Sean ask me to post the following discussing one of our last sticking points on the CCG:
Aside from the question of how to adopt the Essentials magic item changes, there is another issue that we need to finalize before we can release the new LFR Campaign Guide. Specifically, we need to decide exactly how adventure tiering is going to work next year.
Here's the part that IS NOT up for discussion: Starting in Q1 2011, most new LFR adventures will be written to span an entire tier of play (Heroic or Paragon). None of us is super thrilled about this change, but because of the reduced number of adventures per year, we feel like we don't have any choice but to widen the spread of levels. There will continue to be exceptions (such as special events that might have versions for every level band, or intro adventures that only support very low-level characters) but by and large most LFR adventures will span an entire tier in 2011. Epic Campaign adventures are also outside the scope of this discussion.
With that as a baseline, we are in the final stages of deciding what we want to use for level bands and adventure ELs in these adventures. I'm actually deliberately talking about those two topics separately because it turns out that you can combine them in a lot of different ways. Here's how each topic is defined:
Topic 1 - Level bands. A level band is a range of character levels, such as levels 1-4 (H1), or levels 14-17 (P2). In the current LFR model, a normal adventure is written for exactly one level band, and level bands are hard limits (both upper and lower): if you have a level 5 character, you may not play that character in an H1 adventure or an H3 adventure, under any circumstances.
Topic 2 - Encounter Levels (ELs). The ELs supported by an adventure are the targets used to determine the XP budget the author has for building the adventure, and also determines the amount of XP that the PCs will earn for playing the adventure. In the current LFR model, a normal adventure supports two ELs: "low tier" and "high tier" and these ELs are determined by the level band. For example, an H2 adventure has a low tier target of EL 5, and a high tier target of EL 7. ELs determine the budget, but are not hard limits: authors can create individual encounters in the adventure that are above or below the target ELs, but overall, the challenge level of each version of the adventure is going to average out to the target EL for that level band and tier. (Monster level limits are also based on the EL, so you could not use a level 10 monster in the H2-low tier version of an adventure, but you could use it in the H2-high tier version.)
Since the beginning of LFR, I think most people would agree that this system has worked pretty well. There are six level bands (again, only considering Heroic and Paragon) and with low/high tier for each, the DM has a very reasonable amount of support to handle any mix of PC levels at any legal table for that level band. High tier is always +2 EL above low tier, which is significant enough that the players should notice the difference (this wasn't always true, but it should be now, especially with the MM3 damage expressions in play). Authors only need to produce two sets of stat blocks per adventure, which is a fair workload. Mustering is relatively easy because event organizers can divide players up by level band and then by low/high within that level band. It can be hard to find an adventure at any given level band, but the replay rules mitigate this to a certain extent. For all of these reasons, we would rather not change the current system. Unfortunately, as I said at the outset, we don't believe we can maintain the status quo. The campaign is moving to a target of 24-30 adventures per 12-calendar-month release cycle. If we keep the current model it means that we would be looking at an average of one adventure per level band per quarter (and that is assuming perfect distribution) ... and that is just not enough content with enough frequency to keep people playing the campaign.
One option would be to drop support for an entire tier. If we just had 24-30 Heroic or Paragon tier adventures a year, with the current level bands, then it would work. Unfortunately if we drop Paragon tier play then we cut off everyone who has already got a Paragon character and if we drop Heroic tier play then we will not have any way for new players to start the campaign. So that isn't really an option that we can consider without restarting the entire campaign and we are not willing to restart the entire campaign (that has its own advantages and disadvantages, but in aggregate, we believe that the disadvantages outweigh the advantages).
So that brings us to the point of needing to decide how best to administer the campaign where every adventure has to support 10 character levels (either 1-10 or 11-20). For now we are setting aside questions of story and focusing on the mechanics of the situation. I acknowledge that it will be harder on authors to write adventures that make sense if a table of all 1st-level characters go on the adventure and also makes sense if a table of all 10th-level adventures go instead. My recommendation is that authors pick the level band they *would* have written for under the old system and write the story so that it is plausible for that level band, then include notes for the DM (or even optional encounters) for adapting the story to lower or higher level PCs, as appropriate. But I digress.
Here are the questions that we need to answer. I list them under the same two topics as I did above, although they are connected (your answer to the level band question affects your answer to the EL question).
Topic 1 - Level bands. Should the current level bands (1-4, 4-7, 7-10; 11-14, 14-17, 17-20) be retained as hard limits on top of the new "span the entire tier" model? In other words, for a levels 1-10 adventure, does it still make sense to channel players into three level ranges, or would it be okay to allow a table of mixed 3rd, 4th, and 5th level characters to play together (something that cannot happen under the current system)? If we do want to allow a broader spread of levels to play together, what (if any) limits should be imposed? For example, is it OK for 1st-level and 10th-level characters to play together? If not, what is the widest spread that should be legal?
Topic 2 - Encounter Levels. If we retain level bands as they exist today, should we also retain the ELs? (That would mean each adventure would contain EL 2, 4, 5, 7, 8, and 10 stat blocks.) If not, what spread of ELs should we use? How many sets of stat blocks is it reasonable to expect an author to create? If we do not retain the level bands as hard limits, how do we decide what ELs a given table should be able to choose (and how do they choose, if we get rid of the current low/high tier model)? Should there be a cap on low-level characters earning XP and treasure from ELs that are far above their character level (do we want 1st-level characters getting the same XP and treasure as 5th-level characters if they decide to play up to EL 5 and they are fortunate enough to get a tactically inexperienced or generous DM)?
Our primary goals for the new system, in rough priority order (though we could argue the priority too), are:
1 - People need to be able to play the game. If people cannot form tables at all, then there is no campaign. All else being equal, getting to play is better than not getting to play. (However, all else is not always equal; for example, we have decided that playing at a table of 3 or a table of 7 is worse than not getting to play. That isn't up for debate as part of this discussion.)
2 - Once formed, a legal table needs to have some kind of choice about their desired challenge level and they need to have a reasonable idea how likely they are to be successful at that challenge level. (Some of this can be managed by the DM, but a lot of it is based on the fundamental math of the system, so we can't stray too far from that math.) Note that this is less about "risk/reward" and more about just being able to predict TPKs and walk-overs. TPKs will always happen but there is a line beyond which it should be virtually guaranteed, and a line below which it should be virtually impossible.
3 - The system needs to facilitate efficient mustering -- given a group of 10 players, 100 players, or 1000 players, can each of those event organizers see a way to efficiently get tables formed and have people playing in a timely fashion? When the inevitable situation arises where not all of the players can be seated, do the rules provide a way for the organizer to fairly decide which players are turned away, or does it boil down to who has the most forceful personality gets to stay and the quiet people always lose out?
4 - There should be some kind of fair balance between the amount of work the authors and editors have to do vs. the amount of prep time and cognitive workload that is placed on the DM at the table. (For example, we could ask authors to create 10 sets of stat blocks, but that is unfair to the author when it isn't that hard for the DM to add or subtract a level for all the monsters in a given encounter on the fly; at the same time, giving the DM a list of 20 monsters that could appear in the adventure depending on the PCs' levels and asking the DM to mix and match on the fly is unfair to the DM.)
5 - The system needs to be as simple as possible while still serving the other goals. The current level band system is about as simple as it gets. If you are levels 1-4 you can play this adventure with any other group of levels 1-4 characters; if you are level 5 or higher, you can't play at all. We know the complexity has to increase by at least some amount, because now we are saying if you are levels 1-10 you can play this adventure -- but probably not with any arbitrary group of other characters of levels 1-10, so now you have to look at the next level of the rules to figure out who can play together. Thus, there will be at least one added "step" of complexity for the players to figure out; we want to keep that added step as simple as we can.
The Global Admins have been discussing this and we have also read a lot of posts on this topic on the old regional admin list, on this list, and on the WotC forums. We have pretty much decided that there is no way to make everybody happy. But we do have to decide something, and we wanted to give you one more chance to share your thoughts and suggestions before we reach a final decision. There are three main options that we are considering. We would appreciate it if you could tell us what you think of these options, and if you don't like any of them, what you would do instead (bearing in mind that "keep the current system" is not an option for the reasons I outlined above).
System 1 - Keep the current level bands as hard limits, and keep the current EL options associated with those level bands. So a Heroic tier adventure would still have H1, H2, and H3. All characters at the table must be within the same level band. The adventure would have six sets of stat blocks: EL 2, EL 4, EL 5, EL 7, EL 8, and EL 10. Each level band would still have a choice of low or high tier. H1 would choose between EL 2 and EL 4. H2 would choose between EL 5 and EL 7. H3 would choose between EL 8 and EL 10.
This is essentially identical to the current system but there are two main things we don't like about it: the authors have to create six sets of stat blocks (lot of work) and two of those sets only differ by one level (EL4-EL5 and EL7-EL8) which seems like unnecessary work. People with level 4 characters are discouraged from playing H1 (H2 will have better rewards) and people with level 7 characters are discouraged from playing H2 (H3 will have better rewards).
System 2 - Keep the current level bands as hard limits, but only write stat blocks for the even ELs. So a Heroic tier adventure would still have H1, H2, and H3. All characters at the table must be within the same level band. The adventure would have five sets of stat blocks: EL 2, EL 4, EL 6, EL 8, and EL 10. Each level band would still have a choice of low or high tier. H1 would choose from EL 2 and EL 4. H2 would choose from EL 4 and EL 6. H3 would choose from EL 8 and EL 10.
This is similar to System 1 but it drops a set of stat blocks and spreads things out so that every set of stat blocks is +2 EL above the previous set. It doesn't line up as well with the level bands, though, especially in the H2 tier. If you do the math, which I will leave out because this is already long enough, the combination of the current level bands with even ELs does not result in a smooth challenge curve for characters of different levels (sometimes you have good choices, sometimes you don't).
System 3 - Keep the current level bands as hard limits, but write stat blocks for the odd ELs. So a Heroic tier adventure would still have H1, H2, and H3. All characters at the table must be within the same level band. The adventure would have six sets of stat blocks: EL 1, EL 3, EL 5, EL 7, EL 9, and EL 11. Each level band would still have a choice of low or high tier. H1 would choose between EL 1 and EL 3, H2 between EL 5 and EL 7, H3 between EL 9 and EL 11.
This is similar to System 1 but it matches the ELs closer to the level bands (odd ELs also match the levels when PCs gain encounter and daily powers, as opposed to utility powers and feats, so the odd ELs are arguably a better measure of PC power levels). It is back to six sets of stat blocks so more work for the authors (and some people will argue that nobody would ever want to play EL 1, but you can't remove EL 1 if that would mean the lowest level choice available was EL 3 -- it would be TPK city for tables of new players). Unlike System 2, the math of the available EL choices vs. character levels is smoother (of course having six choices instead of five also helps with that).
System 4 - Drop the current level bands as hard limits. Instead, the table could include any group of characters within some specified range (such as 3 levels, i.e. levels 1-4, 2-5, 3-6, 4-7, 5-8, 6-9, 7-10). All characters at the table must fall within that range (so you can't have 1st and 10th level characters playing together). Write the ELs for either even (2/4/6/8/10) or odd (1/3/5/7/9/11) breaks and let the table choose the EL they want to fight, perhaps with an upper bound based on the lowest-level character in the party.
This, of course, is essentially the same as the old APL system that previous campaigns used. If you set the requirements up so that they are based on the lowest-level character in the party, then you can avoid having to actually take the average character level, but functionally, it's pretty darn close. There are some very good aspects about this (I think it results in the highest probability of seating all players given a random distribution of character levels) but there are also some very bad aspects (it is far and away the most complicated, and it makes mustering harder).
System 5 - You could also make System 4 a free-for-all and not have any limits on who can play together or what ELs they can choose, but I think this fails to meet several of our goals. Mustering would be a lot harder than it is with the level bands, and the system is open to abuse. The prospect of a bunch of 10th levels "powerleveling" their 1st-level buddy doesn't really appeal to me. It's too much like work; especially when people can legally create higher-level characters already. So I think in this case we do need to limit what people can do, if for no other reason than to indicate that they should just create higher-level characters if that's the outcome they want. 4th Edition is designed around the notion that everyone in the party needs to be within about 3 levels of each other and of the monsters. So I list this option for completeness but I don't think it would ever be chosen.
Sorry for the length of this post. I wanted everyone to understand what we have been thinking about and the various options that we are debating. It kind of boils down to a decision tree that starts with "keep the existing level bands or not?" and then branches off to "what ELs can people choose from once they form a legal table?"
Thoughts / questions / suggestions appreciated. I can post some of the math behind some of these systems if people are interested, but the one thing that seems pretty clear to me at this point is that there is no perfect option. So ultimately what we are trying to do here is construct a system that is the least bad for the most people. We are hardly the first campaign to ever have these struggles, of course -- it's intrinsic to any campaign model that allows players to advance their PCs at different rates.
Talk to you later --