Dashboard Guide

Your AION dashboard is a quantitative signal terminal. It runs multiple AI and statistical models against real market data, updated throughout the trading day, to give you a probabilistic read on where a given sector or index is heading.

There are 8 cards on the main dashboard, each showing a different angle on the same question: what's the risk/reward setup right now?

The top dropdown lets you switch between profiles: different sectors (Cyber, Semis, Energy, etc.) and broad indexes (SPX, NDX, High Beta). Each profile runs its own independent set of models trained on the tickers in that universe.

This is the headline signal. A deep learning ensemble (multiple neural networks trained on price, volume, breadth, and volatility features) outputs the probability that the sector/index will be higher than today over three time horizons.

A dedicated AI model trained specifically to detect the conditions that precede significant drawdowns. It asks: what's the probability that this sector drops substantially over the next 20 trading days?

A blended probability model built from five traditional quantitative factors, no machine learning. It uses years of historical data to generate probabilities and expected returns over a 30 day averaged window. Each factor measures a different dimension of market health.

This card counts how many of the 9 total models (AI + statistical) are currently reading bullish vs. bearish. A model counts as bullish if its probability is above 50% and its crash estimate is 15% or lower.

This card puts today's conditions in historical context using 2 years of daily price data across every ticker in the sector. It computes 6 statistical factors, breadth, drawdowns, volatility, RSI, and momentum, then ranks today against the full distribution. Two separate composites measure different things.

Market Internals measures breadth, how many individual stocks are participating in the current trend. A rising index with falling breadth is a warning sign. A falling index with stabilizing breadth may signal a bottom forming.

Transparency card. Shows how well each AI model is performing on out of sample (unseen) data. These are AUC scores, a measure of model accuracy where 0.5 = random guessing and 1.0 = perfect prediction.

Tracks the global liquidity environment: the macro backdrop that affects all risk assets. When central banks are easing and liquidity is flowing, stocks tend to rise. When liquidity contracts, risk assets face headwinds.

This card is the same across all profiles because it measures macro conditions, not sector specific data.

Shows which individual stocks within the sector have the strongest and weakest momentum. The composite RS (Relative Strength) score ranks every ticker across multiple timeframes. It's not just "what went up the most" but a blended measure of persistent outperformance.

The AION Index page has two ranking views that score tickers using a proprietary multi-factor composite but against different universes. Think of them as two complementary lenses on the same data.

Accessible from the ⚠ RISK MONITOR button. A dedicated risk analysis page that aggregates multiple risk signals into a single view: crash probability trends, exposure history, volatility conditions, and the historical track record of risk calls.

Use this when you want a deeper dive into the risk picture beyond what the main dashboard Crash Detection card shows.

The Macro Regime monitor tracks three cross-asset ratios to assess broad market risk conditions. Each ratio compares a risk asset (tech or equities) against a defensive asset (gold, commodities, or the US dollar).

When risk assets are outperforming their defensive counterparts (ratio above its moving average), capital is flowing into growth, a positive macro signal. When defensive assets lead (ratio below the MA), money is rotating into safety, suggesting caution.

Charts: Each ratio panel shows a 2-year history with green shading during risk-on periods (ratio above its MA). The white line is the ratio, the dashed line is the moving average.

Composite Score: Equally weights all three signals into a single 0-100% gauge. Above 50% means the majority of macro signals favor risk-on positioning. Below 50% suggests defensive conditions. The yellow threshold line marks the 50% boundary.

How to use it: The Macro Regime is a context layer, not a buy or sell signal. When the composite is risk-on, the broader macro environment supports being invested. When risk-off, consider reducing position sizes or tightening stops, even if individual stock signals look bullish. Think of it as the weather forecast for the market: it does not tell you what to do, but it tells you the conditions you are operating in.

Accessible from the 🎲 SIMULATIONS button. Two modes share the same engine: Probabilistic Outlook (default) blends AI forecast probabilities, five per-ticker statistical models, regime classification, and AION INDEX relative strength into a signal-biased projection. Raw Monte Carlo (secondary) runs unbiased Geometric Brownian Motion from historical drift and volatility only, with no signal layer. Both render probability cones around a forward-looking median path.

Think of it as a "what could happen" tool: not a prediction, but a probability map. It shows median, best case, and worst case scenarios with confidence intervals.

Index Analogues scans the full history of the S&P 500 and Nasdaq 100 (back to 1985) for technical patterns that are active right now. Each pattern is a specific market condition, such as a breadth thrust, a Bollinger Band squeeze breakout, a streak of consecutive rallies, or a move near 52-week highs.

When conditions match, the tool finds every historical occurrence of that same pattern, plots the index chart with each occurrence marked, and shows you what happened next. This is not a prediction. It is a factual record: "This exact setup has occurred X times since 1985. Here is what the market did afterward."

How to use it:

Select an index Toggle between SPX and NDX using the buttons at the top. Each index is scanned independently.

Active patterns Green pills at the top show patterns that are currently firing. The percentage next to each pill is the threshold or measured value that triggered it. Click any pill to see its full history.

Historical chart The main chart shows the index price history with vertical markers at every date the selected pattern occurred. This lets you see at a glance whether the pattern tends to cluster at bottoms, tops, or during trending periods.

Occurrence count and range Below the chart: how many times this pattern has triggered, the date range covered, and when the most recent occurrence was.

Available patterns (34 total):

Momentum and Streaks

Consecutive Up Days / Down Days: N consecutive days of positive or negative returns. Longer streaks indicate persistent directional pressure.

Consecutive Higher Highs / Lower Lows: N consecutive days where the high (or low) exceeds the prior day. Measures trend persistence at the bar level.

N-Day Rally / N-Day Drop: Cumulative return over N days exceeds a threshold. Captures large moves regardless of whether every day was positive.

Oscillator Signals

RSI Oversold / Overbought: RSI crosses below 30 (oversold) or above 70 (overbought). Classic mean reversion conditions.

MACD Bullish Cross / Bearish Cross: MACD line crosses above or below the signal line. Trend following momentum signal.

Moving Average Signals

Golden Cross / Death Cross: 50 day SMA crosses above (golden) or below (death) the 200 day SMA. Classic long term trend signals.

Price Crosses Above/Below 200d or 50d SMA: Price itself crossing a key moving average. More responsive than SMA vs SMA crosses.

Reclaim 200d After N Days Below: Price reclaims the 200 day SMA after being below it for an extended period. Often signals a trend reversal.

Below 200d SMA: Price is currently below the 200 day SMA. Bear market context.

Stretched Above/Below 20d SMA: Price is extended far from its 20 day average. Can indicate overheating (above) or panic (below).

Volatility and Range

Above Upper / Below Lower Bollinger Band: Price outside the 2 standard deviation envelope. Signals extreme extension.

Inside Day: Today's high and low are completely within yesterday's range. Signals compression, often precedes a breakout.

Wide Range Day: Today's range is significantly larger than recent average. Signals a volatility expansion event.

Volume Spike: Volume exceeds its recent average by a large margin. Confirms conviction behind a move.

Realized Vol Spike: Short term realized volatility surges relative to its own average. Captures volatility regime shifts.

Gap Up / Gap Down: Today's open is significantly above or below yesterday's close. Signals overnight sentiment shifts.

52-Week Range

Near 52-Week High: Price is within a small percentage of its annual high. Breakout or resistance context.

Near 52-Week Low: Price is within a small percentage of its annual low. Capitulation or support context.

Drawdown From 52w High: Measures how far price has fallen from the 52 week peak. Quantifies the severity of a selloff.

Composite / Thrust Patterns

RSI Thrust (Zweig-Style): RSI drops below 30 then surges above 60 within a short window. A rare breadth thrust signal indicating a sharp shift from oversold to strong momentum. Historically one of the most reliable bullish signals.

Bollinger Thrust: Price moves from below the lower Bollinger Band to above the 20 day SMA. Signals a volatility compression breakout to the upside.

V-Recovery: A significant drop followed by a full recovery within a defined window. Captures sharp reversals.

Snapback From Drawdown: Price recovers a meaningful percentage after a deep drawdown. Signals potential trend resumption after a correction.

Options Heatmaps show where options market makers are positioned across every strike and expiry. Their inventory creates mechanical forces that push price toward certain levels and away from others. The heatmaps surface those forces directly so you can line them up against the rest of the dashboard.

The two grids:

GEX Gamma Exposure per strike and expiry. Long gamma pins price and suppresses volatility. Short gamma amplifies moves and turns strikes into breakout or cascade triggers.

VEX Vanna Exposure. Sensitivity of dealer gamma hedges to implied volatility. High-VEX strikes act as sticky levels into an expiry as dealers rebalance on IV shifts.

Key Levels panel: The numbers above the heatmaps that distill the full chain into actionable reads.

Magnet Strongest long-gamma level above and below spot. Rallies into the magnet above get sold, dips into the magnet below get bought.

Amp Strongest short-gamma level above and below spot. If price breaks through, dealer hedging accelerates the move rather than cushioning it.

All-Exp Flip Chain-wide level where aggregate dealer gamma crosses from positive to negative. Above the flip, price action stays contained. Below it, moves feed on themselves.

Flip Distance How far spot sits from the flip in percent. A small distance means the regime can shift on a normal day’s move.

Star Nodes: Standout dealer walls per expiry. Each chip carries a tag that tells you the behavior to expect if price reaches it.

support Long gamma above spot. Dealers sell rallies into this strike, capping upside.

magnet Long gamma below spot. Dealers buy dips into this strike, cushioning downside.

cap Short gamma above spot. Once price breaks through, dealer hedging accelerates the move higher.

danger Short gamma below spot. Once price breaks through, dealer hedging triggers cascades lower.

Regime Reads: Front week, near-spot (±10%), and the full chain each classified as long gamma (dampening, dealers stabilize), short gamma (amplifying, dealers chase), or mixed. Checking all three at once shows whether the environment is structurally calm, fragile, or split across timeframes.

How to use it: A bullish AI Forecast on a ticker sitting below its All-Exp Flip with a danger node right under spot is a different trade than the same forecast on a ticker above its flip with a magnet supporting it. The dashboard tells you where the models think the ticker is going. The heatmaps tell you what the options market is likely to do to price on the way there.

The Predictive Outlook is a visualization, not a separate model. It takes the outputs that already exist on your dashboard (AI Forecast probabilities, statistical regime classifiers, crash detection, market stress, exposure, and liquidity regime) and renders them as a forward-looking probability cone using Monte Carlo simulation.

The chart shows the sector's equal-weight index history as a solid cyan line, with a shaded cone projecting forward from today. The cone's shape, direction, and color are entirely determined by the current model readings.

What drives the cone:

AI Forecast The 3-day, 10-day, and 20-day probability-of-upward-move values set the directional drift. Higher probabilities tilt the cone upward; lower probabilities tilt it down.

Statistical Regimes (L1 to L5) The regime states (AGGRESSIVE, CONSTRUCTIVE, DEFENSIVE, MAXDEF) and their expected returns contribute the largest share of directional bias. Regime pills at the bottom of the card show each model's current state and probability.

Crash Detection Elevated crash probability injects downside jump risk into the simulation and widens the cone with heavier tails.

Exposure & Stress Low recommended exposure or high market stress widens the cone (more uncertainty). High exposure and low stress tighten it.

Liquidity A HEDGE liquidity regime nudges the expected drift slightly downward.

Reading the chart:

The dashed line is the expected path, a realistic noisy projection following the model-implied direction, not a smooth line. It resembles actual price action because it is generated with momentum, mean-reversion, and market-like noise.

The shaded bands show probability ranges: the inner band covers ~68% of simulated outcomes, the middle ~80%, and the outer ~95%.

A green cone means the blended outlook is bullish. A red cone means bearish. Cyan is neutral.

Use the dropdowns in the top-right corner to adjust the history length (60 to 180 days) and forecast horizon (10 to 30 days). Changing either re-runs the simulation instantly in your browser.

Stats bar: The row of values above the chart provides a snapshot of the key inputs driving the cone: 3D/10D/20D probabilities from AI Forecast, crash probability, statistical regime consensus (e.g. “3/5 Bull” meaning 3 of 5 stat models are in a bullish regime), market stress level, recommended exposure, and liquidity regime.

The Probabilism Index (PI) ranks every ticker by how heavily its per-ticker statistical signals stack in the same direction. It lives below the cone on the Simulations page as two live leaderboards: Highest Statistical Upside on the left (green) and Highest Statistical Downside on the right (red). Click any row to load that ticker’s cone above it.

Each ticker is scored across several independent per-ticker signals and rolled into a single composite from roughly 0 to 100. When those signals all align in one direction, the score stacks to an extreme. Two names at the same AI Forecast probability often sit far apart on the PI because one has every underlying signal aligned and the other is carrying it on one or two.

What stacks into the score:

TREND How cleanly the ticker is trending on its own price history.

T+MOM Trend confirmed by momentum moving in the same direction.

VOL ENV Where price sits inside the ticker’s own volatility envelope.

MOM Rate-of-change momentum strength.

AION The ticker’s AION INDEX percentile relative to the full universe.

Reading the table: Each row shows the ticker, its composite PI, and the per-component scores that drove it. The upside column sorts descending, the downside column sorts ascending. Each side shows the top 30.

When you open a ticker’s cone, the stats row above the chart includes a PROB INDEX pill carrying that ticker’s current score, color-coded green at 70+, cyan in the middle, red at 30 or below. Same read as the leaderboard, at a glance while you’re on a single name.

How to use it: High PI means the statistical stack is aligned up. Low PI means the stack is aligned down. Use it to find names worth deeper inspection: a high PI in a sector you already like, a low PI on a name you already hold, or a disagreement between the PI and the AI Forecast. Then confirm with the rest of the dashboard before acting.

Click on any individual ticker (in the AION Index or View Tickers grid) to generate a real time AI research report. This uses a large language model to pull together fundamental data, recent news, analyst sentiment, and technical factors into a concise summary.

Each subscriber gets 100 reports per month. Your remaining count is shown in the report dialog.

Each profile in the dropdown is an independent universe of stocks with its own models trained on that specific group. Switching profiles completely changes all signals on the dashboard.

No single card tells the whole story. Here's a practical framework for reading all the signals together: