A long running campaign is a desirable thing. Yet many (most) campaigns crash and burn before becoming memorable. Few get played very long.
For new players the average length of a campaign f is 8 sessions. This rises up to 12 sessions for people playing role-playing games for under 5 years. (source)
So that's what you can expect from a campaign. Two to three months of play. You roll up a character and get him to level 2 or 3 before the campaign explodes. If you're lucky, you might make it to level 5 before the game fizzles out. Real world spanning memorable play there.
And yet the average campaign length of my games is somewhere around 40 games more than double the average of someone of my gaming experience, usually comprising 2+ years of play.
Why is this?
Your game is going to crash.
What you should do is design the game to take advantage of those crash-only systems. For example, leadership positions can not be permanent, because people die. That system is going to crash. Creating a political structure that formalizes the transfer of power rather than just letting it go till it blows up into a war for power has stabilized an inherently unstable system.
This instead of being a negative thing is very very positive. The reason your games end and burn out is because they have accumulated detritus that accumulates through play that eventually becomes more of a burden then the fun that is had during play. This is why rolling up new characters is so enjoyable—they don't have to deal with the fallout from previous actions and choices. At some point this weight becomes overbearing and it drives the game to a crash.
I'm not just talking about choices in game. This is also the fabled Gamer Attention Deficit Syndrome I'm talking about. You are exposed to new ideas and they build up over time until the weight of all those new ideas conflicts with your current campaign. Then you nuke everything and begin again.
You don't have to.
The serial television show
Recently, gamings actor/philosopher in residence Justin Alexander recently talked about bloat in extended narratives—serial television shows. Serial television shows are very much like a gaming campaign. Characters are introduced, plots are driven forward, change occurs and stays. What eventually happens in many of those television shows is that the show gets bogged down in an ever increasing number of characters, side plots, and other events, until it feels like each single episode only contains about 4 minutes of content due to switching between the various storylines.
He also notes some examples of how some shows avoided the problem. He says:
"One TV series that seemed to largely avoid this problem while also enjoying the benefits of arc-plotting was Buffy the Vampire Slayer. The technique they used seems relatively straight-forward: They limited the number of arc-plots they introduced in each season and made sure that most or all of them were resolved by the end of the season. As a result, when they launched season 6, for example, they didn’t have any more balls in the air than they did when they were launching season 3."And what he's talking about there is a scheduled crash.
The gaming lifeI'm not talking about a single game.
Internalizing this crash process is important, because when you start a new campaign you never do so in a state where you have never ran a game before. You will always start with the experiences, successes and failures of your previous campaigns behind you. There is no "new" campaign. Just the next one. And just like moving somewhere for a fresh start won't solve any of your problems with your life, neither will starting a new campaign solve any of the problems with your current game.
What is necessary to resolve this issue is some examination of your previous gaming experience. Why do your games end?
Do you have gamer attention deficit disorder? Have events in the campaign restructure the way things work drastically. Your players reluctant to engage in the tedium of a long quest? Resolve it in an unexpected way and have the campaign head off in a new direction. Want to try out a different style of gaming? Switch it up within the game itself. Getting burnt out on a setting or project? Restructure your time devoted to running the game, take turns, or have something new or different happen. Did the characters lose focus because there's too much for them to do?
There is no perfect game. They are all going to end, and like all crashed endings: divorces, marriages, new jobs, children, there's pain involved. Thinking that changing things will make it avoidable is a dream from a pipe. Directing your game towards that crash gives you some measure of control over the outcome.
Fly it right into the ground and come out the other side.