Methodology | Orange Abacus

Every number on Orange Abacus comes from a disclosed data source and a model you can audit from this page alone. No black-box outputs, no proprietary adjustments. Every calculator’s assumptions are reviewed before it ships, and this page is updated every time a calculator’s model, data source, or limitations change. For the legal framing around how these tools may be used, see the Disclaimer.

Where our data comes from

Bitcoin price — live spot price from CoinGecko; the historical daily-close series is a CryptoCompare backfill, extended by one day at a time via CoinGecko since the refresh automation shipped in July 2026.
Network data — block height, difficulty, hashrate, and transaction fees — mempool.space.
USD/EUR conversion — Frankfurter, which republishes the European Central Bank’s daily reference rates.

Every calculator runs entirely in your browser. Inputs and results never leave your device — Orange Abacus has no accounts and performs no server-side calculation.

Bitcoin DCA Calculator

Model. Each scheduled contribution buys BTC at that exact calendar day’s closing price. Today’s portfolio value is priced at the live BTC spot price.
Data source. Historical daily closing prices from CryptoCompare, extended daily via CoinGecko, stored in a static file on the site. Today’s BTC spot price from CoinGecko, fetched live when you load the page. EUR contributions are converted using the Frankfurter USD/EUR rate.
Refresh cadence. Today’s BTC spot price is fetched live when you load the page and cached for 2 minutes. The historical closing-price file is extended with new days ahead of each deploy, not on every page load. The EUR rate refreshes every 24 hours.
What this doesn’t account for. Intraday price swings, exchange trading fees, and taxes. A schedule date beyond our latest stored close reuses that close rather than projecting forward.

Bitcoin Future Prediction DCA Calculator

Model. A deterministic price path from today’s price to your target. The Growth, Sideways, and Decline presets follow defined curve shapes with light, seeded variation, so the same inputs always produce the same output. The Historical option instead takes our most recent stretch of real daily closes and rescales it to start at today’s price and land on your target.
Data source. Today’s BTC spot price from CoinGecko, fetched live when you load the page, used as the starting point for every scenario. The same daily closing-price file used by the DCA calculator supplies the Historical option’s shape. EUR contributions use the same Frankfurter rate.
Refresh cadence. Today’s BTC spot price (the starting point for every scenario) is fetched live when you load the page and cached for 2 minutes. The historical closing-price file used for the Historical scenario shape is refreshed ahead of each deploy.
What this doesn’t account for. This is a what-if tool, not a forecast. It excludes exchange fees and taxes, and no scenario predicts what Bitcoin will actually do.

Bitcoin Loan Calculator

Model. A generic overcollateralized, fixed-rate, interest-only loan: one liquidation LTV threshold, simple (non-compounding) interest, full liquidation of the posted collateral if that threshold is breached. The structure mirrors common terms across BTC-backed lenders such as Ledn, Unchained, and Arch — it is not pulled from any single lender’s live terms.
Data source. BTC spot price from CoinGecko, fetched live when you load the page. APR, LTV, term, fees, and margin-call buffer are the values you enter.
Refresh cadence. BTC spot price is fetched live when you load the page and cached for 2 minutes.
What this doesn’t account for. Variable-rate or rolling-margin loan structures, partial liquidations (the model liquidates the full position at the threshold, not a partial amount), and any lender-specific fee beyond the origination fee field.

Bitcoin Mining Profitability Calculator

Model. Daily revenue is the protocol-exact probability of finding blocks at your hashrate against current network difficulty, multiplied by the block reward (subsidy plus average fee) and net of your pool fee, then converted to dollars at the current BTC price. Daily cost is your machine’s power draw times 24 hours times your electricity rate. The monthly and yearly columns extend the daily figure holding price and difficulty constant.
Data source. Network difficulty and hashrate, and average fee per block, from mempool.space. BTC price from CoinGecko, fetched live when you load the page. ASIC hashrate and power-draw specs are a hand-maintained table of manufacturer spec-sheet values, updated when new hardware ships.
Refresh cadence. Live data is fetched once per page load and cached for 2 minutes. If both the live fetch and the cache fail, the calculator falls back to a frozen snapshot taken June 10, 2026.
What this doesn’t account for. Difficulty and price changes within the monthly/yearly projection window, plus cooling, hosting, maintenance, and resale value.

Bitcoin Halving Countdown

Model. Blocks remaining to block 1,050,000, multiplied by the network’s live average block interval, projects a wall-clock date. The countdown re-syncs that estimate while the page is open.
Data source. Current block height and average block interval from mempool.space.
Refresh cadence. Re-fetched every 2 minutes while the page is open; the on-screen timer ticks every second between syncs. Falls back to a frozen snapshot taken June 8, 2026 if both the live fetch and the cache fail.
What this doesn’t account for. The average block interval can shift after every sync, so the projected date is always an estimate, never a fixed deadline.

Bitcoin Sats Converter

Model. The sat/BTC ratio — 100,000,000 satoshis per bitcoin — is a fixed protocol constant. USD values multiply the entered amount by the live BTC spot price.
Data source. BTC spot price from CoinGecko, fetched live when you load the page.
Refresh cadence. Fetched on load, cached for 2 minutes for any subsequent load within that window.
What this doesn’t account for. Bid/ask spread or any specific exchange’s fees — the figure shown is a reference spot price, not a tradable quote.

Bitcoin 4-Year Cycle Chart

Model. Cycle windows are fixed boundaries derived from the bundled daily closing data, not a live-fitted model. Each expansion runs from a halving to that epoch’s highest close within the following 18 months; each drawdown runs from that peak to the lowest close that followed (the first cycle’s bottom excludes the February 2014 Mt. Gox flash print). The current drawdown’s end date is a projection — the October 2025 peak plus 12.6 months, the mean drawdown length of the 3 completed cycles — and shifts only if Fabio revisits the boundaries once the current cycle resolves. The chart itself plots the full daily-close history on a logarithmic scale so every epoch’s percentage move stays comparably readable, and marks the “you are here” point on the latest close. The next halving date shown on the chart is projected the same way as on the Halving Countdown: blocks remaining to block 1,050,000, times the network’s live average block interval.
Data source. Historical daily closing prices from CryptoCompare and CoinGecko, stored in the same static bundle used by the DCA calculators. Today’s BTC spot price from CoinGecko, fetched live when you load the page. Current block height and average block interval, for the projected next halving, from mempool.space.
Refresh cadence. The historical closing-price bundle is extended ahead of each deploy, not on every page load. Live BTC spot price and network data are fetched on load and cached for 2 minutes; the halving projection falls back to a frozen snapshot if both the live fetch and the cache fail, same as the Halving Countdown.
What this doesn’t account for. The cycle windows are hardcoded, not recalculated live — they only change when Fabio updates them after a cycle resolves. Four completed halvings is a small sample, and the pattern is a market observation read onto the halving schedule, not a rule the protocol enforces.

Bitcoin Market Cap Calculator

Model. Market cap is Bitcoin price multiplied by circulating supply: marketCap = btcPrice × circulatingSupply. Circulating supply is computed from the live block height using the protocol’s actual issuance schedule, not assumed to be a flat 21 million, so it climbs in step with real mining. The second mode runs the same equation in reverse, dividing a target market cap (typed in or picked from the asset list) by circulating supply to return an implied price per coin. Each result also runs through a comparison engine that scores every reference asset for how “readable” it is against that value (closest to a whole or clean fraction, not too far outside its range) and surfaces up to three: one major asset (metal, company, equity, or real estate), one national economy, and one money-supply or per-capita figure.
Data source. Live BTC spot price from CoinGecko, fetched when you load the page. Circulating supply from the live block height via mempool.space. Gold’s comparison value is recomputed live from the same CoinGecko quote (BTC priced in troy ounces of gold) applied to the world’s above-ground gold stock. Seven mega-cap companies and the S&P 500/global-equities proxies pull live market caps from Finnhub. Every other comparison — national GDP, the remaining company and equity-index valuations, real estate, money supply, and per-capita population — is a curated, dated figure from public sources (IMF, World Gold Council, Federal Reserve, company filings, UN), stored in a static reference file.
Refresh cadence. BTC and gold prices are fetched on load and cached for 2 minutes, same as the site’s other live-price calculators. The Finnhub-sourced company and index figures refresh weekly via an automated job. The remaining curated comparison figures are updated manually, and each one carries the date it was last checked.
What this doesn’t account for. The curated comparisons are approximations for scale, not to-the-dollar figures, and move out of date between refreshes. National GDP and money-supply figures are annual flows, not market values, so lining them up against a market-cap snapshot compares two different kinds of measurement. Any hypothetical price or market cap shown is an illustration, not a forecast.

Bitcoin Cost Basis Calculator

Model. A simple weighted average: total fiat invested across every entered lot, divided by total BTC accumulated, gives your average cost basis. Unrealized gain or loss compares your BTC holdings valued at the live price against that total invested.
Data source. Today’s BTC spot price from CoinGecko, fetched live when you load the page. The same historical daily-close file used by the DCA calculator supplies the counterpart — fiat or BTC — for any lot where you only enter one side. EUR values use the Frankfurter USD/EUR rate.
Refresh cadence. BTC spot price is fetched live when you load the page and cached for 2 minutes. The historical closing-price file is extended ahead of each deploy. The EUR rate refreshes every 24 hours.
What this doesn’t account for. FIFO, LIFO, HIFO, or any other tax-lot accounting method — this is a single weighted average across all entered lots, not a jurisdiction-specific tax calculation.

If I Bought Bitcoin Calculator

Model. Compares the historical daily closing price on the date you enter against today’s live BTC price, and reports the BTC bought, the gain or loss, and the return multiple.
Data source. The same historical daily-close file used by the DCA and Cost Basis calculators, spanning July 17, 2010 to today. Today’s BTC spot price from CoinGecko, fetched live when you load the page.
Refresh cadence. BTC spot price is fetched live when you load the page and cached for 2 minutes. The historical closing-price file is extended ahead of each deploy.
What this doesn’t account for. Exchange trading fees, taxes, or inflation on the original amount — the result is a before-costs snapshot, not a net return.

Bitcoin Difficulty Tracker

Model. Current difficulty and block height are read from the network’s latest block. Epoch progress and the projected change at the next retarget come from the same difficulty-adjustment math the protocol runs, projected forward from the blocks mined so far in the current epoch. The chart plots the network’s full retarget history, toggled between 1-year, 2-year (default), 5-year, and all-time windows.
Data source. Current block height, current difficulty, and per-epoch retarget metrics, all from mempool.space.
Refresh cadence. Re-fetched every 2 minutes while the page is open. Falls back to a frozen snapshot taken July 18, 2026 if both the live fetch and the cache fail.
What this doesn’t account for. The projected next retarget is an estimate that moves as the epoch’s average block interval changes, and only locks in once the epoch’s final block confirms.

Bitcoin Uptime Tracker

Model. The genesis clock counts elapsed time from Bitcoin’s fixed genesis timestamp, January 3, 2009, 18:15:05 UTC, which never changes. The uptime percentage divides that elapsed time, minus the combined duration of Bitcoin’s two documented consensus incidents, by the same elapsed time. Those incidents are the August 2010 value overflow (about 8 hours 27 minutes) and the March 2013 chain fork (about 6 hours 55 minutes), the only periods the network failed to converge on a single valid chain. The live network status reads the current tip block for time since the last block, the current average block interval, and the block height.
Data source. The genesis timestamp and the two incident durations are fixed historical constants, taken from block-explorer and CVE records (CVE-2010-5139 and CVE-2013-3220). Current tip-block height, timestamp, and average block interval from mempool.space.
Refresh cadence. The genesis clock ticks every second in the browser and never resets. Live network status is re-fetched every 2 minutes, and falls back to a frozen snapshot taken July 18, 2026 if both the live fetch and the cache fail. The uptime timeline and percentage are deterministic and recomputed on load.
What this doesn’t account for. Uptime here means base-layer consensus continuity, not the availability of any wallet, exchange, or individual node. Normal variance in block timing is not downtime. Only the two documented consensus incidents count against the total, and because that fixed downtime shrinks as a share of an ever-growing total, the figure holds at approximately 99.99%.

When data can’t be reached

Every live data call follows the same order: a fresh fetch first, then a cached value up to 2 minutes old (24 hours for the historical price file), then a frozen reference snapshot, dated and labeled as such on the few pages that carry one. The site never substitutes a guess for a number it can’t source.

Keeping this page current

This page is updated every time a calculator’s model, data source, or known limitation changes, and gets a new section once a new calculator has shipped, been approved, and is running in production. For the legal framing around how these tools may be used, see the Disclaimer.