ILLUSTRATIVE — NOT A LIVE QUOTE
Stop watching for the signal. Let the rule watch for it.
Every setup you can chart here, you can also leave running. Pick the stocks, pick the indicators, set the conditions — and the terminal evaluates them for you on every new bar, places the order into a simulated account, and keeps the record of why it did. Four stages, and you can read all four.
Your indicators, your symbols
Up to six indicators over up to 25 stocks, combined so that any of them firing counts — or so that all of them must agree. Daily or 15-minute bars. Every price rule is decided on a bar's close, because this app stores no intrabar path and will not pretend otherwise.
Every guard fails closed
Position limits, universe limits, how much may happen without you. If a guard cannot resolve, it refuses rather than waving the trade through — and the refusal is stored with its reason, so “why did nothing happen today” is a question with an answer.
Into a paper account
The same pending order you would place by hand, worked by the same engine against real prices. A limit the session never reached rests, with its reason. Routing to a broker is declared and then refused — there is no live execution path in this app.
The record, not a summary
Every fill, every balance and the trace behind each one. A strategy created today is replayed over the history its indicators needed so it opens with a record — and those seeded fills are counted separately and never alerted on, so a backfill can never read as something that just happened.
ALGO Trade
Your own rules, left running. Choose the stocks, choose the indicators, say what the price has to be doing, set the capital per name — then let it work while you do something else. You get told when it signals, and the fills reach your paper account.
- A price range that means one of four things, and you pick which
- Daily bars, or 15-minute bars
- Email when it fires, entitlement re-checked at the moment of sending
- Editing the rule clears the record — and asks first
Trading agents
A step past a fixed rule: an agent that decides, per session, what to trade inside limits you set. Three things you choose separately — who decides, where the order lands, and how much happens without a person in the loop.
- Rules, select or propose — who picks the trade
- Auto, approve or mixed — how much runs unattended
- Letting a model propose the direction is its own switch, on the top plan, and the scheduler refuses it unless you opted that agent in
The paper account
Place orders against real prices and see what would have happened — by hand, or driven by a rule. Fills are simulated and routed nowhere. Options too: up to four legs, all-or-none on a signed net, with Reg-T style collateral.
- An oversell is refused, the way your broker would refuse it
- Expiry, exercise, assignment and cash settlement are handled as four events — not as four kinds of sale
- Reset the account and the old ledger stays readable
A factor model, in public
Value, quality, momentum, size and low-volatility scores, computed across the whole US universe every night — and a book that rebalances on them quarterly, which any signed-in account can read. It demonstrates a method rather than showing somebody's trading.
- Forward, never backtested — the ledger starts the day it was switched on
- Every fill is charged commission and slippage
- The growth curve is rebuilt from that ledger on every request, never stored
Every data source. One terminal.
Live charts, real indicators, a bank of machine-learning forecasts, options flow and greeks, insider filings, institutional 13F positions, order-flow classification, basket risk, the economic release calendar, a market-wide dashboard and a server-side screener — fused into one view, so every call is backed by evidence rather than a hunch.
Insight
Four independent lanes on one symbol — price dynamics, fundamentals, news and the macro backdrop — read separately, then fused into one card that says whether they agree. Not a score out of ten: a verdict about alignment, the readings underneath it, and an honest count of how much of the evidence actually resolved. “Not enough evidence” is one of the answers it can give.
Economic indicators
91 series — inflation, jobs, growth, the yield curve, credit spreads, housing, the consumer, the dollar — charted back 25 years with the release calendar beside them. Series that share a unit share an axis and nothing else does, so a rate near 78 never flattens a change near 3.
Machine-learning forecasts
A bank of models forecasts where the price could go next, each drawn on your chart with its own confidence band. Where they agree, you have conviction. Where they split, you have a warning — and you see both, instead of one number dressed up as certainty.
Multi-chart grid
Every cell in the grid runs its own full indicator engine — watch four setups develop side by side, not four screenshots.
TradeWickAI proprietary signals
Wave-count overlays plus a reversal early-warning signal with structure and persistence tracking, built in-house — so a call isn't gone the next candle.
Server-side screener
Scan the full instrument universe against any screenable indicator — the same calculation engine that draws your chart, not a separate approximation.
Stock screener
Filter the market by sector, market cap, dividend yield, beta, and 52-week range — the fundamentals workspace you'd otherwise pay for as a separate subscription, built in and free.
AI chart analysis, built in
Ask what a setup means in plain English — the AI panel picks indicators relevant to what's actually on your chart and explains the terms it uses, right there.
Alerts that reach you
Price levels, percent moves, or an indicator signal firing — delivered by SMS, WhatsApp, or Telegram, your choice, with a full delivery log.
Options chain & Greeks
Full chain with live Greeks, Max Pain, Put/Call Ratio, and Average IV — plus options-specific Bull/Bear setup indicators.
Any overlay, any pane
Drop a moving average directly into your Momentum or RSI panel instead of the main chart — compare shape and crossovers side by side with the oscillator itself.
Market dashboard
The whole market in one screen: index candles that go live during the session, advancing-vs-declining and 52-week breadth, gainers, losers and unusual volume, and a sector map sized by how many names each sector holds. Every panel is computed in one pass over the whole US stock list, so the advancing count and the movers list can never disagree.
Sector rotation, up to four years back
Median quarterly return per sector, ranked into a grid where leadership changing hands reads as a colour band walking up or down. Choose 4, 8, 12 or 16 quarters, and hover any cell to trace one sector across all of them. The quarter still running is marked and never ranked as though it were finished.
Playbook — 48 prebuilt scans
A shelf of ready-made screens across 11 groups: insider buying clusters, dividend and quality filters, drawdown and momentum cohorts, Elliott-wave catalysts, reversal sweeps, crowded shorts, and 18 industry themes. Each card states the exact rule it runs — including what it excludes and why — and every play scans whichever market your display currency is set to.
Insider filings — and who's talking
SEC Form 4 activity per symbol: who bought, how many shares, at what price, with open-market trades kept apart from routine vesting. Then the Podcast panel finds long-form interviews by those same filers and has the model watch one and summarise it, every point timestamped back to the video.
Institutional holdings, 13F
What the funds actually own, straight from their quarterly filings — holders per symbol, a fund browser, sector mix and a filing-date backtest. Reported as filed, 45 days after quarter end, so you are reading a real position rather than a guess at one.
S&P 500, sized by weight
A squarified treemap of all 500-odd constituents: tile area is market cap, colour is the session's move, blocks are sectors. The index is cap-weighted, so a big NVDA tile is a fact about the index rather than a flourish. Membership is a dated file refreshed on demand, and every view shows that date rather than implying it's live.
Dashboard — build your own screen
79 widgets you arrange yourself, on as many named dashboards as you like, and the layout persists: quotes, mini charts, breadth, movers, earnings, insiders, 13F ownership, options positioning, economic indicators, your positions and your stake in a name. The same catalog powers Symbol Overview, a fixed one-symbol window over every tile at once.
Portfolio on a real ledger
Positions are derived from a transaction ledger, not typed in as a summary — so cost basis is computed, both FIFO and average cost, with fees where they belong. Switch method and the numbers change the way your broker's would. It's the one part of the app checked by its own test suite, because it's the only number you read as your own money.
Build your own indicator — no code
Your edge, as an indicator, in a few clicks. Pick the indicators you already trust, set the conditions, name it. It lands on your chart as an ordinary indicator — screenable, alertable, backtestable — sitting alongside the 235 built in across 14 groups. Build a signal, a strategy (which has to declare its own exit, so it can be tested honestly), or a plotted line other rules can read. No syntax to learn and nothing to compile.
Already have a TradingView script? Import the .pine file and run it here — our own Pine engine compiles and plots it on the chart, and on the top plan runs it across every listed symbol server-side. Pine here is an import path, not a place to write one: a script arrives as a file and the editor holds it read-only. Authoring happens in the builder above, where a rule can call any of the 235 indicators — something Pine cannot do. On import you get a compatibility report that names every unsupported call in the script in one pass, read straight off what the engine actually implements, so you learn where it stops before you trust a plot rather than one error at a time.
Basket risk, six ways
Measure a portfolio, a watchlist or a screen's results as one book: correlation and diversification, statistical and fundamental factor exposures, stress scenarios, performance statistics, historical VaR, and risk contribution per holding — which is a different question from weight. A holding with no price history is excluded and named, never quietly given a zero weight.
Options, past the chain
The implied-vol surface across strikes and expirations, skew, ATM term structure, IV rank and net dealer gamma by strike. A multi-leg strategy builder with payoff, breakevens and net greeks — and its inverse, a scanner that takes a target price and ranks the structures that express it by return on risk. Plus what the quoted calls jointly imply about where the price lands, by finite difference: no volatility model, nothing fitted, nothing smoothed.
Which side traded
Live cumulative volume delta, every print classified against the quote that was resting before it. A print with no quote behind it is left unclassified and counted as such rather than guessed at. Beside it, FINRA's daily off-exchange and short-sale volume — a flow, not short interest, measured against off-exchange volume rather than the consolidated tape, because on AAPL those two differ by about 3.4x.
Today — what actually changed
One feed over the names you hold and follow: insider filings, 13F moves, alerts that fired, contracts about to expire, off-exchange spikes, earnings ahead, economic releases. Every row carries its own date and its own source, and the feed composes sentences rather than figures — so nothing on it can be mistaken for a live price.
Four kinds of evidence. One question: do they agree?
Price says one thing, the filings say another, the headlines say a third, and the macro backdrop is doing something else entirely. Insight reads all four separately — no lane can see what the others concluded — and fuses them into a single card about whether they line up. Ask for one on any symbol, or have it delivered before the open on weekday mornings.
-
1
Four lanes, read independently
Price Dynamics, Fundamentals and News are about the company. Macro is the backdrop, and is marked as backdrop rather than being folded in as if it were a fact about the business.
-
2
“Not enough evidence” is a real answer
Below three directional readings across all four lanes, there is no verdict — because two agreeing readings inside one lane is one observation counted twice. The card says so rather than producing a direction from thin evidence.
-
3
There is no confidence score
A number out of ten would be invented, and it would be the most quoted thing on the card. What sits in its place is coverage: how many readings resolved out of how many were available, and the age of the oldest input. Both are reconstructable from the card itself.
-
4
Counted or judged — and it says which
Ledger derives every direction from a published figure and calls a model once, for the summary paragraph. Council instead has a model judge each lane and reports that it did, badged on every reading. Same card, so the two can be compared rather than described to each other.
Price Dynamics
Trend, order flow and the factor scores lean the same way.
Fundamentals
Margin and revenue growth improving; last earnings beat.
News
Sentiment positive across 22 days with coverage.
Macro
Backdrop, not a reading about this company — direction by stated convention.
Lanes agree — constructive
Three lanes directional and pointing the same way. Coverage: 9 of 12 readings resolved, oldest input 3 days old.
48 scans.
Every Playbook card states the exact rule it runs — the bounds, the units, what it excludes, and how many names actually survive it — so a short list reads as a short list rather than as a verdict.
-
1
Screen on what the signal said
New: scans can now match a marker's own label, not merely the fact that it fired. “Just finished a Wave 2” finds names whose Elliott count currently reads 0-1-2 — where Wave 3 is still ahead — rather than every chart that has ever printed a Wave 2.
-
2
Signals crossed with fundamentals
New: “Reversal signals on quality names” runs all 33 screenable reversal indicators over a week, then keeps only profitable, growing companies with a return on equity of 15% or better and debt-to-equity at or under 1.0×.
-
3
The funnel is on the card
That quality filter takes the US list from 1,487 names to 236, and the India list from 479 to 10 — and the card says so. A scan that returns four names has told you something; a scan that returns four names without saying it started from 1,487 has not.
Both markets, one shelf. Every play scans whichever market your display currency is set to, and the price floor is in the listing's own currency — a real bound on a US line, barely one on an India line. The cards that depend on US-only data say which they are.
A signal is only useful if you can see why it fired.
Most chart tools show you a triangle and ask you to trust it. TradeWickAI draws the reasoning: the swing points, the confluence factors, and the crossover that actually triggered the call — evidence lines, on by default, not a setting buried three menus deep.
-
1
Structure forms
Swing highs and lows build up across the lookback window your indicator is watching.
-
2
Confluence crosses
Eight independent reversal signals, tallied bar by bar — the line crosses the moment enough of them stop disagreeing and swing the same way.
-
3
Evidence stays on the chart
The points behind the call stay drawn, so you're checking the reasoning, not taking it on faith.
One watchlist. Four asset classes.
US and India-listed equities side by side in the same currency-aware watchlist and screener — no separate app for NSE/BSE names — plus crypto and 29 spot FX pairs, each charted on its own terms. A rate isn't an amount of money, so pairs are never re-quoted into your display currency, and FX volume is shown as unknown rather than as zero.
Start free. Upgrade when you're ready.
Four plans, and the line between them is what a feature costs us to run — not an arbitrary split. Free costs nothing and isn't a trial.
- Live charts, US & India, with the real-time price stream
- The full indicator library and the reversal menu
- Market dashboard, S&P 500 treemap, sector rotation and the screeners
- Playbook scans, watchlists, Dashboard, portfolio and alerts
- The machine-learning price forecasts, on the chart
- Analyst ratings, and the no-code indicator builder
- ALGO Trade — your own rules, left running
- The paper account, stocks and options
- Options mode — chain, Option Charts, options indicators
- Institutional holdings (13F)
- Economic indicators — 91 series and the release calendar
- Chart AI — ask about the symbol you're looking at
- Insider Podcasts, summarised
- Your own built indicators on the chart, and the options indicator set
- Trading agents — an agent that decides per session, inside your limits
- Insight — four evidence lanes on one symbol, and whether they agree
- Earnings interpretation in plain English
- Ask AI — describe a screen, the model writes the SQL
- Sector analysis and crypto AI reads
- AI impact colouring on the news feed
- The options screener, and AI commentary on your basket risk
- Pine import — bring a TradingView
.pinefile, compile it, plot it
- Strategy backtesting with trades and an equity curve
- Screener beta — multiselect indicators, explicit date ranges
- Pine market scan — one imported script over every symbol, server-side
- Full-universe scans, run server-side across every listed name
- Model-proposed trades — the agent arm where the model picks the direction, badged on every order it produces
Pricing and the upgrade flow are still being finalised, and nothing is charged today — signing up locks in your account on whichever plan you pick, with no surprise charges. Every plan sees the same market data; the tiers differ in what you can run over it.