This montage is based on a storm image by Artur Pawlak, with lightning strikes from an image by skeeze and another from an image by AIAC Interactive Agency. Editing and color effects added by Mike. The button is from a set by 538778. All the images were sourced from Pixabay.

I Spent much of the weekend a few weeks ago performing my regular data backup.

Everyone in IT has horror stories resulting from inadequacy of backups. Even IT professionals and past professionals like myself – who know how essential backups are – have lost irreplaceable material to inadequacy of backups. That’s because there are two parts to the backup equation and people focus so intently on the first that they tend to ignore the second.

The first part:
     data → backup process → storage

The second part
     storage archive → restore process → restored data

When my old computer’s hard drive was failing (about 10 years ago), I very carefully backed up everything that I considered essential. Much of the most irreplaceable data went to a DVD-ROM that I had burned. This was a backup performed under considerable time pressure; the PC could fail completely at any moment because the Operating System itself was being rapidly corrupted. As a result, all my focus was on the first process.

Swapping the old hard disc into a new computer wasn’t an option; the cause of the Operating System corruption was impending hard disk failure – or so it seemed at the time.

Thus, there was total reliance on the backups to preserve my data.

I had a low level of confidence that everything could be saved, but I had to try.

This pessimism turned out to be Justified: the DVD-ROM could not be read by the drive in the new PC. I still don’t know why.

I persevered, having little other option. I was able to recover about 75% through other archives. Eventually, on a third machine, I managed to find a settings tweak that enabled recovery of the archived file from the failing Hard Disk – only to discover that the low-level program used to compile the archive would not function under my new operating system. It was Windows 3.1 software, and Windows 7 didn’t want to know about it. Nor does Windows 10, for that matter. But it worked perfectly with XP.

Essentially, it was a hundreds-of-Gb locked filing cabinet – with misplaced keys.

The remaining information was lost.

My documents and web links were saved. Everything CM-related was saved. Some artworks were lost, and some corrupted. Some material relating to old and unplayed campaigns was lost, but everything relating to the then ongoing campaigns was preserved. Some of my carefully-curated music files were lost – my Rolling Stones collection, for example – but a lot was preserved through other archives, too. And I still had the original CDs for a lot of what was lost. Lost and irreplaceable were all my original musical compositions with only a few exceptions. That includes the half-mixed MP3 versions that I was planning to put out as my first CD.

Even if the best-case scenario had prevailed, and all the data had been restored, or at least recovered, there would still have been some losses. A Backup is a snapshot of the way things were at the moment of the backup; at best, you can return to that moment in time. It’s a bit like reloading a saved game when computer-gaming – anything done (for good or ill) since the last save never happened.

But I’m not actually here to talk about the vagaries of computer dependence. Suffice it to say that lessons have been learned and the current backup protocols are unlikely to yield the same outcome should the worst occur!

I was preparing for this particular backup the day after my superhero campaign had resumed after a lengthy Covid-19 -induced hiatus, and the two thoughts connected to raise an interesting question: Backup and restore of a campaign? How?

Two processes

It’s clear that we’re talking about two processes – focusing on the first will get you nowhere unless you have already solved the second. The lessons of the computer world are directly applicable.

The first is making preparations for a potential future disaster. What prep-work is required will depend on the solution to the problem, and has to be specified as part of the description of any solution. Let’s call it the “backup process”.

The second is rebooting the campaign using that prep. This process will vary according to the prep process, which should have been designed with this usage in mind.

It’s not enough to be able to back up the campaign, in other words – you need to be able to use that backup to press the great reset button in the sky.

Different Emergencies, Different Protocols

What may not be as clear at first glance is that different problems will require different solutions. Specifically, there are two different types of problem and a host of different circumstances under which each can apply.

Problem #1: The PCs are about to make a mistake of campaign-killing proportions. This may or may not result in a total party kill (TPK). The GM can see it coming, and has hinted as broadly as he dares, but they aren’t taking any notice. A train-wreck is inevitable.

Problem #2: The PCs are killed. All of them, a TPK. This might be at the hands of the main antagonist (better) or in a random encounter (worse). They may have been unlucky, or may have made a mistake, or it might be the DM who has made the mistake. The train-wreck is either in progress or is complete.

Seven Solutions

I have seven solutions to offer, some better suited to one of these problems than the other. With some solutions, it will be necessary in the case of Problem #1 to let the disaster play out – at least to the point where the players can see that defeat/failure is inevitable – before the reset button can be pushed.

I have to admit to hating some of these solutions with a passion – but not so much that if it was the only way out, I wouldn’t use it; I would just try to impart some clever and original spin on it.

It only makes sense to me not to rely on any one or two solutions to the crisis. Between the two subtly-different problems and the range of possible circumstances, any one of them might be “the best answer” or “the right answer”.

That mandates that they all be low-impact low-prep sustainable solutions, so that it’s practical to have them all on standby. So that’s an additional factor to be taken into consideration.

The solutions I’ll be looking at are:

1. The Nightmare Is Ended
2. Dystopian Recovery
3. New PCs, Same Old Problems
4. Back In Time
5. The Jailbreak Adventure
6. Temporus Interruptus
7. The Great Reset Button In The Sky

1. The Nightmare Is Ended

One of the solutions to the problem that I especially despise is “it was all a dream”. Nevertheless, this can be a vital escape clause, and times when – because what took place wasn’t a single catastrophic error but an accumulation of smaller mistakes over a longer period of time – when this is the best answer.

One example is where the players have completely misinterpreted something the GM has said. If you notice this discrepancy at the time, you can correct the problem – but are then fighting Confirmation Bias, which is often a losing fight. The article to which I have linked offers possible solutions to the problem, but they generally require game time to implement – and that’s a luxury you might not have. So there are times when this is the best answer.

Prep Component

Prep consists of three things:

• keeping your campaign well-documented;
• selecting points within the game play that will be suitable platforms for a restart;
• having, on tap, some sort of original twist on the “it was all a dream”. This may in fact be a full adventure in its’ own right or a single scene.

The first item you should be doing already.

The second is a little trickier to offer advice on; it has to be before the “critical mistake” was made or the situation became irrecoverable, but there are a host of other considerations – what the PCs were doing at the time, what the state of the campaign was, and so on. There are so many variables that the decision will be half-instinctive. Ideally, you would like to be able to sum up the PC’s situation, and the current state of the campaign, in a single paragraph of no more than half-a-dozen lines, but that’s rare. Lulls in the emotional intensity are also preferable to moments of high intensity, because the restart will inevitably be such a lull. Ultimately, each GM will have to assess his own campaign as it proceeds to locate the most recent ‘reset point’, and such determinations will be as much instinctive as they are logically justifiable.

The third item is probably the most difficult. In a sci-fi or superhero campaign, you could have “glitches” in the artificial reality around the PCs start to show up in an adventure following the train-wreck. It gets harder in a fantasy campaign, or in one more tightly rooted in reality. Even a pulp campaign would struggle with plausibility with such a plot device despite the weird science element of the genre.

That’s not to say that it can’t be done – a divinely-sourced “cautionary tale” works in a fantasy context, for example, or a “tortuous nightmare” from a demon or devil. Such a revelation might be achieved by simply starting another adventure (after the party have ‘died’) in which they are all killed, and then another – with the deaths seeming increasingly improbable, or the enemies they face increasingly overpowered (Vorpel Bunnies?)

So there are ways of doing it in virtually every campaign – the literal “it was a bad dream” (‘What did you put in those beans last night?’) is the worst choice but is always there if nothing else will serve.

Without such a twist, I would prefer any of the other solutions to this one.

Implementation

Implementation is comparatively simple – everything after point “X” in the campaign didn’t happen – but the PCs get to keep experience and objects acquired since, and anything consumed is gone. The PCs, in other words, are healthy versions of the characters as they are, despite everything else being reset.

This compromise with reality is necessary to make the option palatable to players, and because any alternative risks desynchronization of characters, which occurs when PC#1 resets to a point X game sessions previous, PC#2 resets to a different point in the campaign, (because those are when their backup copies derive from), while PC#3 doesn’t reset at all (no backup copies). Since that’s obviously unfair to some, the only solution is to keep the characters as they are.

2. Dystopian Recovery

The second solution is to let the campaign die the resulting death, and start a sequel campaign in which the bad guys have won, and the new PCs have to find a way to undo it or overthrow them – in other words, to recover from the consequent dystopia.

Prep Component

This requires you to always know what the antagonists are trying to achieve, which is something that I’ve been recommending as good practice for a decade or more.

It also requires some awareness of the significance of the PCs within your campaign. There’s a whole article to be written on that question, but for the moment, suffice it to say that some campaigns treat the PCs as extraordinary, or in privileged positions by happenstance or opportunity or fate.

Another question that often feeds into the status of the PCs is how the campaign treats prophecies and ‘destinies’. Do they have an underlying reality, or are they the products of arrogance and wishful thinking?

Many GMs won’t be able to answer those questions, because they haven’t considered them important – until now, when they manifest in tangibly different outcomes. If the PCs weren’t unique, their loss won’t end opposition to the antagonists. If they were in a privileged position but were otherwise normal representatives of their societies, such opposition becomes a lot more easily discounted. And if they were exceptional by virtue of being PCs, then such opposition becomes irrelevant.

The answers don’t have to be the same from one campaign to another; but there will be manifest differences in campaigns resulting from the answers. Sometimes, these differences are more noticeable than other times, is all.

The prep required is, therefore, to know and understand your campaign – not just what has happened, but why it has happened, and why it has happened to these specific characters.

Implementation

It may be tempting to start the new campaign at the moment (approximately) that the old one ended, and to actually have the antagonists success occurring in “real time” within the new campaign. This means that most of your campaign notes, maps, etc, will apply 100% to the new campaign.

Charting the course of a war that you already know will be successful can be a lot of work, however, and risks a premature confrontation between the new PCs and the old enemies – which would bring the “rebooted” campaign to another train-wreck moment. There can also be a problem with the players knowing more about what’s going on than their characters should.

All these problems are avoided – at the cost of more prep work – by setting the new campaign some time after the first. That means that all those campaign notes will need to be updated at once (instead of piecemeal as the campaign unfolds) but offers the opportunity for what the players know to become “common knowledge”. Any logical failings within the conduct of the conquest can be written off as history being fogged, distorted, or lost after the fact. As the winners, the antagonists will surely have rewritten history to their tastes and perceptions!

A big factor is how much time you have to complete this work, because your campaign is shut down until it’s done! Experience tells me that the shortest such interval that’s realistic is two or three months, and possibly longer.

That’s a shutdown that’s avoided by an immediate restart. So the situation is not as black-and-white as it might at first appear.

3. New PCs, Same Old Problems

Which gives rise to the next solution: new characters, at the same experience level and capabilities achieved by the old ones, who simply take over the problems faced by the old PCs. This can provide an opportunity to revisit those “big picture” decisions embedded in the old campaign – perhaps the old PCs failed because they weren’t the ones “predestined” to confront the antagonists, no matter what they thought, but the new ones are.

The big advantage that this brings to the table is that little or no prep is required (other than the generation of new characters and their backstories). What’s more, it keeps “Dystopian Recovery” in your back pocket should this group of PCs also fail to stop the antagonists.

The players, knowing more about the antagonists than they probably should, can actively design their characters to face the challenge – which fits in nicely with the concept of a “predestiny”.

Prep Component

There is still a little prep to be done – aside from the generation of new characters. If these PCs are to be predestined to confront the antagonists that wiped out the previous crop, do they know it? What do they know about it? Are they to have specific roles to play in overcoming the adversaries? The answers will influence the design and creation of the new PCs. (Another key question: why hadn’t the old PCs heard of these New ones?)

But that’s about it. Everything else is normal adventure prep.

Implementation

The assumption would probably have to be made that this group of new PCs came together at some point historically. The new campaign should probably start with the confrontation with the antagonists and will need to be written specifically for the new characters. Do they get to witness the defeat of their old characters? Or do they arrive some short time after that has happened? Did the old PCs have any gear that was essential to overcoming the antagonists? These are decisions that can be taken quickly – while the players are generating their new characters, in other words – but are probably better taken with some deeper thought. It depends on how much of the old campaign was pre-planned, and how much was improvised.

4. Back In Time

Another variant on the same theme is to actually go back into the campaign’s history and roleplay the development of the new PCs from their first meeting / first adventure. If the problem arose because the GM was overwhelmed by the complexities of the more powerful PCs, this might be the right choice.

Prep Component

Most campaigns evolve over time. History gets added and fleshed out and supplemented. All that canon needs to be integrated into a new campaign background. If you’ve been maintaining good records electronically, this can be largely a cut-and-paste operation; anything else is more work.

You can get away without contemplating the big questions posed in “New PCs, Same Old Problems”, but you might be better off taking the time, anyway. It will help guide you in answering the biggest headaches that are sure to arise: the new PCs not liking the way something turned out in their old PCs past experience and deciding to do something about it.

Ultimately, this is a time-travel campaign, with all the attendant problems – even if the PCs are unaware of it.

Can the new PCs change the history of the old campaign? If not, what stops them? What are the consequences, either way? If not, what does that do to player/character freedom of will? If the PCs have a destiny, who else has one? Will it inevitably happen in some fashion, or do the PCs have to embrace it first – and what does that entail? What if they decide to leave the antagonists alone? What if they decided to do the equivalent of killing Hitler’s mother?

The more the new PCs change the game world experienced by the old PCs, the less value and relevance their ‘inherited’ player-knowledge will have. The less that they are free to change it, the more value and relevance it will have – and the more the new PCs may be able to take advantage of that.

Big decisions. Big problems. Answer them in advance or be caught out by them.

I know of some GMs who require all players to generate a “backup character” in case the primary falls – it’s not an idea that I really approve of, because it also brings behavioral problems with the players, such as deliberately suiciding a character to get to the backup.

I also know of some GMs who run two different groups of PCs in the same game world at the same time with the same players, just so that if one falls, the other group can take ownership of the more significant issues belonging to the first group. If you can make this work, more power to you – the potential for problems is extraordinary, and the list of such problems far too extensive to go into here! Suffice it to say that they all stem from character interaction (overt, covert, or metagame).

Implementation

Implementation of this situation is relatively straightforward, especially if you’ve done the prep before you need it. You just tell the players, “your characters are all dead. Get out your backup characters / roll up some new ones. Here’s what you need to know…”

5. The Jailbreak Adventure

The name of this solution will make a lot more sense when I tell you that the “Jail” is imminent death. Without warning, the PCs (still alive) find themselves somewhere else – it might be the afterlife they are expecting, or not. Wherever it is, they have to “break out” and get back to their mortal bodies with the means to refresh / rejuvenate them.

This could be done as a shared delusion, with no real healing/rejuvenation – the characters simply went above and beyond what mortals are supposed to be able to do, winning a victory for the ages – and then succumbing to the wounds they have received.

It could also be done as an objective reality – the healing / rejuvenation is real, but the opportunity to draw on this reality is constrained in some way and will not be granted a second time.

This really is a “get out of jail” for the campaign, but the PCs have to earn it ‘the hard way’ – hence the title.

Prep Component

This is, essentially, a mini-adventure that you are dropping whole into the old one to give the PCs a way out That could be with the cooperation or opposition of whoever runs this place. There may or may not be a price to pay, but it should definitely NOT be a freebie.

You may be tempted to make the price tag something to be paid in the future – I’ve tried this and it doesn’t work very well; it smacks of being exactly what it is, an escape hatch being dropped in to save the PCs. Verisimilitude is lost, Suspension of Disbelief is broken, and your Credibility is shot to pieces, in one fell swoop.

Avoiding those three undesirable outcomes – which can happen even if you don’t make the ‘deferred payment’ mistake – has to be the number one design consideration of your adventure. The more ‘tacked on’ it feels, the more you flirt with one or more of them.

Sow the seeds of plausibility early, and water them with attention in the campaign from time to time. This is ultimately a limited Divine Intervention, whether you label it as such or not – prepare accordingly. Do whatever you can to avoid this being perceived as a dues-ex-machina.

The place to start is with the “big questions” posed above. From the answers, you can determine who has intervened, and what the price will be. If this IS the afterlife, then the intervention has occurred post-mortum – but people have been brought back from the dead, before.

In some old-school gaming, the long lives of the Elves was bought at the price of them not being subject to resurrections. Thinking about this situation became a cornerstone of my Rings Of Time campaign, in which the PCs goal was nothing less than Ascension. The path to that for the Elven character was for him to die and be “rescued” from the Elven Afterlife by his Dwarven companion (the other PC).

You need this stand-by drop-in adventure ready to go at a moment’s notice – so make sure to update it / review it anytime that it might become necessary.

Implementation

Implementation is nothing more than the PCs waking up somewhere else, feeling fine, as you pull out your adventure notes from wherever you store them.

But implementation is also learning to live with the consequences – and whatever you decide in terms of those ‘big questions’ will have consequences. If you aren’t prepared to pay that price, put this solution back on the shelf.

6. Temporus Interruptus

This is a variation on the above. “I have intervened a instant before you perish from your mortal existence because no-one will notice your absence. Agree to my terms or I shall return you hence, in condition unchanged from that you experienced previously. Do a “small” task for me, and it might be that when you return, it will be in markedly better condition, with what you need to have a chance of avoiding the downfall that was so imminent.”.

Prep Component

The credibility of the rescue depends on two things: convincing the players that their characters were dead and gone from the moment battle was joined (even if it wasn’t), and making the NPC making the offer, and the offer itself, as credible as possible. That means that whatever the “task” is, it has to be difficult but within the players’ capabilities, and yet it has to be something that the being who has interceded cannot do himself. Get these things right, and it will seem like this was always your intention.

That often means starting your ‘sales pitch’ before the PCs have even fallen – the flavor text that you use to heighten the drama of the conflict can do that job, especially if you save the really good stuff for such life-and-death battles.

Adding those components to the requirements listed under item 5, “The Jailbreak Adventure,” completes the recipe.

Implementation

This solution bypasses most of the credibility issues described in “The Jailbreak Adventure” by crystallizing them into the specifics described above. As I said, do it right, and it will look like this was always what you expected to happen – a hidden extra chapter between the confrontation and the victory.

You are giving the PCs their lives, and a victory to boot – make sure that the price demanded is commensurate.

7. The Great Reset Button In The Sky

This is the most analogous to the computer-based inspiration for this entire article, and is the only solution to the problem that I haven’t seen written up somewhere in some form over the years – which is why I have lent it’s name to the title of the article as a whole.

Prep comes in three parts: “Checkpoint Adventures,” “Character Backups,” and Synchronization of the Backups with the Adventures. The latter is how the problems described in earlier solutions are avoided.

Implementation is simple: “It is X days/weeks/months ago. Your characters are as they were. Nothing from this date forward is set in stone, so don’t rely too heavily on what happened last time after the point. Your characters are currently….”

Checkpoint Adventures

A checkpoint adventure is one that is designed to be a good platform for a campaign reset. That means that all the ongoing plotlines at this point have to be in a stable position at the end of the adventure – not resolved, necessarily, but ‘on hold’ for a period of in-game time, literally “a problem for another day”. Ideally, the checkpoint adventure will touch base with each of those plot threads in some fashion, so that your notes for the adventure will function as a player briefing on the state of the campaign.

Furthermore, the adventure’s primary plotline should come to a concrete and definitive conclusion. For this reason, Christmas-themed adventures seem particularly suited to the purpose. “Peace on earth and goodwill to all” – at least temporarily. “No Boom Today. Boom Tomorrow. There’s always Boom Tomorrow,” to quote Ivanova from Babylon-5.

Another good time is at the end of a major plotline, for obvious reasons.

None of this will happen by accident, or at least, should happen by accident.

Character Backups

Players have to be told that the purpose of the adventure is to provide a “Checkpoint” to which the campaign can revert if things fall apart before the next checkpoint is reached. Before you next play, you need a copy of their character sheet, dated – because that’s what they will be required to revert to in the event of a catastrophic disruption of the campaign.

These are easily produced using ‘copy file’ and ‘rename file’ if they are electronic in nature, but in most cases, will require a photocopy.

Make sure that players know that if you have to reset the campaign, they WILL reset to the last backup version provided – even if that’s the one you made when the campaign first started – so it’s in the player’s interest to ensure such provision in a timely fashion.

…Synchronized with the Checkpoints

It’s obviously vital for the character backups to synchronize with the end of the Checkpoint adventure fairly closely if not perfectly.

It can also be useful getting players to synopsize what’s happened to their character since the last checkpoint and where they want to go in the near future with their characters. No matter how succinct they may be, these will help the players recapture their character’s frame of mind at the time of the backup. This isn’t essential, but it can be useful.

The Do-overs

That caution offered regarding relying on player knowledge needs to be more than an idle threat, and should become apparent with the first post-reboot adventure. NPCs who have not already been established should be different, their functions within the plot should be different, the plot itself should be different… you may end up getting to the same point in the campaign that caused the reset, but you should get there by a different path.

Don’t forget that you, as GM, can (and should) also be capable of learning from the past experience of running the subsequent adventures and be making an attempt to do them better this next time around. “Don’t do it the same, do it better” should be your mantra.

Play On! (With Twists)

Of course, that’s good advice that can be applied (to some extent) to all the offered reset buttons.

Make no mistake – the less experience you have as a GM, the more prep work will be required to prepare for the worst, but the more likely you are to need that prep. But even seasoned campaigners like myself get caught out from time to time.

I have the advantage that I not only like to think about “the big questions” in advance, but that I formalize the answers and even implement changes to the game mechanics (if necessary) to implement those decisions and their consequences. That, plus a willingness to improv if I have to (especially if I’ve prepared in advance for it) put most of these solutions at my fingertips, should I need them.

GMs, do your homework – you never know when you’ll be called to the front of the class!