GoMining: 4 Years of Real Payouts vs. Model

One real account, 1,183 days of actual payouts, retrieved from the account directly. Everything below is real data — and the model on this site had to predict it, day by day. The gaps are real gaps. The result: 798 of 1,183 days within ±2% of pure physics — through the halving, the bull market and the bear.

What this proves

The model reproduced the payout history of this reference account closely (798/1,183 days within ±2%) across multiple market conditions — bear market, halving, bull run. In plain English: on most measured days, the model came within 2% of what the account actually produced. The formulas and fee handling are sound, out-of-sample, over years. Once the farm stopped growing (January 2025 onward) the daily fit sharpens further: 95% of 581 stable-era days within ±2%, 98.5% within ±5% — details on the Validation Log.

What this does NOT prove

It does not guarantee that a new investment will produce the same future returns — difficulty and platform fees can move, the farm grew over the years, and a future exit price is an assumption (resale lever), not a payment.

The whole ride, week by week

Each bar is one week of the real account: gold = actual gross production, hairline = what the physics model predicted from the previous week's farm size alone (an out-of-sample forecast — the model must hit today using only yesterday's knowledge). The halving and the big market phases are marked. Gaps in the data are shown as gaps — never interpolated.

Day by day — the full four years at daily resolution

The same data, not averaged: every single payout day from October 2022 to August 2026. The 219 days without payouts (10 gaps, largest: May–Sep 2023 while the account migrated miners) stay visibly empty — the x-axis is real calendar time, so gaps compress nothing. The model line is the physics-only prediction, run day by day.

Cumulative — what four years actually added up to

Same data, accumulated: bitcoin earned (top) and its pre-fee dollar value (bottom). The yellow area between model and real lines is how far pure physics was off over the whole run: 0.8% on total volume.

The market phases — the model's report card per era

What this page proves — and what it cannot

Proven: this is one real account, payouts retrieved directly, day by day, over the longest span any public page shows for tokenized mining — through a bear market, the April 2024 halving (reward 6.25 → 3.125 BTC per block), a bull run and the following correction. The model on this site was built from documented rates only; the day-by-day agreement on this page is the test whether those rates describe reality, in every market weather.

Not proven by this page: that your account would match. The farm size grew over the years (visible in the chart), the account holds discounts a newcomer doesn't have yet, and bitcoin's price path was what it was — none of that is a promise. What carries over is the method: the formulas, the fee handling, the difficulty handling — and the fact that they held up for four years of real money.

The resale market — the exit lever, measured

Data honesty

The record has gaps — platform outages and app changes — shown as empty stretches, never interpolated: the largest is about four months in mid-2023. Days missing from the source are excluded from every statistic rather than assumed. Rewards before the halving were earned at 6.25 BTC per block, after at 3.125 — visible as the step in the gold line. Farm size is reconstructed from each day's payout and that day's network difficulty (not entered by hand); its drift upward is the reinvestment the account actually did. Full method: the Method page; the recent months live in the daily feed.

Four years, real — one account, every day, checked