Synthetic Indices Explained: How Broker Priced Markets Work

A synthetic index looks like a chart, moves like a chart, and answers to none of the things that move a chart. The price series is produced by software the broker runs, so there is no order flow behind a candle and no outside price to check a print against.

That single fact decides almost everything else worth knowing: which analysis still applies, who sets the cost, and which firms are allowed to sell you one.

What a Synthetic Index Actually Is

The instrument you trade is a contract for difference whose reference value is a number the broker publishes. Nothing is bought, nothing settles, and the reference is not a claim on anything outside the broker’s own system.

Compare that with a stock index, where the level is computed from the traded prices of listed companies by an index provider who publishes the method. Two brokers quoting the FTSE 100 are quoting the same underlying object, and either can be measured against it.

With a synthetic index there is no such object. The name may borrow the vocabulary of volatility, and the chart may carry the same candles, but the series exists only inside the firm that created it.

Where the Price Comes From, and Who Generates It

Deriv, which publishes the best known range of these instruments, states on its synthetic indices page that a random number generator of cryptographic strength produces the series, and that the charts imitate the behaviour of markets while nothing in the news or in live market conditions reaches them. That is the mechanism, described by the firm that operates it.

Read carefully, it says two things. The series is random by design rather than by accident, and the randomness is the product, not a defect in a data feed.

The second consequence is structural. In a real market the broker is either passing your order to someone or taking the other side of it, a distinction our page on broker as counterparty sets out. Here that distinction collapses, because the firm taking the other side is also the firm generating the number the contract settles against.

Nothing in that arrangement is hidden. It is published on the product page. What it removes is the appeal: there is no exchange print, no third-party feed and no reference rate to point at when a fill is disputed.

It is worth being precise about what the cryptographic wording settles. A generator of that strength is one whose next value cannot be worked out from the values already published, which is a claim about the sequence and nothing else. It says nothing about the spread quoted around that sequence, the margin required against it or the swap charged to hold it, and those are separate decisions made by the same firm.

The Number in the Instrument Name Is a Setting, Not a Measurement

Instruments in this family carry a number: a volatility index at 10, 25, 50, 75 or 100, a boom or crash index at 300, 500 or 1000. Deriv presents the volatility figures as a choice offered to the trader, and describes its jump indices as leaping roughly every twenty minutes on average, at around thirty times the normal volatility of the index.

An average interval and a chosen volatility percentage are parameters. They describe how the generator was configured, and they were true before a single tick existed.

That is the opposite of what a volatility figure normally means. The volatility index most traders know is computed from live option prices and can only be read after the fact. One is an output of a market; the other is an input to a program.

The practical effect is that a rising figure tells you nothing has changed. There is no regime to shift, because the setting is the regime.

Which Regulated Entity Can Offer These, and Which Cannot

On its regulatory information page, checked on 12 August 2026, Deriv lists subsidiaries licensed in the British Virgin Islands, the Cayman Islands, Mauritius, Vanuatu, Samoa, Saint Vincent and the Grenadines, and Labuan, with a Guernsey holding company. No company holding an EU or United Kingdom authorisation appears anywhere on that list.

That is not an accusation, and offshore licensing has its own obligations, which our page on offshore licence obligations covers. It is a fact about where the product is sold from, and the arithmetic behind it is visible in the leverage.

ESMA’s product intervention put a ceiling on retail leverage for contracts for difference, running from 30:1 at the most liquid end of the ladder down to 2:1 at the least, with the rung decided by what the contract refers to. The whole ladder is set out on our page about tiered leverage.

The FCA adopted the same range for retail clients in the United Kingdom, and added a close-out at 50 per cent of required margin and a floor that stops losses at the money in the account.

What matters here is that no rung on that ladder describes an index a broker generates for itself. Such an instrument falls to the residual rung, the one covering other reference values, where the ceiling is 5:1. Set against it, the specification Deriv publishes shows maximum effective leverage of 1:600 on its Boom 1000 Index and 1:4000 on its Boom 50 Index.

The restriction that applies to a general CFD is set out on our contract for difference page. What matters here is narrower: these instruments are sold on terms that no EU or UK retail permission allows, which is why the entity offering them is registered where it is.

What Stops Working: Calendar, Sessions and Correlation

Three familiar inputs disappear at once, and none of them degrades gracefully.

The economic calendar has nothing to release. A generated series has no central bank, no inflation print and no employment report, so a method timed around scheduled events has no events to time around.

Session structure goes the same way. Liquidity does not thicken when London opens because there is no liquidity in the ordinary sense, only a quote the broker publishes, so the session hours that shape a currency pair are simply absent.

Correlation is the third. Two generated series share no economy and no flow, so the diversification and hedging arguments that rest on currency correlation have no basis, in either direction. A position in one says nothing about a position in another.

Volume and market depth belong on the same list. A depth ladder shows resting orders from other participants, and a generated series has no participants, so a depth window on one of these instruments is displaying the broker’s own quote rather than a book. Any method that reads size at a level is reading a presentation, not a queue.

What survives is anything defined purely on the price series itself: a moving average, a range, a breakout level. Whether such a rule is worth running is a separate question, and the honest answer is that it is being tested against a process whose parameters are published rather than discovered.

Costs Are Set by the Firm That Sets the Price

On a currency pair a spread can be compared with an interbank reference, and a swap traces to two policy rates. On a generated index neither comparison exists, so the cost is whatever the specification says.

The figures below are taken from the Deriv trading specification page on 12 August 2026, for two instruments on the same platform at the same firm.

Two rows carry most of the meaning. The currency pair charges one direction and credits the other, because a swap on a real pair reflects an interest differential that has a sign; the generated index charges 18 points whichever way the position is held, because there is no differential to reflect.

The hours row is the other. A pair closes because the market it belongs to closes. A generated series never closes, and the same specification shows margin ranging from 0.03 per cent to 1.00 per cent across instruments in this family, so the terms are set instrument by instrument rather than by any external convention.

A Decision Framework: When These Instruments Fit and When They Do Not

The question is not whether the product is legitimate. It is published, specified and openly described. The question is whether what you intend to do with it survives the way it is built.

Start with your method. If it reads economic releases, waits for a session, or sizes across correlated positions, none of its inputs exist here and it should not be transplanted. If it is defined entirely on the series in front of it, it can at least be stated.

Then look at what a disagreement would look like. There is no exchange record behind a spike, so a dispute about a fill has only the broker’s own logs on both sides of it.

Then read the leverage as information rather than as an offer. A figure well beyond any EU or UK retail permission tells you which regime the account sits under and which protections, including negative balance protection and the 50 per cent close-out rule, are not automatically part of it.

Those two protections are worth naming separately, because they do different jobs. The close-out rule fixes the point at which positions must be shut, so it limits how far an account can run down before the broker acts.

Negative balance protection limits the outcome after that point, guaranteeing the loss stops at the funds in the account. Where neither is mandated, both become terms in a contract rather than obligations on the firm, and they belong in the account documents a trader reads before opening a position.

Finally, treat weekend availability as neutral. It is convenient, and it is the arithmetic result of there being nothing to close, not evidence that the instrument is more liquid or more accessible than a market.

Who This Is Not For

Anyone whose analysis begins with why a price moved will find nothing to work with, because the answer is always the same and it is not about the world.

Anyone who needs the protections attached to an EU or UK retail permission should check which entity the account is opened with before anything else, since that is settled by the licence and not by the platform.

And anyone comparing costs across brokers should note that there is no shared underlying to compare on. Two firms quoting a volatility index are quoting two different products with the same style of name.

Frequently Asked Questions

What is a synthetic index in trading?

It is a contract for difference on a price series that the broker generates with software rather than collecting from a market. No asset stands behind the number, and the series exists only inside the systems of the firm that publishes it.

Who sets the price of a synthetic index?

The broker does. Deriv states that a random number generator of cryptographic strength produces its synthetic index series, so the same firm creates the reference number, quotes the spread around it and takes the other side of the position.

Can synthetic indices be traded at the weekend?

Yes. The trading specification of Deriv shows hours of Sunday 00:00 to Saturday 24:00 GMT on its Boom indices, against Sunday 21:10 to Friday 20:55 on AUD/USD. Nothing closes because no market is open in the first place.

Are synthetic indices offered by brokers regulated in the EU or the UK?

The regulatory page of Deriv, checked on 12 August 2026, lists companies licensed in the British Virgin Islands, the Cayman Islands, Mauritius, Vanuatu, Samoa, Saint Vincent and the Grenadines, and Labuan, and not one holding an EU or United Kingdom authorisation. Leverage of 1:600 sits far outside the range those regulators permit for retail clients.

Does technical analysis apply to synthetic indices?

Any rule defined on the price series alone can be calculated, since the candles are ordinary candles. Anything that reads news, sessions, volume or correlation has no input here, because a generated series carries none of those things.