Idle·Ledger
Some links here are referral links and pay me if you sign up. It does not change what you pay, and it does not decide the order of the ledger, which is sorted by measured payout. Apps I removed for underperforming stay listed, with the date I removed them.

How every number here is produced

Stating the method is the point. A payout figure without a date, a location and a conversion basis isn't a measurement, it's a claim.

The setup being measured

One host, running continuously, on one residential connection in Ohio. No VPS, no proxy farm, no multi-IP arrangement. That constraint is deliberate. Multi-IP setups produce much larger numbers that nobody reading this can reproduce at home, and reporting them without that caveat is the most common way this category misleads people.

Host Geekom GT1 Mega, 24/7 · line FILL: down/up Mbps
Region Ohio, US · classification residential
Measured draw FILL W at the wall · utility rate FILL ¢/kWh

The six rules

  1. Read, never estimate. Every figure comes off a platform dashboard I log into, on the first of the month. If I didn't read a cell, it's marked unmeasured rather than filled with something plausible.
  2. Spot rate on the recording date. Token payouts convert at the price on the day I record, not the price when they were earned and never at a peak. This means token rows move with the token, and a good month for a token is not the same thing as a good month for network demand. Those two get conflated constantly and it's the main reason crypto-paid apps look better than they are.
  3. Electricity measured at the wall. Plug meter, not the manufacturer's spec sheet, billed at my actual utility rate. Net figures subtract it. A lot of stacks in this category are close enough to breakeven that power is not a rounding error.
  4. Signup bonuses excluded from monthly rows. They're listed separately. A one-time bonus inflates month one and it is the single most common trick used to make a first-month screenshot look impressive.
  5. Removals stay published. When I stop running something, the row stays with the date and the reason. Survivorship bias is what makes every other table in this category useless.
  6. Uncertainty stays visible. Where a platform's reporting is unreliable, or where I can't reconcile a dashboard against what actually landed in a wallet, the page says so instead of smoothing it.

Known limitations

Sample size is one host. Figures are specific to a US residential connection and don't transfer to other regions, because demand for residential IP addresses varies enormously by country and US addresses sit near the top of that range. Token-denominated rows carry price risk independent of anything the network does. And platforms change their rates without announcing it, so a figure from six months ago describes six months ago.

If you cite a number from this site, cite it with its date.