skill my group has, seems to lead to the results I described, roughly 7 months of regular play to get to leve 5-7, or about 1 level per month or 1 level per 4 sessions.
Over at EN World, Raven Crowking has [a thread about leveling assumptions] in qhich he quotes Gygax from the Strategic Review issue 2.2: “It is reasonable to calculate that if a fair player takes part in 50 to 75 games in the course of a year he should acquire sufficient experience points to make him about 9th to 11th level, assuming that he manages to survive all that play.”
@Tavis, that thread is a good read. Personally, I would be very satisfied if PCs were leveling every 4-8 sessions. I suspect that advancement, like a lot of other parts of the game, was subject to DM fiat. Even in (or especially in?) Gary’s own games.
@Alexander, your sandbox sounds very cool. Formulaic treasure distribution is pretty appealing to me. Even with only 5PCs and 4 henchmen (ie the minimum), they’d need to earn 3,500xp per session to wind up with 500xp per PC (ie 25% of 2nd level). That’d be difficult to pull off with the treasure tables in B/X. If you’ve got a wiki or anything for your game, I’d love to check it out.