Methodology
How TAO Analytics helps you navigate Bittensor subnet investing
Every investment decision follows five steps. Our analytics are designed around this loop.
Steps 1-3 of the Investment Loop
"Is the Bittensor ecosystem healthy and growing?"
Combines ecosystem breadth, TAO trend context, capital distribution, and emissions activity. Used as macro regime context only.
"Can you trade safely?"
Measures liquidity depth and price stability. Used as a filter to identify subnets where you can enter and exit without excessive slippage. Not a return predictor.
"Is the price rising?"
Measures price momentum using moving average analysis. Higher scores indicate stronger upward trends. Used for selection in momentum-focused strategies.
"Is capital flowing in?"
Measures capital flow changes over time. In AMM pools, flow and price are linked, but flow gives a distinct lens for weighting and risk control.
"Is buy-side demand supporting miners?"
Subnet tokens are OpEx-replacement assets. Miners have real costs and are natural sellers. We measure whether buy-side demand is absorbing this sell pressure.
Step 4: Systematic allocation strategies
Each portfolio targets a specific investor objective. All use our tradability filter (SHI) to ensure positions can be entered and exited. Momentum-focused portfolios use SMI for selection. Balanced portfolios use STI for weighting (safer allocation, less crash risk than pure price momentum).
Alpha — 10 positions, weekly
Objective: Maximum returns through concentrated momentum bets. Top 10 subnets by composite SMI+STI score, weighted by momentum. 0-15% TAO buffer. Highest potential return and risk.
Growth — 20 positions, weekly
Objective: Strong returns with better diversification than Alpha. Top 20 subnets by momentum (SMI), weighted by SMI with 8% cap. 0-20% TAO buffer. Weekly rebalancing responds to trends.
Balanced — 25 positions, bi-weekly
Objective: Steady growth using capital flow signals. Top 25 eligible subnets, STI tilt weighting (capital flows, not momentum). 15-35% TAO buffer. Different signal from Growth for true diversification.
Defensive — 35 positions, monthly
Objective: Capital preservation with upside participation. Top 35 eligible subnets, equal weight, no signal tilt. 40-70% TAO buffer absorbs volatility. Monthly rebalancing minimizes turnover.
Dynamic TAO Buffer: TAO allocation adapts to concentration risk (pool-HHI) and execution conditions.
Risk Integration: Portfolios automatically exclude subnets at DANGER level during rebalancing (see Risk Monitoring below).
Passive baselines for comparison
Benchmarks provide neutral market baselines without active signals. Compare portfolio performance against these to understand alpha generation.
TAO Only
100% TAO, no subnet exposure. Answers: "Did subnet investing add value vs. just holding TAO?"
Investable Equal-Weight
Subnets with ≥1,000 TAO liquidity and 14-day seasoning, equal weighted, monthly rebalance. Answers: "How is the investable universe doing?"
Total Market Equal-Weight
All subnets with ≥100 TAO liquidity and 7-day seasoning, equal weighted, monthly rebalance. Answers: "How is the broad market doing?"
Step 5: Ongoing surveillance
After you invest, conditions change. Our Risk Sentinel monitors for deterioration using Bittensor-native signals and alerts you before problems escalate.
Risk Signals
Sudden drops or reversals in emission flow can signal underlying issues
High stake concentration in few validators increases systemic risk
Low reserves increase slippage and make exits difficult
Multiple signals together indicate more serious concerns
Risk Levels
No active concerns
Monitor closely
Review position
Consider exit
Primary: Our Subtensor Node
- • Subtensor node on dedicated infrastructure
- • Block-level timestamps for provenance
- • 15-minute price and pool snapshots
- • Emission and stake distribution data
Secondary: External Sources
- • Historical backfill for early periods
- • TAO/USD pricing (CoinGecko, Taostats fallback)
Subnet identity (names, logos, links) is pulled directly from on-chain SubnetIdentity data.
- • Simulated performance: Historical results before live inception are based on backtesting. Simulated performance has inherent limitations.
- • No guarantee: Past performance does not predict future results. All indicators are probabilistic, not deterministic.
- • Execution reality: Model portfolio returns are theoretical. Actual execution depends on liquidity and market conditions.
- • Data limitations: Some subnets may have incomplete data during early periods.
- • Not investment advice: This is analytical infrastructure for research purposes. We do not provide personalized investment recommendations.
All methodologies are versioned for reproducibility. Version changes require validation before deployment.
Indicators
Portfolios
Benchmarks
Risk Systems
Indicators updated every 15 minutes. Portfolios and benchmarks calculated daily at 00:00 UTC.
Questions about our methodology? Contact us for institutional factsheets with full specifications.