A flood of filings, prices and headlines, turned into four clear numbers per stock after every trading day. Free to look at all of it.
Free to start · No credit card · Research and simulation software — not investment advice, and not a recommendation to buy or sell any security.
Example: four measurements on one company, one of them deliberately blank.
You can ask a chatbot about one company and get a fluent answer. What you cannot do is measure a whole market against itself, using last night’s numbers, every night. That is a different kind of problem — and it is the part that costs money.
Company accounts, prices and published coverage for a global list of stocks — collected, cleaned and kept in line every single night. The maths is the well-understood bit. Keeping the feed honest is the work.
Scoring the whole list costs us the same whether one person looks at it or ten thousand do. Doing it privately, for yourself, means carrying that entire bill alone — for one portfolio.
Nothing here is tailored to you, and that is deliberate. A stock’s score is the same for every person looking at it — so it is a measurement you can compare and check, not advice aimed at your situation.
Four separate models, each answering one question, scored 0–100 and refreshed after every trading day. Three are trained on historical market data; News is a language model reading published coverage. They are deliberately allowed to disagree — you see four independent readings, not one verdict.
Whether a stock has been rising, how steadily it has done it, and whether the move still has strength behind it.
What the shares cost today, set against what the company really brings in — its profits, what it owns, and the cash it generates.
Whether the crowd has already piled in, which industries money is moving toward, and how the stock compares with its closest rivals.
An AI reads the published coverage and turns a wall of headlines into one number and a short summary you can read in a few seconds.
0–100 is a rank, not a grade. 80 means it scores higher than 80 out of every 100 stocks we cover on that one measure — nothing more than that.
Illustrations of the actual screens, built from the same components. Tickers and figures shown are examples.
The whole universe ranked by four models that are allowed to disagree — so you can see why a name sits where it does, not just that it does.
Every score we have for a name, side by side. A strong trend and an expensive price are visible at the same time.
Sort and filter the whole universe by any of them.
A dash means we have no honest read for that model on that stock. We leave the gap open rather than invent a number.
Set a rule: any score, any threshold, any rebalance. The engine runs it over market history and shows you what it did, good stretches and bad.
Pick the score, the cut-off, the region and how often it rebalances.
Reconstructed history and live results are drawn differently, so you can always see which part is which.
Our own models' records are in there too — judge them the same way you judge your own rule.
A list of names you want to follow before committing anything. Rescored every night, so you see what changed while you were deciding.
Your list against the universe average, so you can see whether it is actually any good.
Overnight score moves per name, not price noise.
Nothing here places a trade. KhaidoLab is research and simulation; the order goes to your broker.
Every possible mix of your holdings, plotted risk against return. Fees and trading costs are included, so the estimate is closer to what you would actually get.
Drag along the curve to any mix; the lowest-risk, best-balance and highest-return points are marked.
Mean-variance optimisation, applied to the holdings you chose.
Simulated before you commit anything — then you place it at your broker, not here.
If you want it, your paper portfolio is scored against everyone else’s — anonymously, on percentage return, so account size never matters.
Ranked on return, risk-adjusted return, drawdown control and consistency.
Everyone appears under a generated handle. No names, no account sizes.
Entirely opt-in — the rest of the product works without it.
Three steps, every trading day. Nothing here tells you what to buy.
After the close, four independent models read a global universe of stocks and put a 0–100 number on each one. Where a model has nothing honest to say about a stock, it returns nothing — you see a dash, never a filled-in guess.
A language model turns each score into plain sentences: which measurements moved it, and what the model is reacting to. So a number is never just a number you are asked to accept.
Treat a score as a starting point, a contrarian signal, or nothing at all. The backtesting engine is there to test our rules — or to ignore them entirely and test yours.
KhaidoLab is research and simulation software. It does not give investment advice, does not make recommendations, and does not execute trades.
Sort the whole covered list by any of the four measures.
One page per stock, per measure, in plain sentences.
Build the position on paper first. Nothing is bought.
Scores are redone nightly, so you see what moved.
Weigh risk against return across your holdings, costs included.
None of this is new, and none of it is secret — large institutions have worked this way for a long time. What kept it out of reach was the cost of the data and the effort of keeping it running, not the ideas.
Every plan sees the whole scored universe and runs the same nightly engine. You pay when you want the product working while you are not looking — alerts, portfolios — and then for the analysis tools on top.
Prices exclude VAT, which is added at checkout at your local rate. Cancel any time. 14-day money back, no questions asked.
We are working on individual screening and on plans for institutional clients. If that is relevant for your team, get in touch and tell us what you need.
Get in touchNot an offer, a product commitment, or a delivery date. We will only contact you about KhaidoLab product availability.
The whole scored list and the backtesting engine are free to use. No credit card to begin.
Start free