So what is accretion? To simplify it down, it's just something getting bigger by adding small pieces over time. Accretion is actually a non-trivial problem for any game that's been around for a while. As more and more stuff gets added you start to run into a number of issues from unexpected combinations of game elements, to power creep and eventually even simply adding anything new becomes a problem. A simple example of accretion in relation to X-Wing miniatures is the card Deadeye. Deadeye was released in wave 2 and immediately forgot about. It made missiles and torpedoes slightly easier to use, but those were pretty useless at that point and nobody wanted to spend a Elite Pilot Talent to make them less bad. Fast Forward all the way to wave 8 and the release of the JumpMaster5000 and the addition of Guidance Chips and all of a sudden, not only was this throwaway card seeing play, it was seeing play in very common meta build. In fact, this build became so common that Deadeye, the card that was largely unplayable for it's entire lifespan to that point, got nerfed so that JumpMaster5000's couldn't use it. Even later, with the advent of really good missiles in the form of first Cruise Missiles and then Harpoon Missiles, this card is seeing pretty regular play. This also presents an ongoing problem, that is, FFG had to assume that any Torpedo or Missile capable ship with an Elite Pilot Talent will be carrying Deadeye which dramatically changes how useful the ordnance being carried by that ship is. This one decision, a card in wave 2, impacted how FFG has to design all ordnance and ordnance carriers in 1.0 X-Wing.Â
What's amusing is that this isn't uncommon. Elements from the very early life of the game are frequently, over time and releases, either amazingly good or mostly unplayable. For examples, let's look at the very beginning, the original core set for X-Wing 1.0. Biggs was a very common meta pilot to build a squadron around right up until he was nerfed and R2D2 still saw regular tournament play right up to the end of 1.0...but everything else from the original core set was mostly out o fthe game by wave 5? This is typically because, even if designers know what they want the core experience of the game to be, they don't always have the best grasp on what they want it to look like 1, 2, 4 or 5 years down the line, so elements that make perfect sense in the early days are either overshadowed or amplified by newer elements that they were never intended to interact with.
So how does a game solve accretion? Well, there have been a few different approaches.
Set Rotation
Set rotation is the process by which older elements are rotated out in favor of newer ones. This is mostly only viable with card games, so we won't discuss it that much, but set rotation does accomplish basically all of the goals to keep accretion in check. It gives you a chance to regularly revisit the cost/power ratio and pull it back or crank it up as needed. It gives you design space to work with as cards only have to compete with their contemporaries and not every card ever released. It keeps the meta constantly changing as the rotation of cards make new decks viable and continuously forces players to rethink common archetypes and that archetypes place in the newer sets. It also allows you to let problem cards die by simply waiting until they rotate out and not reprinting them. This is fine for a card game, but for various reasons, I don't think it will ever be viable for a miniatures game. Miniatures tend to be larger and harder to store than cards on the off chance they'll rotate back in and players get more attached to miniatures (and for games like Warhamemr 40,000 or Warmachine there's a lot of time investment in terms of building and painting).
Landfill
The landfill approach is not as bad or desperate as it appears, but, broadly speaking, it's very unhealthy for the game. The landfill approach basically just boils down to constantly adding new elements, but with a strict caveat: it doesn't work without power creep. Power Creep is the change over time of the ratio of cost to power. In a landfill strategy, each release has a slightly better cost/power ratio (or in extreme cases massively better). The goal is to bury the old stuff under the new stuff. If you release anything game breaking, no worries, it will eventually be buried under other elements that are just costed better. You're probably asking why the hell you'd want to do this as it seems bad on every level. Simply put, the landfill approach doesn't require the same resources because you don't have to worry about the finer points of balance or fine tuning all of the elements. As long as you keep the steady climb upwards, the game continues even if the peaks and troughs will get rough at times. The landfill philosophy is also the poor-man's way to creating a vibrant meta even if that meta almost exclusively contains the last few iterations of game elements, it will be in constant motion.Â
The biggest reason for the landfill approach has been that a game system is failing. Even though this approach is unhealthy for long term viability, that doesn't matter if the game is already dying. The worst result is the game dies a little faster and that's pretty irrelevant. The best result is that the shake up reinvigorates the game and keeps it going. This will create an issue later if you do actually get the resources to do all the balance fine tuning because most of the landfill elements will have to be scrapped, but if you actually get to that point, then you're pretty happy as a game designer. Usually the landfill approach won't be initiated by a design team, it will be the company or, in the case of video games, the publisher and they'll just be trying to get what profit they can out of the property before it goes. As a general rule, if a game starts doing this, it's in pretty dire straights.
The Codex System
It's fair to say that Games Workshop and the Warhammer IP have been around for a while and have had some major problems with accretion. The closest the game really has to a solution is the Codex system. In Warhammer 40,000 all of the armies have a codex. In that codex are all of the rules for that faction (barring supplements, etc.). The codexes are on a constant rotating release schedule and with each new codex, all of the models in the range get new rules and new models usually get added to the range to boot. In theory, this actually solves some of the problems with accretion reasonably well. You get to regularly re-examine and adjust power levels to bring older models back to curve. If there's a problem ability or model you'll get the chance to adjust it as needed. It also keeps you releasing new models on a semi-regular basis to keep the cashflow going even from players who already own most of the faction. To top it off, it also regularly adjusts what's good in the game resulting in near constant meta movement even if that movement isn't particularly fast.
The problems I've seen with the codex system are mostly in how Games Workshop has handled it i.e. Space Marines getting 4 new codexes before Necrons get their second, outright deleting game elements that people have spent time and money modelling or simply making the new hotness better than the old stuff. These aside though, I can't think of a reason as to why a codex system, broadly speaking, couldn't work. What's interesting is not the system, but how gamers react to it. You'll get the pretty common response that they have to buy whole new models, but that's how it should work. You probably owned the above curve game elements because they were above curve and didn't bother investing in the below curve element because they were below curve. If both of these get adjusted properly, yeah, you'll probably have to buy new models.
The Book System
This system was used by Malifaux and, up until Mark 3, was used by Warmachine/Hordes (and that's an interesting story that we'll talk about a little later). The Book system involves releasing expansion books that have some story elements to the game and that also introduce new models. When Privateer Press started using the book system for Warmachine, it was largely a response to the codex system and players being unhappy with having to wait a long time before getting anything new. The book system offered regular releases for all factions and the releases were generally symmetrical in nature...this is actually pretty similar to the X-Wing release model. But this system has a small problem. If all you're doing is releasing books with new content, how do you address old content? The only real way the games have come up with is either by Errata or FAQ or adding new elements that specifically interact with older ones. Quite a few models in Malifaux were hotfixed by adding 0 cost upgrade cards only specific models can take...stop me if this is also ringing any bells for 1.0 X-Wing players.
The only real way the book system can adjust older models is with a new edition as trying to handle it exclusively with FAQs, Erratas or new releases is just a band aid for a wound that will continue to bleed. These transitions through editions are something games want to avoid if possible mostly because the cost both in terms of money and man-hours is pretty substantial and the net benefit, at least financially, is not very high. In fact, the book system, to a large extent, makes accretion worse. There are only so many ways to make a ranged unit or melee unit before you start to get substantial overlap that renders one or more mostly useless and without the ability to easily make major changes to older models over the course of the normal release cycle, those models will continue to be useless or a problem. As an example, the epic incarnation of the Warcaster Asphyxious had to be nerfed 3 separate times in Mark 2 Warmachine...and was still a good caster? The book system also has a major problem in that, once an archetype is cemented in the meta, it usually stays there and new elements only serve to augment this. Without a large scale reworking of some of the moving parts, you're going to see the same archetypes no matter what gets released unless something that creates an entirely new archetype drops.
Community Integrated Developement
The Community Integrated Development system, or CID, was started by Privateer Press and appears to have replaced the book system altogether. The concept is that the CID runs in cycles where a faction goes into a pseudo open beta where older models get re-examined and new models added for the community to test...if this sounds familiar then you get a cookie. This is basically the codex system by a new name, but using the game's electronic app called Warroom to update cards instead of having to release and physical media such as cards or books. The aspect of this system that I feel is hands down the weakest, is the community part. To put it bluntly, my experience with open betas and the average player in general has me in a state of skepticism about how useful the input of most of the players is and some of the results that have come out of CID have only cemented this thought. That reservation aside and much like the codex system, there's no reason why this couldn't work as a countermeasure to accretion. It keeps the game moving, has the potential to change the power level of list archetypes, can adjust or remove problem abilities and still keeps you on a semi-regular release cycle. Since this system is basically just an electronic codex system, it's hard to really talk too much more about it, but Privateer Press has also looked into another route to combat accretion.
Theme Lists and Formations
So Warhammer 40,000 used what were called Formations in 7th edition (and dropped them entirely in 8th edition). Formations had a specific requirement of things to take (1 of unit Y, 2 of unit Z, etc.) and offered benefits that ranged from "When would I ever use this?" to fairly obscene. There are some obvious applications to this, that is to say, offering you a benefit for using a narratively interesting combination or underutilized element that could overcome some of it's issues...but that's not really what happened.  The Optimized Stealth Cadre is a pretty prime example. If you took a Ghostkeel unit and 2 Stealth Suit units you got a benefit that was absolutely obscene. While the Ghostkeels were decent enough in their own right, the Stealth Suits were pretty terrible in every iteration of the Tau Codex including the 7th edition one. While on paper, this encouraged the use of this unit, all it really did was force a tax on taking Ghostkeels of having to drag around 180pts of wet blanket Stealth Suits, so naturally you would only ever field the largest possible Ghostkeel unit and the smallest possible Stealth Suit units. It didn't so much solve problems as create new ones where the underpowered formations were ignored and the overpowered ones showed up everywhere making for a lot of copy-paste lists even if you weren't trying to net list. I'm not surprised the mess of formations were dropped for 8th edition.
Privateer Press took a slightly different approach for what they call theme lists. In Mark 3 of Warmachine and Hordes, Privateer Press has basically gone all-in on the theme list concept to the point where the goal appears to be for every list to be a theme list. Theme lists only allow you to take specific models, but give you some fairly solid benefits (most of them directly equate 15-20% extra effectiveness either through bonus abilities or just getting free models). I can see the theory behind this. Accretion makes it increasingly hard to add new elements because those new elements have to compete with everything ever released for that faction and there's only so many ways to make a ranged unit, melee unit, warjack etc, before you start to see serious overlap. So instead of having them all compete with everything else, they only compete with other models in their chosen theme. If they were to release a new melee Dawnguard unit, that unit wouldn't have to compete with every other melee unit, just the other Dawnguard ones. The problem is that the theme lists have come with unintended consequences. The big consequence that I've seen is meta full of inflexible lists that can't naturally adjust to a shock. Having everything be a theme list means that if something starts dominating the meta, even if an answer exists, it might not work. The answer might be in one of the weaker themes or split across multiple themes resulting in most factions not having a meaningful answer which is what happened with the Ghost Fleet theme list.
Closing thoughts
To put it simply, the book system where you just do roughly symmetrical releases for everyone and only address problems with FAQs, Erratas or adding new elements, seems to have a service life. Privateer Press definitely appeared to hit it with Warmachine/Hordes. Malifaux still ran it in 2E, but that's a much smaller game and company and is less about finely tuned balance (but that would be another lengthy discussion). X-Wing had to change it's model into digital points and do a full-on new edition. X-Wing's move to virtual rules wasn't quite as extreme as Privateer Press' (Privateer Press doesn't even include physical cards with their models anymore), but has helped it a lot and allowed for a very healthy meta where problems are easily fixed.
For all the downsides of accretion, there is one upside. Once a game gets big enough, nothing can ever be dominant because there will always be an answer floating around somewhere in all the obscure and underutilized elements. I'm not sure if X-Wing is there yet, but it does make for a meta constantly rotating between measure and countermeasure.