Guide
Mining Profitability Modeling: Hashrate, Difficulty, and Power Cost
mining profitability • power cost modeling
Mining profitability is the output of a few measurable variables: hashrate, network difficulty, block rewards (including fees), and operating cost—primarily power. If you treat mining as a “black box,” you’ll overreact to price swings and miss the real drivers of margin.
1) Start with expected production. Convert hashrate into expected coins/day using the current network difficulty and block schedule. Use rolling averages rather than single snapshots; difficulty can change, and luck variance affects short windows.
2) Model power as a first-order constraint. Measure watts at the wall, not just what the miner reports. Then compute daily power cost: (watts ÷ 1000) × 24 × $/kWh. If your fleet includes mixed tuning profiles, model each profile separately because efficiency (J/TH) can vary dramatically.
3) Include real-world overhead. Cooling, networking, switchgear losses, facility overhead, and downtime reduce effective output. A practical model applies an uptime factor (e.g., 95–99%) and a small overhead cost line item.
4) Sensitivity analysis beats single-point estimates. Run scenarios for difficulty up, price down, and power cost changes. A good operator knows the break-even power price for each tuning profile and can adjust frequency/voltage accordingly.
Once you can explain your profitability in these terms, you can make disciplined decisions: when to tune for efficiency, when to pause marginal rigs, and how to size growth without overextending.
1) Start with expected production. Convert hashrate into expected coins/day using the current network difficulty and block schedule. Use rolling averages rather than single snapshots; difficulty can change, and luck variance affects short windows.
2) Model power as a first-order constraint. Measure watts at the wall, not just what the miner reports. Then compute daily power cost: (watts ÷ 1000) × 24 × $/kWh. If your fleet includes mixed tuning profiles, model each profile separately because efficiency (J/TH) can vary dramatically.
3) Include real-world overhead. Cooling, networking, switchgear losses, facility overhead, and downtime reduce effective output. A practical model applies an uptime factor (e.g., 95–99%) and a small overhead cost line item.
4) Sensitivity analysis beats single-point estimates. Run scenarios for difficulty up, price down, and power cost changes. A good operator knows the break-even power price for each tuning profile and can adjust frequency/voltage accordingly.
Once you can explain your profitability in these terms, you can make disciplined decisions: when to tune for efficiency, when to pause marginal rigs, and how to size growth without overextending.