To a lot of people, getting crashed is just part of OGame. Part of the cycle. You get crashed, you rebuild, you get crashed again.
It isn't a law of the game. It's a habit.
Proper fleeters hold their fleets for years. Not because they're lucky, and not because nobody's trying. Because they follow rules, and they follow them when it's inconvenient.
This isn't a technical guide. I'm not going to explain how to fleetsave. This is about how you think — and it isn't only for fleeters. Fleeters have huge fleets. Miners have huge savings. The mistakes look different and cost the same.
The question isn't if. It's when.
You're human. You will fall asleep. You will log in too late. You will misread a timer, mistype a coordinate, get stuck somewhere with no signal.
That's not pessimism, it's arithmetic. Over a long enough run, every mistake you're capable of making, you will make.
Now the second half, and this is the one people skip: when it happens, someone will be there. Assume you are being watched by somebody patient, somebody who has been waiting weeks for exactly this. Because that person exists, and they are better at waiting than you are.
So the question is never "what are the chances." The question is what happens to you when the chance comes in.
Part I — Don't make the mistake cheap to exploit
You can't stop yourself from slipping. You can decide in advance how much a slip is allowed to cost.
Everything in this half is about that: closing the openings you'd otherwise leave lying around, so that the night it goes wrong, there's less sitting there to take.
The math is never in your favour
Look at what you're actually betting when you stay up for one more raid.
You're risking your entire fleet. You're winning a fraction of it. That's the trade, every single time. It doesn't matter how good the target is — the upside is a slice, the downside is everything. Nobody would take that bet if they wrote the numbers down. People take it because they never write the numbers down.
Same for the last half hour of expeditions. Same for the fleetsave you shortened because you wanted to squeeze in one more round.
Miners, this is your section too. Most miners save for the next project by simply keeping everything in the fleet. Resources go in, resources stay in, and when the pile is big enough it becomes an astro or a mine. If you get crashed the day before that, you don't lose a raid's worth of profit. You lose the entire project.
And the reason people do it isn't that they've weighed it up. It's that hiding resources properly is annoying. Dumping into constructions, cancelling them later, pooling everything at the end — it's micromanagement, and nobody enjoys it. So they take the easy route and carry weeks of progress around in the open.
You didn't decide the risk was worth it. You just decided you couldn't be bothered. Those aren't the same thing, and only one of them is a decision.
No risk is small enough
A dirty save off some random moon looks fine. What's the realistic chance someone pops every moon you have?
Wrong question. The chance doesn't matter, because you're not being paid enough to take it — see above. Anything above zero is already too much when the stake is the whole thing.
The practical version of this rule is buffers. Assume delays. Assume traffic. Assume you get held up, the train is late, the meeting runs over. Don't set a return for the exact minute you expect to be back at your desk, and then buy yourself thirty more minutes of farming with the margin you had left.
And never, ever build a plan on the idea that nobody will notice. Somebody notices. That's the whole job for some people.
A fleet you can't fuel is a fleet you can't use
There is no deut bill too high to keep your fleet safe.
Taking the cheaper, less safe route because the proper one costs too much is one of the most common ways good players lose everything. You'll do it once and get away with it, which is the worst possible outcome, because now you'll do it again.
If your income can't cover safe saves, you don't have a fuel problem. You have a fleet you don't know how to use.
A fleet has running costs. Whether it ever repays what you spent building it is a different argument, and one I'm not making here — plenty of fleets never do, and people build them anyway. The bar I'm setting is much lower than that: it should bring in enough to keep itself safe.
If it can't clear even that, it isn't earning, and that isn't the fleet's fault. Nobody forced those ships on you. You built something you can't operate, and now you're cutting corners on safety to cover the gap between what you own and what you're actually able to do with it.
Scrap it. Put it into mines. Build it again when you can feed it.
If tonight doesn't need it, put it out of reach
Before anything else, ask what the job actually requires.
Staying up to farm? Going after someone's moons? Work out the minimum that operation needs and leave the rest where your mistake can't touch it. Most people bring everything out of pure habit, then spend the whole evening with their entire net worth on the table for something that needed a fraction of it.
Resources are the same principle, and for miners this is the whole game. If you're not spending them, they shouldn't be sitting in a fleet waiting for you to slip. Hide them. Pool them only when you're actually ready to spend, so that a bad night costs you the last few days instead of the last two months.
Don't put all your eggs in one basket
Literally and metaphorically.
To be clear about what this isn't: it isn't "split your fleet." Sometimes splitting is the right call, and sometimes it's how you turn one untouchable fleet into two crashable ones. Ships have their own logic and it doesn't always point the same way.
The point is the basket, not the eggs. Go through everything you own and ask which single failure takes all of it. One alarm not going off. One person — you — not being awake. One routine that everyone can read. One pile of resources that represents every hour you've played this month.
Most people have far more riding on one thing than they realise, because the basket usually isn't a place. It's a moment. Everything you own converging on the same minute, depending on the same person to be at the same screen.
Find those moments and take things out of them. Your worst night should cost you a piece. If it can cost you the server, you haven't found the basket yet.
Part II — Assume you made the mistake anyway
Everything above shrinks the odds. Nothing above makes them zero. So build the second layer for the night the first one fails.
Your absence should look exactly like your presence
Earlier I told you never to bank on nobody noticing. This isn't a contradiction, and it's worth being precise about why.
Deception is not a safety measure. You never count it in the plan. You build a setup that survives being watched by somebody who knows exactly what you're doing — and then you make sure they don't.
The first half is the obvious one: don't be predictable. Activity at the same hours every day, saves that always go the same way, from the same places, in the same shape. Every pattern you show is a free piece of the puzzle, and somebody is assembling it whether you believe it or not.
The second half is the one most people never think about. If every scrap of activity anyone can see on you is real — you, at your screen, doing something — then your activity is a live feed of whether you're there. The moment it stops, everyone watching knows you're gone. You've handed them a light that switches off the instant you become vulnerable.
So cover it. Generate activity during the hours you aren't there. The game gives you more than one way to produce it without a human behind the keyboard, and finding them isn't hard. Do that and the feed stops being useful to them: your absence looks like any other quiet evening. Whoever is watching now has to burn time working out whether you're actually gone — and time is the one thing you need most on the night you slip. Often they won't work it out at all, because from the outside nothing changed.
But understand what this is made of. Fake activity is worth exactly as much as it is indistinguishable from the real thing. If anything else visible contradicts it — something that only happens when you genuinely aren't there — then you haven't hidden your absence. You've labelled it. You've told the person watching precisely which hours you're bluffing in, and you've done it more clearly than silence ever would have.
Activity is one signal. It isn't the only one you produce.
You are readable in more ways than you're currently counting, and the ones you've never thought about are the ones still telling the truth about you. Which is where a cover starts working against you — because the most likely thing to contradict it is the cover itself.
Whatever you're using to manufacture activity is doing something, and doing something leaves a mark somewhere. The trap is that the convenient way to set a cover up is the tidy way: the same thing, the same size, at the same spacing. Nothing about a person at a keyboard is tidy. Build it the easy way and you haven't covered your absence — you've drawn a line around it, and made the moment you sit back down easier to spot than it was before you started.
A cover is never something you trust. It's something you check. Work out what yours leaves behind — then go wider, and work out what moves when you're at the screen and what moves when you're not. Assume somebody has already noticed the difference.
Which means the cover has to be complete or it shouldn't exist. Half a disguise is worse than none.
Somebody else has to be able to save you
You know the mistake is coming. Decide now what happens when it does, because you won't be in any state to improvise at the time — you'll be asleep, or stuck somewhere, or ten minutes too late.
A number you can call. A number other people can call, so that whoever spots your fleet sitting in the open at 3am has a way to wake you up. A sitting token with somebody who knows what to do with it.
Real allies are worth more here than anything you can build for yourself. Four eyes are better than two, six are better than four. Not as sentiment — as infrastructure.
Which starts with being honest about who that is. An alliance tag is not trust. Alliances are looser than anybody in one likes to admit — people are in them because they're friendly with somebody else who's in them — and plenty of players extend trust on the strength of the tag alone. They're the ones surprised when somebody leaves one day and turns around. Your real list is shorter than your member list, and you should know exactly who's on it.
Then it becomes a routine, and the routine is all of it. Times posted and times read. Eyes on each other's landings. Somebody willing to probe you and say you've left something sitting. An alarm on the hours that matter.
None of that is clever, and that's the point. What stops people isn't difficulty — it's that there's no reward on the nights it wasn't needed, and that's nearly all of them. This is what following the rules when it's inconvenient actually looks like. Not the dramatic night. The eighty before it, when nothing happened and nobody needed you. The night it matters, you don't get to start.
It does something else too, quieter. Writing "back at 03:40" means you had to work out when 03:40 is — the decision got made while you were awake and unhurried, instead of improvised at the hour you're least able to.
None of which means telling your allies everything. Need-to-know applies inside the circle too. And your times shouldn't be worth much even to somebody who turns, because you shouldn't be fleetsaving to your real life's timetable in the first place. Shuffle them. Somebody reading a month of your saves should come away with less than they expected.
Everything else in this manual is you covering your own back, and there is a hard limit to how far that goes, because the failure you're planning for is you. Somebody has to be able to cover for you, because one night somebody will have to.
The short version
The players who keep their fleets for years aren't lucky, and they aren't paranoid. They just stopped taking bets that were priced against them.
You will make the mistake. Build the setup that survives it.
— RiV-
Edited posts together by King Rabbit
