Product & Data.

FOCSAL is a market intelligence platform for crypto traders and investors. It aggregates live order flow, options analytics, smart-money positioning, and macro context from seven exchanges (Binance, Bybit, OKX, MEXC, Coinbase, Hyperliquid, Deribit) plus CFTC and FRED, then runs the data through a five-engine deterministic ensemble to produce directional bias verdicts with confidence scores.

FOCSAL isnot a trading bot. We don't place orders. We deliver analytics; you make the call.

No. FOCSAL is an analytics platform. We provide data, structure, and interpretation. Nothing in the product, the AI Analyst, or our marketing constitutes investment advice, a solicitation, or a recommendation to buy or sell any asset.

Trading crypto carries substantial risk of loss. You are solely responsible for your decisions and outcomes. If you're unsure, consult a licensed financial advisor in your jurisdiction.

Active crypto traders — intraday, swing, options — who want the kind of informational edge usually reserved for prop shops and institutional desks: depth heatmaps, gamma exposure, CFTC institutional flow, regime detection, scenario verdicts.

We assume you're comfortable with options Greeks, futures basis, and orderflow basics. If you're still learning to read candles, FOCSAL will overwhelm you — start with the basics first, then come back.

No. FOCSAL is purely an analytics and intelligence platform. We don't custody assets, we don't execute trades, we don't have access to your exchange balances. Your funds remain on your exchange of choice, in your own control.

FOCSAL never asks for exchange API keys with trading permissions.

Three principles:

1.Deterministic over magical.All core analytics use classical algorithmic logic — same input, same output. We use AI only for interpretation, never for generating signals.

2.Interpretation over data.Raw metrics are commodity. Context ("is this funding rate unusual?") and synthesis ("what does this combination of signals mean?") are the product.

3.The trader decides.We never write "I recommend" or "you should." We surface evidence, weight it, and explain it. The pull of the trigger is yours.

Those platforms show raw metrics — funding rates, open interest, liquidations — and leave interpretation to you.

FOCSAL adds three layers on top: -Context enrichment— every metric comes with percentile, z-score, regime tag, and historical analog ("funding is at 92nd percentile of last 30 days; last 4 times this happened, mean reversion within 8h"). -Ensemble verdicts— five engines vote on direction with weighted confidence. -Smart Money & macro overlay— CFTC TFF positioning and macro regime (DXY, liquidity) aren't side data, they're inputs to the verdict.

We also focus ondeterministic, reproducible logic— no ML black boxes for the core engines.

TradingView is a charting platform — you draw, you analyze, you decide what's happening.

FOCSAL is an interpretation platform — we tell you what's happening across order flow, options, smart money, and macro, with confidence scores and explicit reasoning. The charts are a small piece; the engines are the product.

Think of FOCSAL as what you'd build if you put a quant team, an options desk, and an orderflow analyst in one room and asked them to brief you every hour.

It's an acronym:Futures ·Options ·Crypto ·Strategy ·Analytics ·Liquidity. Six pillars, one operating system.

Yes.7 days of Pro tier access, no credit card required.

During trial you get: - 200 assets across 6 venues - Full Market Intelligence (5-engine ensemble, Futures + Options reports) - AI Analyst (10 queries/day) - Strategy Lab, Journal, Telegram alerts - API Layer 1 access

At the end of the trial your account drops to Free unless you choose to subscribe. You won't be charged automatically because no card was ever required.

One trial per email address.

Go toapp.focsal.com/signup, enter your email and a password, and accept the required consents (Terms of Service, Privacy Policy). You'll receive a verification email — click the link to activate the account.

After verification, you land on a 5-step onboarding wizard (experience level, trading style, preferred assets, alert preferences, language/timezone), then your dashboard.

Signup is free. You can stay on the Free tier indefinitely or start a 7-day Pro trial without entering a credit card.

Three steps for the fastest "aha moment":

1.Open Pulse.This is the morning-check dashboard — funding, liquidations, macro, volatility regime, smart money — all in one screen. Spend 60 seconds reading it.

2.Read the latest Market Intelligence verdict for BTC.This is the 5-engine ensemble's current take. The rationale section explains *why* each engine voted the way it did.

3.Ask the AI Analyst "Explain this report."It walks you through the verdict in plain language. Free users get a preview; Pro gets 10 queries/day.

If you want depth: spend 30 minutes in the Compendium first.

No. FOCSAL pulls only public market data (orderbook, trades, options chains, funding rates). Exchange accounts aren't required.

You don't connect any exchange API keys to FOCSAL — we never ask for them, and the platform never executes trades on your behalf.

Two reasons:

1.Lower friction.We want you to evaluate the product, not commit before you've seen it. Most platforms require cards to make cancellation painful — we don't think that's a fair trade.

2.Cleaner intent.When the trial ends, the question "do I subscribe?" is asked deliberately, not by default. We'd rather have a smaller, intentional customer base than a churning one.

The consequence: trials are one per email address, and we can't extend them automatically. If you genuinely need more time, emailsupport@focsal.com.

No. The platform is a web app with a visual interface — point and click, no code.

Programming skills are only useful if you want to use theAPI(Pro and Enterprise plans). API access lets you pull data into your own scripts, backtests, or trading bots. The OpenAPI documentation includes Python and JavaScript examples.

Roughly ten minutes if you're familiar with options and orderflow terminology. The 5-step wizard takes 2-3 minutes; getting comfortable with the dashboard layout takes a few more.

We ship a built-inCompendiumwith plain-language definitions for every term we use — gamma flip, max pain, basis trade, BOS/MSS, CFTC TFF, premium/discount. Open it side-by-side with the platform when you encounter something unfamiliar.

Hardware: 8 GB RAM, dual-core CPU, >10 Mbps connection. Heavy modules (Liquidity Map, Footprint Canvas) use Canvas API + WebGL — older GPUs may drop frames.

Browsers: Chrome 120+, Firefox 120+, Safari 17+, Edge 120+.

For mobile: the platform works on tablets and large phones, but the UI isn't tuned for under 1024px viewports yet. A dedicated mobile experience is on the roadmap.

The Co-Pilot interprets FOCSAL's reports in natural language. Three modes:

• Explain— "What does this Market Intelligence report mean?" Walks through the verdict, the engine votes, and the key context drivers.
• Quality— "Is this report internally consistent?" Flags when engines disagree, when a verdict relies on a single signal, when confidence is fragile.
• Insights— "What's worth noticing here?" Highlights anomalies — unusual percentiles, regime transitions, contradictions vs. recent history.

The Co-Pilot doesn't generate signals.It interprets the deterministic engines' outputs. The hard rule: it never uses phrases like "I recommend" or "you should." The decision always sits with you.

Limit: 10 queries/day per user (resets at midnight UTC).

Eleven core modules grouped by purpose:

Read the Market— Pulse, Charts, Orderflow & Footprint, Liquidity Map, Options Map.

Research & Scenarios— Market Intelligence (5-engine ensemble), Smart Money (CFTC TFF), AI Analyst.

Plan & Execute— Strategy Lab.

Track & Learn— Journal, Alerts.

Plus aCompendiumencyclopedia of terms accessible from any module.

Smart Money tracksregulated institutional positioningon CME crypto futures via CFTC's Traders in Financial Futures (TFF) report.

Five categories of traders: -Asset Managers— pension funds, mutual funds, insurance companies (long-term capital) -Leveraged Funds— hedge funds, CTAs (speculative) -Dealers/Intermediaries— sell-side market makers -Other Reportables— large non-institutional positions -Non-Reportables— small traders (everyone below reporting threshold)

For each, you see: net position, weekly change, percentile rank vs 1Y and 3Y history, and contextual interpretation.

Covers BTC, ETH, SOL (the assets with CME futures and TFF reporting). Updates weekly, Friday after the CFTC release (~3 days after the Tuesday cutoff).

Spot + perpetuals: Binance, Bybit, OKX, MEXC, Coinbase, Hyperliquid.

Options: Deribit (effectively the only deep crypto options venue).

CME (via CFTC): BTC, ETH, SOL futures for Smart Money positioning.

Macro (via FRED): DXY (dollar index), walcl (Fed balance sheet), M2, 10Y yields.

Plan limits: - Free: 1 venue - Plus / Pro / Enterprise: all 6 spot+perp venues + Deribit options

Alerts is your inbox-side digest of Market Intelligence consensus.Tell the system which assets you care about and what "strong" means — FOCSAL watches the MI snapshot every 5 minutes and emails (or Telegram-pings) you only when a high-conviction signal actually appears.

The module operates on the same deterministic MI ensemble that powers the live platform. No additional algorithms — Alerts is thesame verdict, delivered to where you are.

Adaptive bracket-compressed orderbook depth aggregated across five perp exchanges (Binance, Bybit, OKX, MEXC, Hyperliquid).

Bracket compressionmeans: high resolution near the spread (every $1 of price for BTC, say), low resolution in the tail (every $100 further out). This keeps the map readable while preserving the information that matters.

Visualizations: -Depth heatmap— color intensity = aggregate liquidity at a price -Persistent levels— bids/asks that have stayed in the book over time (likely real, not spoofed) -Liquidity clusters— accumulations where many orders sit close together (potential magnets for price) -Smart DOM— depth of book reading, but cleaned and exchange-aggregated

Updates sub-second via WebSocket.

A toolkit for building, testing, and journaling strategies — not for executing them.

Strategy Builder: visual rule construction. Combine engine outputs, levels, and conditions ("when regime = trend_up AND funding < 25th percentile AND gamma flip below price").

Backtest: run the strategy against FOCSAL's historical data with realistic slippage and fee assumptions. Output: P&L curve, win rate, max drawdown, Sharpe, exposure stats.

Journal: log trades you actually take with full context snapshot (regime, dominant timeframe, engine confidence at entry). Retrospection later is grounded in what you saw, not what you remember seeing.

Free tier gets a basic builder. Plus gets backtest. Pro gets AI optimization suggestions. Enterprise gets custom algo development.

Asset coverage depends on plan:

• Free: 1 asset (BTC)
• Plus: 20 assets — the most liquid majors and mid-caps
• Pro: 200 assets — the full coverage list including alts with futures markets
• Enterprise: 200 + custom on request

Smart Money is limited to BTC / ETH / SOL (the assets with CME futures and CFTC TFF reporting). Options analytics is limited to BTC / ETH (the assets with deep Deribit options markets).

Let us know if you need an asset that's not yet covered — we add based on demand.

No. FOCSAL is purely an analytics platform. We don't have trade execution, position management, or order routing.

When you're ready to act on FOCSAL's analysis, you place the trade on your exchange of choice manually (or via your own automated system). Log it in our Journal afterward to capture the context.

This is a deliberate product choice. Execution is a different competence — exchanges and dedicated platforms (TradingView, GTC, custom systems) do it better. We focus on the layer above.

Four conditions must hold simultaneously:

• Directional bias— onlyUPorDOWNfires.RANGEandUNKNOWNnever do (no actionable edge).
• Confidence ≥ 70%— the aggregator's blended score across all five engines must reach the high-conviction floor.
• 4 of 6 timeframes aligned— at least four of15m, 1h, 4h, 1d, 1w, 1momust lean the same direction.
• Cooldown respected— same(asset, TF, bias)won't email twice within the plan window:Plus 4h, Pro 2h, Enterprise 1h.

All four conditions hold → email fires within seconds of the 5-minute MI scan completing.

Two subscription types, both fire on the same MI trigger but format the email differently for the trader's instrument:

• Futures Strong Signal— for spot/perp traders. Multi-engine consensus + trade plan (entry, invalidation, target, R:R). Clean and minimal.
• Options Strong Signal— same trigger + full options drill-down:GEX regime, gamma flip, max pain, 25-delta skew, ATM IV (near/far), realized HV, IV/RV ratio. Use when you trade through options.

You can subscribe toboth— the trigger delivers two separate emails. Each kind has its own cooldown, so one doesn't block the other.

When you log a trade — entry, exit, asset, direction, size — the Journal automatically captures themarket context snapshot at that moment: dominant regime, ensemble verdict and confidence, key levels (gamma flip, premium/discount), Smart Money state, macro tag.

Later, when you review the trade, you see not just "I bought BTC at $X and sold at $Y" but "I was long in a regime tagged trend_up with high confidence, premium half of range, gamma flip below price." That's what makes retrospection actually useful.

Free tier: basic journal. Plus: full analytics (win rate by regime, by timeframe, by setup). Pro: AI-generated insights from your patterns.

Orderflow shows trade-by-trade aggression: which side initiated the trade (taker vs. maker), price, size. Footprint is the per-candle aggregation: within each candlestick, how much volume hit at each price, broken down by buy/sell.

Key readings: -Delta— net aggressive volume (positive = aggressive buyers) -Imbalance— when buy or sell volume at a price is 3×+ the other side -Stacked imbalance— multiple imbalances at adjacent prices, often a strong reversal signal -Absorption— high volume with little price movement (a large player is fading the move)

This is standard orderflow trader's toolkit. FOCSAL aggregates across exchanges and tags exhaustion patterns.

Pulse is the morning-check dashboard. Seven widgets for a 60-second status read:

1.Macro Strip— DXY, walcl (Fed liquidity), 10Y yields 2.Volatility Regime— implied vol level + trend (VIX-style for BTC and ETH) 3.Funding Pulse— funding rates across venues with z-score context 4.Smart Money Mini— latest CFTC TFF deltas 5.Liquidations Pulse— past 24h liquidation map (long vs. short by venue) 6.Top Movers Quality— gainers/losers with momentum quality tags (sustained vs. spike) 7.System Health— green/yellow/red status per data source

It's intentionally dense — for fast scanning, not deep research.

Our charts are tuned for FOCSAL's data, not general charting:

• Gamma profile overlay— see strike-by-strike dealer gamma exposure as you scroll
• Premium/discount zones— the Wyckoff-style range partition that feeds Market Intelligence
• Liquidity cluster highlights— where stops likely sit
• Funding heatmap— color-coded funding rate evolution
• Smart Money position changes— CFTC TFF deltas overlaid on price
• Engine verdict timeline— see when biases flipped historically

If you want standard TA (Fibonacci, Elliott, classical indicators), keep using TradingView — we link out for that. Our charts are for things TradingView can't show.

A built-in encyclopedia of every term, method, and concept FOCSAL uses. Plain-language explanations with examples and links to the relevant modules.

Covers: orderflow primitives (delta, imbalance, absorption), options Greeks and concepts (gamma exposure, GEX, charm, vanna, max pain, basis trade), CFTC reporting (TFF categories, COT report logic), Market Intelligence engines (how each one works internally), Wyckoff/Smart Money concepts (premium/discount, BOS, MSS), and macro indicators (DXY, walcl, M2, real yields).

Free for all tiers. Searchable, cross-linked, and updated as new methods ship.

Overview is the landing dashboard — your one-screen status check.

It shows: current Market Intelligence verdict (bias + confidence) for your selected asset and timeframe, summary of where each of the five engines voted, key levels (gamma flip, premium/discount boundaries, liquidity clusters), latest Smart Money snapshot, and macro regime tag.

Think of it as "if I only had 30 seconds, what do I need to know right now?"

Two different worlds:

• Smart Money (FOCSAL)shows *regulated futures positions* from large institutions (BlackRock, hedge funds, market makers, dealers) on CME. These are funds with mandates, risk limits, and quarterly reporting obligations.

• On-chain whale trackersshow *addresses holding large amounts of crypto*. The whale might be a single rich individual, a hot wallet, a custodian holding many clients' funds, or a smart contract. You can't easily tell who or why.

Institutions rarely hold BTC on-chain — they use ETFs (spot exposure) or futures (synthetic exposure). So on-chain whales and Smart Money are almost orthogonal datasets.

Yes — single-click asset switcher in the top bar. The dashboard, charts, and reports update instantly. Your watchlist (Pro+) lets you star assets for fast access.

Keyboard shortcut:Cmd/Ctrl + Kopens the global command palette where you can type an asset symbol and jump.

1.Structure Engine— price action, swing highs/lows, break-of-structure (BOS), market structure shifts (MSS). Pure Smart Money / Wyckoff style.

2.Indicator Engine— classical TA filtered for reliability: trend strength, momentum exhaustion, volume confirmation. Not a single indicator but a consensus across families.

3.Liquidity Engine— orderbook depth analysis. Where are stops likely? Where's the magnetic level? Is liquidity being absorbed or stacking?

4.Options Engine(BTC/ETH only) — dealer gamma exposure, charm, vanna, max pain. Computes whether dealers are positioned to suppress or amplify volatility, and where price is mechanically attracted.

5.COT Engine(BTC/ETH/SOL only) — CFTC TFF positioning analysis. Are institutions accumulating or distributing? Detects divergences between price and institutional flow.

Each engine outputs anEngineSignal(bias direction + score + confidence + reasoning). The aggregator weights and combines them.

Market Intelligence is FOCSAL's flagship feature: afive-engine deterministic ensemblethat produces directional bias verdicts with confidence scores, per asset, per timeframe.

Outputs: -Bias: UP / DOWN / RANGE -Confidence: 0.0 — 1.0 -Rationale: which engines voted which way and why -Context filters: premium/discount status, parent-timeframe agreement, key level proximity

Updates every 15 minutes on the fastest timeframes, less frequently on higher ones.

Available timeframes: 15m, 1h, 4h, 1d, 1w, 1mo. Cross-timeframe agreement is part of the confidence calculation.

Yes — all five engines are classical algorithmic logic. Same input → same output, every time. No neural networks, no opaque ML models, no "the AI says so."

This is deliberate:

1.Reproducibility: you can replay any historical moment and get the same verdict the system gave at the time.

2.Explainability: every component of the verdict can be traced to a specific rule.

3.Robustness: ML models drift, overfit, and fail silently. Deterministic rules fail loudly when their assumptions break.

The AI Analyst uses Claude Haiku 4.5 for interpretation — but it never generates the underlying verdicts. The math is human-auditable; the language layer is AI-assisted.

Each engine emits a confidence in [0, 1] based on the quality of its own evidence (how clean the signal is, how many independent inputs agree internally).

The aggregator computes:score = Σ (bias_direction × weight × confidence)across the five engines, with weights from the active profile (Default, Futures, Options, or Calc).

If|score| ≥ 0.25, a directional bias is emitted (UP or DOWN). Otherwise RANGE.

Final confidence =|score|normalized to [0, 1], with modifiers: -Parent-child gating: if a higher timeframe disagrees, lower-timeframe confidence is demoted by 40%. -Engine agreement boost: if 4+ engines align, confidence gets a small boost. -Data quality penalty: if any engine is operating on partial data (e.g. exchange downtime), confidence is demoted.

Disagreement is normal and informative. The aggregator weights and combines anyway, but therationale sectionof the report calls it out explicitly: "3 engines vote UP (Structure, Liquidity, Options), 2 vote DOWN (Indicator, COT). Confidence dampened to 0.42."

When disagreement is severe (engines split 2-3 or 3-2 with similar weights), the bias is often RANGE — i.e., the system tells you it doesn't have a strong directional read.

The AI Analyst'sQualitymode is designed for exactly these cases: "explain why these engines disagree and what evidence is in tension."

A confidence-reduction mechanism that prevents over-confidence on lower timeframes when higher timeframes contradict.

Example: 4h ensemble says UP with confidence 0.7. 15m ensemble also says UP. Both confidences stay as computed.

Now: 4h says DOWN. 15m says UP. 15m confidence gets demoted by 40% (multiplied by 0.6), because the higher timeframe — which has more information — disagrees.

The trade you might want to take on 15m bias is mechanically less attractive when 4h disagrees. The system encodes this directly into the number.

A Wyckoff / Smart Money concept. Define the current trading range (e.g., last 4 weeks). The midpoint is equilibrium. Price in the upper half =premium. Lower half =discount.

The trade-implication rule: buying in premium is unfavorable (you're paying a premium to longs); selling in discount is unfavorable. The good entries are buying discount and selling premium.

When Market Intelligence emits UP while price is in premium, a yellow pill is appended: "long-in-premium" — a contextual warning. Not a hard veto, but a reason to size smaller or wait for a pullback to equilibrium.

Yes. The Structure Engine tags the current regime per timeframe:trend_up,trend_down,range_high_vol,range_low_vol,transition.

Regime affects how other engines' outputs are weighted internally. For example: intrend_up, momentum indicators get more weight; inrange_low_vol, mean-reversion logic dominates.

The regime tag is shown on every verdict, and historical regime sequences are available in Strategy Lab for backtesting.

Four built-in weighting profiles that change how the five engines are combined:

• Default— balanced across all five engines. Good general-purpose.
• Futures— emphasizes Liquidity and Structure. Best for spot/perp directional trading.
• Options— emphasizes Options and Structure. Best for options strategies (especially around gamma flips and max pain).
• Calc— equal weight across the five. For when you want to see pure consensus without any bias.

You pick a profile per asset and per timeframe. Changing profile changes the verdict — it's an honest representation that "bias depends on how you look at the world."

Enterprise customers can define custom profiles.

Yes. Every engine has round-trip unit tests in pytest (310+ tests at last count). Live engine performance is published in post-mortem documents after each major version.

A full historical equity-curve backtest with realistic execution costs is on the roadmap for Q4 — currently you can backtest individual strategies in Strategy Lab using the engines' outputs.

Directly from native APIs of the source venues:

• Crypto exchanges(Binance, Bybit, OKX, MEXC, Coinbase, Hyperliquid, Deribit) — public WebSocket and REST endpoints for orderbook, trades, options chains, funding
• CFTC(Socrata API) — TFF reports for Smart Money
• FRED(St. Louis Fed) — macro indicators (DXY, walcl, M2, yields)

Wedon'trely on data aggregators (CoinGecko, CMC, etc.) for live data. We ingest raw streams, aggregate ourselves, and store derived state.

Every numerical metric (funding rate, OI, gamma, etc.) comes with five extra fields:

• percentile_30d— where this value sits in the distribution of the last 30 days
• percentile_90d— same, 90-day window
• z_score— how many standard deviations from 30d mean
• interpretation— qualitative tag (normal / elevated / extreme_high / depressed / extreme_low)
• regime_history— "last N times this metric was at this level, the following happened"

This is what we mean by "FOCSAL doesn't just show metrics, it interprets them." Raw funding = 0.01% is meaningless. Funding at 92nd percentile of last 30 days = signal.

By data type:

• Orderbook depth— sub-second (WebSocket push)
• Trades / Footprint— real-time push
• Funding rates— every venue's native interval (usually hourly or 8-hourly)
• Open interest— venue-native interval (typically 5-minute updates)
• Options chain (Deribit)— 30-second gateway cache
• Gamma exposure snapshot— 60-second rebuild
• Liquidations— real-time stream
• CFTC TFF (Smart Money)— weekly (ingested Friday after CFTC publishes; data is Tuesday cutoff)
• Macro (FRED)— daily
• Market Intelligence verdicts— recomputed every 15 minutes on 15m+ timeframes, less frequently on higher ones

A specific pattern we detect in CFTC TFF for ETH (and increasingly BTC):

Leveraged Funds increase CME ETH shortswhileDealers increase longs, with spot price stable or rising. This is the signature of the classiccash-and-carry arbitrage: funds short the future (selling at a premium), buy spot ETF or actual ETH (long the underlying), and pocket the basis.

The alert is *not* bearish for ETH price. It's a flow pattern — funds aren't expressing direction, they're harvesting basis. Useful context: ETH shorts are massive on TFF, but they're hedged.

We label this explicitly in Smart Money reports so you don't misread fund short positioning as bearish conviction.

FOCSAL hasgraceful degradation: a single source going down doesn't bring the platform down.

• The affected venue gets a yellow/red status pill in System Health and on every module that touches it
• Aggregated reports (Liquidity Map, Market Intelligence) continue with surviving sources
• The engine that depends most on the broken source notes "operating on partial data" and demotes confidence accordingly
• Auto-reconnect with exponential backoff (1s, 2s, 4s, 8s, capped at 30s)
• After reconnect, REST snapshot pulls the missed window

This is encoded into the confidence calculation — you'll see lower confidence when data is degraded, not a false-positive verdict.

From signed dealer gamma exposure per strike:

signed_gamma(strike) = (call_OI × call_gamma) − (put_OI × put_gamma)

We accumulate signed gamma from the lowest strike upward. Thegamma flipis the price level where cumulative exposure crosses zero — below it dealers are short gamma (price moves are amplified), above it they're long gamma (price moves are dampened).

Window: ±20% from spot. Minimum |signed gamma| threshold: 0.01 to filter noise. Updated every 60 seconds from the Deribit options chain.

The gamma flip is one of the most consequential mechanical levels in crypto options — price tends to be attracted to it.

CFTC publishes the TFF report everyFriday around 15:30 ET, reflecting positions held as of theprior Tuesday close.

FOCSAL ingests within 5 minutes of publication. So on Friday evening (or Saturday morning depending on your timezone), the report reflects Tuesday's data — a ~3-day lag baked into the source.

We don't simulate or interpolate between reports. Smart Money state on Wednesday is the previous Friday's report; we update on the Friday release.

Different retention per data type:

• Trades (raw)— 7 days (high volume, mostly used live)
• Aggregated orderbook snapshots— 30 days
• Funding, OI, basis— 1 year+
• Options chains and gamma snapshots— 90 days
• Smart Money (CFTC)— full history since 2017 (BTC) / 2021 (ETH) / 2024 (SOL)
• Engine verdicts— 1 year+
• Macro (FRED)— full available history

Longer history is on the Enterprise roadmap.

Direct from the Deribit options chain. For each expiry, we identify the strike whose delta is closest to +0.25 (call) and −0.25 (put), then compute:

BF_25 = (call_25_iv + put_25_iv) / 2 − atm_iv

A positive BF means the wings are richer than the body (smile shape). Mapped to standard tenors: 1D / 1W / 1M / 2M / 3M / 6M / 9M / 1Y, with linear interpolation for any tenor in between.

The 25-delta risk reversal (RR) is computed similarly:RR_25 = call_25_iv − put_25_iv. Positive RR means calls are bid over puts (bullish skew).

For each potential expiry-day price level: sum the total dollar value of in-the-money options. The price level that minimizes this sum ismax pain— the price at which option buyers (in aggregate) lose the most, and option sellers profit most.

Useful because dealer flows often (not always) pull price toward max pain as expiry approaches, especially in the last 24-48 hours.

FOCSAL computes max pain for every active Deribit expiry. Available in the Options module and as input to the Options Engine.