Solo Odds

Quantitative risk modeling for solo mining decisions (BTC / BCH).

Most mining calculators focus on expected revenue.

Solo mining outcomes are probabilistic. Expected value alone is insufficient.

Solo Odds models variance first.

It answers questions like:

• What is the probability I mine zero blocks?
• What is the probability I mine at least one block?
• What is the probability I lose money?
• What is the probability solo underperforms pool mining?
• How sensitive is risk to price and electricity assumptions?

What This Is

Solo Odds is a deterministic variance modeling engine built around:

• Poisson block arrival modeling
• Snapshot-frozen network state
• Deterministic shareable links
• Explicit risk metrics (P(0), P(loss), regret probability)
• Sensitivity heatmaps

This is not an ROI marketing calculator.

It is a distribution modeling tool.

Project Structure

The project has two primary surfaces:

1. Web App (Primary)

Deployed example: https://solo-odds.hefftools.dev

Features:

• Solo variance report (tokenized)
• Solo vs Pool risk comparison
• Deterministic share links (snapshot-frozen)
• Probability histograms
• Risk heatmap: P(net < 0) vs price and electricity
• Deterministic Monte Carlo (seeded by token)

Key endpoints:

• POST /api/v1/compare — solo vs pool risk comparison
• POST /api/v1/compare/heatmap — deterministic sensitivity grid
• POST /api/compare/share — generate snapshot-frozen share link
• GET /api/compare/{token} — retrieve frozen comparison
• GET /api/report/{token} — retrieve frozen solo variance report

Share links freeze:

• All user inputs
• Full network snapshot
• Deterministic random seed

This guarantees reproducible shared results even as network conditions change.

2. CLI

The CLI focuses on analytic probability outputs and research workflows.

Install (editable):

pip install -e .

Refresh network snapshot:

solo-odds refresh --coin bch

Compute analytic Poisson odds:

solo-odds odds \
  --coin bch \
  --hashrate 9.4TH \
  --days 365 \
  --json

Model network growth (drift):

Step growth:

solo-odds odds \
  --coin bch \
  --hashrate 9.4TH \
  --days 365 \
  --drift step \
  --step-pct 2 \
  --step-days 14

Linear growth:

solo-odds odds \
  --coin btc \
  --hashrate 50TH \
  --days 365 \
  --drift linear \
  --daily-pct 0.15

Monte Carlo (optional):

solo-odds odds \
  --coin bch \
  --hashrate 9.4TH \
  --days 365 \
  --mc 20000

Generate probability curves:

solo-odds curve \
  --coin bch \
  --hashrate 9.4TH \
  --days 365 \
  --interval-days 7 \
  --json

Core Modeling Concepts

Poisson Block Process

Block arrivals are modeled as a Poisson process with intensity:

μ = integrated block rate over the selected horizon

Outputs include:

• Expected blocks
• P(0 blocks)
• P(at least one block)
• Time-to-first-block distribution (p10 / p50 / p90 / mean)

Drift Modeling

Network growth can be modeled deterministically:

• flat
• step (% every N days)
• linear (% daily)

Drift modifies integrated intensity across segments.

Solo vs Pool (Web)

The compare model computes:

• Solo net distribution (Poisson block count × block reward − electricity)
• Pool net expected value (deterministic in v1)
• Regret probability: P(solo net < pool net)
• Probability of negative net
• Sensitivity heatmap across price and electricity

Pool variance is not modeled in v1 (treated as deterministic EV).

Example Analytic Output

{
  "analytic": {
    "expected_blocks": 0.42,
    "probability_at_least_one": 0.344,
    "probability_zero_blocks": 0.656,
    "time_to_first_block_days": {
      "p10": 12.3,
      "p50": 110.4,
      "p90": 410.7,
      "mean": 238.1
    }
  }
}

Data Sources

Snapshots are fetched from public network APIs and cached locally:

• BTC: Blockchain.com Query API
• BCH: Blockchair stats API with FullStack.cash fallback

Snapshots are stored at:

data/<coin>/latest.json

Each snapshot includes:

• network_hashrate
• difficulty
• blocks_per_day
• block_reward
• source
• source_url

Snapshots are frozen into web share tokens.

Assumptions

• Blocks follow a Poisson process.
• Snapshot inputs are treated as fixed during a run.
• Drift is deterministic.
• Hardware downtime and reliability are not modeled.
• Electricity is modeled linearly from watts × $/kWh.
• Pool modeled as deterministic expected value (v1).
• Fees are not dynamically modeled beyond static pool_fee_pct.

This tool models probability distributions, not guaranteed financial outcomes.

Not financial advice.

Intended Audience

• Serious hobby miners
• Small operators evaluating solo vs pool
• Engineers interested in variance modeling
• Data-driven decision makers

Not optimized for:

• GPU coin hopping
• Yield-maximization dashboards
• Short-term speculative mining
• Turnkey revenue calculators

Non-Goals

• Mining pool operation
• Fleet management
• Hardware optimization
• Tax reporting
• Investment advice

License

MIT