Visualizer Guide

A tour of every section of the RiskState Visualizer — the free, read-only view of the same risk engine that powers the /v1/risk-state API.

The visualizer does not execute trades, hold funds, or require a wallet connection. It is a transparency layer: inspect the inputs, see the caps and regimes that come out, and understand exactly why the engine permits (or restricts) exposure.

For the formal description of what each number means and how it is computed, see Methodology.

Layout overview

The visualizer is organized in stacked sections, roughly in priority order of what matters most for a pre-trade check:

1. Ticker bar — headline metrics at a glance
2. S1 row — three side-by-side panels: Market State · Risk Permissions · Rules Engine Alerts
3. Key Indicators — individual signals with normalized scores
4. Composite Signal — the single weighted tactical number
5. Cycle Context — where we are in the broader structural cycle
6. Macro Risk — traditional-finance context
7. Institutional Flow — corporate treasury tracking
8. ETF Flows — spot ETF flows
9. Price Chart — interactive candlestick chart
10. Funding & Open Interest — derivatives positioning intelligence

Asset (BTC / ETH) and market-type (Spot / Perps) selectors live in the header and adapt the view globally.

Ticker bar

The horizontal bar at the top shows the most important market metrics in real time: total crypto market cap, active-asset price (BTC or ETH), BTC dominance, Fear & Greed index, 24h liquidations, and the U.S. Dollar Index (DXY).

S1 row: Market State · Risk Permissions · Rules Engine Alerts

The S1 row is three equal-width panels that together compress the full engine state into one horizontal strip. Read left-to-right: the market, then what it permits, then what to watch.

Market State (2×4 grid)

Eight boxes classifying the current environment from eight complementary angles.

Each box is a visual summary. When they agree, the engine has confluence. When they disagree, the policy engine's caps and penalties kick in.

Risk Permissions

This is the operational output — the binding numbers and actions that govern a pre-trade decision. The card adapts to the active market type:

Spot mode shows:

• Max Position Size — % of portfolio allowed as a position
• Leverage: N/A — Spot
• Direction bias: LONG_PREFERRED · NEUTRAL · SHORT_PREFERRED
• Allowed actions (chips) and Blocked actions (chips). Leverage-only actions are hidden.
• Policy level with 5 colored dots (1 = BLOCK Survival → 5 = GREEN Expansion)
• Caps row — the 4 underlying caps and the quality / volatility multipliers
• Binding constraint — which cap is limiting (RULES / DEFI / MACRO / CYCLE)

Perps mode adds:

• Label changes to Max Notional Exposure
• Margin Guide grid at 3x / 5x / 10x / 25x reference leverage levels, with implied margin as % of portfolio
• Funding Carry badge showing current funding rate + 30-day percentile, with contextual warning when carry cost is extreme

For the formal mapping from max_size_fraction to policy level, see Methodology → Policy engine.

Rules Engine Alerts

A list of deterministic alerts that fire automatically when specific conditions are met — pure if-then logic, no AI. Rules are sorted by severity:

• CRITICAL — extreme conditions requiring immediate attention
• DANGER — significant risk signals
• WARNING — caution-level signals worth monitoring
• OPPORTUNITY — favorable conditions detected
• INFO — contextual information

Rules cover macro conditions (yields, equity correlation), positioning extremes (funding, basis, OI), on-chain metrics, ETF flow patterns, and trend-vs-composite divergences. The count in the header badge is the number of active alerts.

Key indicators

A grid of individual market signals, each as a small card with current value, a normalized score (0–100) and a directional label (bullish / neutral / bearish). The set of indicators adapts per asset.

BTC mode: RSI · MVRV · NUPL · Futures CVD · Spot CVD · Funding Rate · Long/Short Ratio · Exchange Netflow · Whale Score · Trend Strength.

ETH mode: RSI ETH · ETH/BTC Ratio & Trend · Futures CVD ETH · Funding ETH · Long/Short ETH · Whale Score · Trend Strength. BTC-native metrics (MVRV, NUPL, exchange netflow) are replaced by ETH-native signals.

Indicator color reflects direction: green (bullish), red (bearish), yellow (neutral). These feed into the composite score.

Composite signal

A single 0–100 score aggregating the normalized indicators into a weighted tactical average. A semicircular meter visualises the score with colour zones:

Below the meter, a breakdown bar shows how each subscore contributes. A "disposition" label translates the number into plain language: RISK-OFF / LEAN RISK-OFF / NEUTRAL / LEAN RISK-ON / RISK-ON.

Structural score & tactical score

Alongside the composite, the engine emits two parallel scores that drive the policy combiner (see Methodology → Structural vs tactical):

• Structural score (0–100) — slow-moving fundamentals: Cycle, Supply, Demand, Macro. Label: STRONG / STABILIZING / NEUTRAL / FRAGILE / WEAK.
• Tactical score (0–100) — fast-moving positioning: Positioning Pressure, Momentum, Volume/CVD, Derivatives extremity, L/S velocity, Whale Pressure. Label: STRONG_BULL → STRONG_BEAR.

Direction bias & direction layer

The Risk Permissions card shows a direction_bias (LONG_PREFERRED / NEUTRAL / SHORT_PREFERRED). This is produced by the policy combiner, which reads both scores and applies regime-dependent weighting.

Alongside it, the engine surfaces a direction_layer audit field indicating which layer decided the direction:

When direction_layer differs from tactical, the engine is explicitly warning: the tactical tape and the structural context disagree, and the safer view has won.

Cycle context

Where the market sits in its broader cycle.

BTC mode — classifies current position into one of eight phases (BOTTOM · EARLY · MID · LATE · EUPHORIA · CORRECTION · POST-PEAK · LATE-CORRECTION) using on-chain valuation, time since last halving, drawdown from ATH, and distance from the 200-week moving average. Also shows BTC Supply in Profit and Long-Term Holder supply.

ETH mode — shows ETH/BTC ratio as primary structural metric, ATH drawdown, 200W SMA distance, and an Altseason Score (0–100) gauging whether capital is rotating into alts (dominance shifts, ratio trends, small-cap equity performance).

Macro risk

Traditional-finance context affecting crypto risk appetite:

• U.S. Treasury yields (10Y, 2Y) and the yield curve spread
• S&P 500 and Gold performance (Russell 2000 in ETH mode)
• Real rate (nominal yield − inflation expectations)
• Fed liquidity regime (balance sheet trend)
• Correlation between crypto and equities (BTC/SPX in BTC mode, ETH/NASDAQ in ETH mode)

Classifies the macro environment as RISK-ON / NEUTRAL / RISK-OFF, and shows the coupling level (how tightly crypto tracks equities). High coupling during a RISK-OFF macro regime means equity selloffs are more likely to drag crypto down.

Institutional flow

Tracks publicly reported corporate treasury holdings of the active asset (BTC or ETH). Shows the largest holders, detects buy/sell changes between snapshots, and classifies the overall institutional trend as ACCUMULATING · DISTRIBUTING · MIXED · NO CHANGE.

Data is sourced from public filings, aggregated by CoinGecko. Changes are detected by comparing the current snapshot with the previous one stored server-side.

ETF flows

Daily and weekly net flows into spot Bitcoin or Ethereum ETFs, including the streak metric (consecutive days of inflows or outflows).

Sustained ETF inflows indicate institutional demand; outflows signal distribution. The streak helps distinguish one-off events from persistent trends.

Price chart

An interactive candlestick chart (Lightweight Charts) with selectable timeframes (1H, 4H, 1D) and asset toggle (BTC/ETH). Includes EMA overlays (21, 50, 200) and trade level lines when positions are active. The chart updates independently of the asset selector, so you can inspect ETH price action while the rest of the dashboard stays on BTC context.

Funding & Open Interest

A compact positioning intelligence panel showing derivatives data for both BTC and ETH.

• Funding Rate — who pays whom on perpetual swaps, plus a percentile tag vs. the 30-day history
• Basis — premium (or discount) of perpetual price vs. spot. Premium = longs dominant.
• Open Interest — total leveraged positions, with a z-score vs. 30 days
• Long/Short Ratio — positioning crowding with a label

Liquidation Bar — visual split between long and short liquidations in the last 24h.

Squeeze Detector — combines multiple weighted signals (funding, basis, crowding, liquidation imbalance and more) to classify directional squeeze risk as UPSIDE (short squeeze probable), DOWNSIDE (long squeeze probable), or TWO-SIDED (bilateral tension). Includes a confidence percentage and a one-line actionable recommendation.

Selectors

Asset (BTC / ETH)

The first dropdown in the header switches the entire dashboard context. All sections adapt: prices, indicators, composite weighting, cycle classification, macro correlations, ETF data and institutional holdings. ETH mode replaces BTC-specific metrics (MVRV, NUPL, exchange netflow) with ETH-native signals (ETH/BTC ratio, ETH structural score, altseason score).

Market Type (Spot / Perps)

The second dropdown switches only the framing of the Risk Permissions panel. The underlying risk assessment — Market State, Rules Engine Alerts, composite, structural, tactical — is unchanged. The engine evaluates the market independently of how you intend to trade; only the operational presentation adapts:

• Spot — Max Position Size + leverage hidden + leverage-only actions filtered out
• Perps — Max Notional Exposure + Margin Guide (3x/5x/10x/25x) + Funding Carry badge

Auditability: policy hash

Every state the engine produces is anchored by a policy_hash — a SHA-256 hex string over a deterministic serialisation of the policy inputs. The visualizer surfaces this hash so you can verify reproducibility:

1. Note the current policy_hash and reference_time from the Risk Permissions footer.
2. Call POST /v1/risk-state with the same reference_time in the body.
3. Compare the policy_hash returned by the API with the one displayed.

If the hash matches, both views computed the policy from bit-identical inputs. If it differs, one of the inputs (price, macro data, on-chain signal, cap) changed between the two reads.

The hash covers: API version, scoring version, reference timestamp, prices, indicators, positioning, macro, volatility, cycle, composite, structural score, tactical score, policy permissions and binding constraint. See the full field list in Methodology → Hashed fields.

Data sources

The visualizer aggregates public data from:

• Binance — prices, funding, OI, klines
• CoinGlass — CVD, ETF flows, liquidations
• CoinGecko — dominance, treasuries, correlations
• FRED — yields, SPX, Fed balance sheet
• Yahoo Finance — gold, Russell 2000
• DefiLlama — TVL, DEX volume, fees
• Alternative.me — Fear & Greed index

Data refreshes on page load and can be manually refreshed with the REFRESH button. A 120-second server-side cache prevents excessive upstream calls.

Coming soon

The visualizer is free and will remain free. The following features are planned for paid tiers (see pricing):

• AI Analysis (BYOK) — dual-mode AI analysis (tactical + structural) powered by Claude, using your own Anthropic API key.
• DeFi Loan Monitoring — connect a wallet to monitor Spark / Aave positions, health factors and liquidation buffers in real time.
• Liquidation Engine — automated stop-loss evaluation, take-profit limit orders, OCO execution via CoW Protocol.

Disclaimer

The visualizer is an informational tool. It does not provide financial advice, execute trades, or custody assets. Scores and classifications reflect a quantitative model's output, not a human analyst's judgment. Always do your own research. See Legal & Disclaimer for full terms.