Start with how long you can actually stream
Work backwards, not forwards. Decide the longest subathon you are physically willing to do — three days, a week, whatever is honest — and set the rules so a realistic night of support lands comfortably inside it. Picking a generous seconds-per-sub first and finding out what it implies later is how people end up trapped by their own timer.
The starting time matters less than streamers think. Anything from two to twelve hours works; long enough that the stream does not feel like it might end mid-sentence, short enough that the first few subs visibly move it.
Seconds per sub: the number that decides everything
The honest way to pick it is arithmetic. Take the number of subs you get on a good stream, multiply by your seconds-per-sub, and look at how many hours that adds. If a normal night would add more time than you can stream the next day, the rate is too high — no amount of "it will be fine" changes that.
- A small channel can afford generous rates: five minutes per sub is exciting and rarely runaway.
- A mid-size channel usually lands between one and three minutes per sub.
- A large channel often needs well under a minute, or a cap, or both.
- Bits, donations and gifted subs should convert at a rate that feels fair next to subs — not an afterthought bolted on during the stream.
Test the arithmetic against your best night, not your average one. Subathons attract support precisely when the timer is visible, so your "good stream" numbers are the floor, not the ceiling.
Caps: the least popular rule that saves the most subathons
There are three kinds worth knowing. A hard cap on total remaining time stops the timer from ever exceeding, say, 48 hours no matter what comes in. A daily cap limits how much can be added per stream day. A tapering rate cuts the seconds-per-sub once the timer passes a threshold, so support still counts but counts less.
Whichever you choose, say it out loud before the first sub lands and put it in the panel. A cap that is announced reads as a plan; the same cap introduced on day four reads as moving the goalposts, and that is the one thing a subathon audience genuinely resents.
Sleep, breaks and pauses
Decide in advance whether the timer pauses while you sleep. Both answers are legitimate and both need to be stated. Running through the night is the purer version and the one that produces the famous marathon runs; pausing is the version people actually survive. What does not work is deciding at 4am, on stream, while tired.
The same goes for short breaks. A stated "the timer keeps running during breaks, I take one every four hours" is a rule. An unstated pause that chat notices is a controversy.
How it ends
Every subathon ends with the timer at zero, and the last hour is the part people remember. Plan something for it: a finale, a reveal, a thank-you segment, anything that turns "the clock ran out" into the intended ending. Announce the format early, so the final hours feel like a countdown to something rather than the stream simply stopping.
And leave the final total up afterwards. It is the single best piece of evidence for the next one.
A rule set you can copy
- Starting time: 6 hours.
- Subs add 2 minutes; gifted subs count individually.
- Bits and donations convert at a published, consistent rate.
- Hard cap: the timer never shows more than 48 hours remaining.
- The timer pauses between 2am and 10am, announced up front.
- Final hour is a planned finale, not whatever happens to be on screen.
Then put the timer somewhere it is always visible. A subathon timer that viewers have to ask about is a subathon timer that is not doing its job.