Methodology
Where our figures come from, what each number measures and what it does not. This page explains the method and the vocabulary; it does not describe how the tooling is tuned internally.
Where the figures come from
Every accounting figure we publish comes from the report the company itself filed with its regulator. Nothing is taken from aggregators, third-party summaries or press releases. When a report says a company earned a given amount, that amount sits in a document signed by its officers and anyone can check it.
«Line by line» is literal: the full income statement, balance sheet and cash flow statement are extracted, with their detail accounts, not a four-figure summary. That is what lets a diagram show where every dollar goes instead of jumping from revenue straight to the bottom line.
Prices and market data arrive through a chain of sources with fallbacks: if the first does not answer, the second takes over, and then the third. An isolated failure degrades how fresh the figure is; it never invents one. When a figure is unavailable we show a dash — never an estimate dressed up as a measurement.
What gets calculated, and where
Calculations run on the server, not in the browser. That has a consequence worth knowing: what you see on screen is the output of a computation, not a template filled in with numbers.
The same engine produces the figures in the public research and the figures in the subscription tools. There is no simplified version for free content. And the accuracy of a calculation never depends on the plan: a plan limits how many lists or how many assets you can follow, never the quality of the number.
How uncertainty is expressed
The future of a price is not a number, it is a range of possibilities with differing likelihood. That is why our own engine — Q-PathEngine™ — does not compute «the price it will reach». It simulates fifty thousand complete price paths, day by day, out to the date under study, and describes the resulting distribution.
Two distinct things come out of that, and they are routinely confused. One is where the price ends up at expiry; the other, very different, is whether it touched a level at any point along the way. The second is always more likely than the first, and a calculator returning the same number for both is simulating a single jump instead of the journey.
Results are published as probabilities and percentiles, never as a verdict. A 5th percentile is not «the floor»: it is the value below which five per cent of the simulated paths ended. The distinction matters, because the first reading implies a guarantee no model can offer.
What this method does not do
- It does not predict prices. No number on this site claims where an asset will be.
- It does not issue buy or sell signals, and does not sell them in any form.
- It does not score assets with an overall grade, nor rank them as «best» and «worst».
- It does not promise returns, nor publish trading track records as a sales argument.
- It does not replace the reader's judgement. It describes conditions and probabilities; the decision is always the reader's.
Known limits
Stating what a method cannot measure is part of the method. These are the limits that affect what we publish today, and they are here because anyone reading a number is entitled to know what constrains it.
- The options feed we use does not distinguish the price at which you buy from the price at which you sell: it returns a single price per contract. Direct consequence: the cost of that difference cannot be measured, so no yield we publish deducts it. We say so rather than estimate it.
- Probabilities are computed under the risk-neutral measure, the standard for pricing options. That means they do not incorporate the asset's historical drift. A tool that does incorporate it will return different numbers: neither is wrong, they answer different questions.
- The simulation advances in daily steps, not continuously. That slightly understates the probability of touching a level, because a move that happens and reverses within the same day goes unrecorded. The bias is conservative and known.
- Open interest tells you how many contracts exist, not who holds them. Every aggregate exposure measure — GEX and DEX among them — requires assuming which side the dealer is on, and that assumption is not observable. Whenever we publish one of those numbers, we state the convention it was computed under.
- Stock indices are not available from our data sources; freely quoted equivalent instruments are used instead. Exchange-traded funds do not report the same fundamentals a company does, so those fields appear empty rather than filled with a substitute.
- The regulator's public repository does not always expose every tag a company used, even when it is present in the original filing. When one is missing, we show the gap. It is not filled by interpolation or by analogy with another company.
Glossary
Each term is explained on its own, together with the misreading it most often attracts — usually more useful than the definition itself. Knowing what something is called is half of being able to argue about it.
Options contracts
The minimum vocabulary needed to read an options chain without depending on someone else to interpret it for you.
- Open interest(OI)
The number of contracts in a series still alive — open, neither closed nor expired. It updates once a day, after the close.
It is not the same as volume. Volume counts how many contracts changed hands today; open interest, how many exist in total. High volume with flat open interest means positions were opened and closed, not that new ones are accumulating.
- Strike price
The price at which the contract allows you to buy or sell the underlying if exercised. It is fixed from the moment the contract exists and defines the whole geometry of the outcome.
- Premium
What the contract costs. The buyer pays it; the seller collects it and takes on the corresponding obligation in exchange.
- Break-even
The price of the underlying at which the position stops losing and starts making money at expiry. It is not the strike: it also accounts for the premium paid or received.
- Implied volatility(IV)
How much movement the market is pricing into that specific contract, inferred from its price. It is a reading of expectation, not of history.
It says nothing about direction. High implied volatility means movement is expected, not which way. And it varies between contracts on the same asset: two different strikes can carry very different implieds.
- Historical volatility(HV)
How much the asset actually moved over a past window, computed from its daily returns. Measurement, not expectation.
Implied sitting above historical is the norm, not an anomaly. The gap between the two is a subject of study in itself, not a signal.
The Greeks
They measure how a contract's price reacts to changing conditions. They are derivatives — sensitivities: they say «if this changes, that changes by this much».
- Delta
How much the contract's price moves if the underlying moves one unit. A delta of 0.30 means that for a one-dollar rise in the underlying, the contract rises about thirty cents.
It is widely used as if it were the probability of finishing in the money. It resembles it, but is not the same thing, and confusing them means reading a price sensitivity as a probability.
- Gamma
How much delta itself changes as the underlying moves. It is the acceleration: it measures how fast a position changes character while the price travels.
- Theta
How much value the contract loses from the mere passing of one day, everything else held equal. It runs in favour of whoever sold the contract and against whoever bought it.
- Vega
How much the contract's price changes if implied volatility rises by one point. It is exposure to a change in the expectation of movement, independently of the price moving at all.
Aggregate positioning
These measures do not describe a contract but the picture of the whole. They all share one assumption worth keeping in mind before reading any of them.
- GEX(gamma exposure)
The sum of gamma across every open contract on an asset, weighted by open interest and expressed in money. It describes how hedging across the whole would tend to behave as the price moves.
It requires assuming which side of each contract the dealer is on, and that cannot be observed: open interest says how many contracts there are, not who holds them. It is computed under a market convention, and therefore reads as «under that convention, this would imply…», never as a measured fact. In money terms it is also not comparable across assets: two companies of very different size can produce the same figure and mean opposite things.
- DEX(delta exposure)
The same exercise applied to delta: the aggregate directional exposure across open contracts, in money.
GEX and DEX do not answer the same question and are frequently mixed up. Aggregate delta points to a directional lean; aggregate gamma, to how that lean would change with movement. The same asset can carry the two with opposite signs.
- Liquidity wall
A strike where open interest concentrates far above its neighbours. It marks where activity has piled up, which is a fact about structure.
A concentration is neither a magnet nor a barrier. It describes where open contracts sit, not what the price will do on reaching them.
- Gamma flip
The price level at which aggregate gamma exposure changes sign. Computing it properly means revaluing every contract's gamma at different prices, not reading the snapshot at today's price.
Probability and simulation
- Path
One complete run of the price, simulated day by day out to the date under study. The engine generates fifty thousand of them and describes the set; none of them is a forecast, and neither is the most frequent one.
- Percentile
The value below which a given share of the paths ended. A 5th percentile of one hundred dollars means five per cent of the simulated runs finished below that figure.
It is not a floor or a ceiling. By definition, five per cent of cases fell outside it.
- Risk-neutral measure
A pricing convention that assumes the asset grows at the risk-free rate, without incorporating its historical drift. It is the standard for pricing options because it keeps the model consistent with the price the contract actually trades at.
It is why these probabilities differ from those of a tool using the asset's past return. Neither is in error: they answer different questions.
- Q-Riesgo
The probability that the price finishes below the level under study on the expiry date. It looks only at the outcome.
- Q-Barrera
The probability that the price touches that level at some point during the period, even if it moves away afterwards. It looks at the whole journey.
It is always greater than Q-Riesgo, and that inequality is a check that the simulation is walking complete paths. If a tool returns the same value for both, it is simulating a single jump to expiry.
Financial statements
Company research rests on these distinctions. Almost every common misreading comes from conflating two of them.
- Operating margin
What share of revenue survives the cost of producing and of operating, before interest and tax. It describes the business itself.
- Net margin
What share survives everything, once interest, non-operating items and tax are through.
It can swing sharply without the business changing: a different tax rate or one extraordinary item is enough. When operating and net pull apart, the explanation is rarely in the operation.
- Gross versus net revenue
Two companies can call «revenue» magnitudes that cannot be added together. A trading desk may report the gross amount passing through it; a services company reports what it bills.
It is the most expensive comparison error there is, and the easiest to make, because the label is identical. Before putting two figures with the same name on the same chart, open each one.
- Deferred revenue
Cash already collected for a service not yet delivered. It sits as a liability, because an obligation remains outstanding.
It is an obligation to deliver a service, not to return cash. A liquidity ratio that drops it into the denominator without distinguishing it can paint a picture considerably worse than reality.
- Stock-based compensation
Pay settled in shares rather than cash. It does not leave the treasury, but it dilutes existing shareholders.
That is why it is studied alongside buybacks: a company can repurchase shares and increase its share count at the same time.
Frequently asked questions
The ones that come up most often, answered plainly.
Does Quantika Research recommend buying or selling?
No. Quantika Research issues no buy or sell recommendations, sells no trading signals, markets no indicators that anticipate price movements and offers no automated execution systems. It publishes quantitative analysis and tools so that each person decides for themselves. It is not registered as an investment adviser with any securities authority.
Does Quantika Research hold client funds?
No. Quantika Research is subscription software: it does not hold funds, does not execute orders and does not manage third-party portfolios. Users trade on their own account through the broker of their choice; Quantika Research takes no part in any transaction.
Do I have to pay or register to read the research?
No. The research is free, complete and accessible without an account or registration. There is no paywall, no cut-down version and no obligation to leave an email address. The subscription gives access to the tools for running that same analysis on assets of your choosing, not to the contents of the research.
Where do the figures in company research come from?
From the financial reports each company files with its regulator. The complete statements are extracted line by line, not a summary, and the published figures reconcile arithmetically with each other before a single sentence is written. When a figure is unavailable we show a dash rather than estimate it.
Why do the probabilities differ from another calculator's?
Because they most likely answer different questions. Quantika Research computes under the risk-neutral measure, the industry standard for pricing options, which does not incorporate an asset's historical drift. A tool that does use past returns will give different figures. It is also worth checking whether the other one computes the probability at expiry or the probability of touching the level at some point: those are two different numbers, and the second is always larger.
What does it mean for an asset's GEX to be positive?
GEX sums the gamma of open contracts weighted by open interest, to describe how hedging across the whole would tend to behave as the price moves. Its sign depends on an assumption about which side of each contract the dealer sits on, and that is not observable: open interest indicates how many contracts exist, not who holds them. It therefore reads under a stated convention and never as a measured fact. The money figure is also not comparable across assets of different size.
Do I need to know how to program to use the tools?
No. The tools run in the browser, with no code to write. What is genuinely required is conceptual knowledge, which is why the glossary and the open research are free: someone who understands what a number measures will use it better than someone who only knows where to click.
Does Quantika Research publish returns or trading track records?
No. No trading results are published as a sales argument and no return is promised. The research describes figures companies have already published and explains methods of analysis; it neither shows nor implies achievable results.
Is it related to other companies using the name «Quantika»?
No. Quantika Research (quantikaresearch.com) has no corporate, commercial or operational relationship with quantikalevels.com, quantika-ai.com, quantika.co or any other organisation with a similar name. The difference is not only one of ownership: Quantika Research does not sell trading signals, predictive indicators or execution bots.
Are calculations less accurate on the cheaper plans?
No. The accuracy of a calculation never depends on the plan. What changes between plans is how much you can do — how many lists, how many assets, which modules — and never the quality of the number. A calculation degraded by price would be a wrong figure, and a wrong figure is not worth having even cheaply.