Ticker Council is a stock research tool built around one idea: a single AI answer hides its disagreements, so show them instead. You type a ticker, twelve AI analysts each look at one kind of evidence, they argue it out, and the Council tells you which way it leans for the next week, the next 3 months and the next year, with how sure it is. Later, every call is checked against what the price actually did.
This page explains each step, in the order it happens.
Step 1: twelve seats, each with one kind of data
Each "seat" is an AI analyst that is only allowed to see its own data. That's deliberate. When one model sees everything at once, the loudest input (usually the latest headline) tends to drown out the rest. Splitting the evidence up makes each kind of data speak for itself.
The twelve seats are:
- Price Chart: price and trading volume, trend and momentum.
- Financials: revenue, profit, debt and valuation.
- Analyst Targets: Wall Street price targets and rating changes.
- Earnings Estimates: forecasts for future earnings and how they are moving.
- Insider Trades: what executives and directors buy and sell (SEC Form 4).
- Congress Trades: trades disclosed by members of the US Congress.
- News: headlines and upcoming events.
- SEC Filings: buybacks, new shares, debt and lawsuits.
- Institutional Holdings: what big funds hold (13F filings) and short interest.
- Options Market: what options traders are betting on.
- Economy: interest rates, inflation and the wider economy.
- Related Markets: similar stocks, the index and overseas markets.
Each seat says up, down, even (its evidence genuinely cancels out) or "no read" (it couldn't get usable data), and gives its reasoning in plain words. A seat that has nothing to say is allowed to say so, rather than inventing a view.
Step 2: the debate
After the seats vote, two advocates take over. A bull advocate makes the strongest case that the stock goes up, and a bear advocate makes the strongest case that it goes down, each drawing on what the seats found.
Then the Challenger goes after whichever side is winning. Its only job is to find the hole: the risk the majority skipped, the data point that doesn't fit, the earnings date three weeks out. Investors call this steelmanning the other side. It's the step people most often skip when they research a stock alone, because once we lean one way we tend to look for evidence that agrees (confirmation bias).
Step 3: the verdict
A final judge, the Grand Master, reads the seats, the debate and the objection, and writes the verdict. For each period (next week, 3 months, a year) you get:
- A chance, not a slogan. For example "55% up". A call near 50% is close to a toss-up, and the site shows it that way on purpose, with words like Barely and Leaning instead of hype.
- How much the seats agree, and which ones leaned which way.
- A likely price range for the end of the period, based on how much that stock usually moves.
- The reasoning, in plain words or with every number.
Nothing in the verdict says buy or sell. It's a second opinion for you to weigh, not an instruction.
Step 4: every call gets scored
This is the part most tools leave out. Every call is saved where it can't be edited, then scored against what the price actually did a week, 3 months and a year later. The scoring is explained in How Ticker Council scores its calls, including where the record stands: it's too early to tell, because the first 3-month calls aren't scored until December and the first 1-year calls until next autumn.
Every call stays in the record, right or wrong, because a tool that only shows its wins can't be trusted with anything.
What it isn't
- Not financial advice. Every call is the same for everyone who runs that stock, and it can be wrong.
- Not a broker. It doesn't place trades or manage money.
- Not a crystal ball. Markets are mostly noise over a week. The useful part is often the argument (which evidence points which way), not the final number.
What it costs
Free accounts get all twelve seats, the debate and the verdict, using your own free Google Gemini key (about two minutes to set up, no card). See how to get a free Gemini API key.
Questions
Why use twelve AIs instead of one?
Because one model that sees everything tends to over-weight the most recent or most dramatic input. Giving each seat a single kind of data, then making them argue, shows you where the evidence actually agrees and where it splits.
Which stocks can I run?
US-listed stocks and Canadian (TSX) stocks. Funds and other listings can run too, but some seats sit out when their data doesn't exist for that kind of security.
How long does a run take?
A few minutes. Each seat's card fills in as it finishes, and you can tap any card for its full reasoning.
Is Ticker Council financial advice?
No. It's a research tool. Its calls can be wrong, and you should make your own decisions and only invest what you can afford to lose.