Do brokers manipulate synthetic indices? What happens at Deriv (and what doesn’t)

In the world of online trading, there is a question that lurks in every forum, every comment section, and every trader’s mind after a bad streak: Is the game rigged?

Specifically, when it comes to synthetic indices—markets that exist entirely on our servers rather than on the floor of the NYSE or NASDAQ—the concern is natural. If we build the track, owning the cars and the finish line, what’s stopping us from moving the goalposts?

As the Head of Quants at Deriv, I spend my days dealing with probability, variance, and algorithms. Today, I want to step away from the code to address the issue of "Index Manipulation" head-on. I want to explain not just why we don't manipulate prices, but why our architecture makes it mathematically and technically impossible to target your specific trade.

The "broadcast" vs. "narrowcast" misconception

The most common myth I hear is: "The moment I placed my BUY order, the chart dipped just enough to hit my stop loss, then skyrocketed. The system was hunting me."

I understand the frustration. It feels personal. But from a quantitative engineering perspective, here is why that scenario is impossible.

Our synthetic indices (like volatility 75 or crash/boom) function as a global broadcast.

Think of it like a live football match on TV. Millions of people are watching the same goal happen at the exact same second. If the referee blows the whistle, he blows it for everyone. We cannot show you a version of the match where your team loses while showing your neighbour a version where they win.

Similarly, the price feed for Volatility 75 is generated by a single central algorithm and broadcast to all 2.5+ million Deriv clients simultaneously. If we manipulated the price to hunt your stop loss, we would have to manipulate it for every single trader around the world at that exact millisecond. Doing so would trigger thousands of arbitrage opportunities and anomalies that would collapse the mathematical integrity of the entire system immediately.

The engine: Cryptographically secure RNG

You might ask, "Okay, everyone sees the same price, but do you control that price?"

This is where the math comes in. Our indices are not drawn by a person; they are generated by code. Specifically, we use Cryptographically Secure Pseudo-Random Number Generators (CSPRNGs).

This is the same grade of technology used in banking security and blockchain encryption.

‍The Seed: The algorithm starts with a "seed"—a complex initial value.

‍The Generation: It produces a sequence of numbers that determines the next tick (up or down) and the magnitude of the move.

The Audit: These algorithms are not black boxes we hide in a basement. They are rigorously audited by independent third parties to ensure fairness.

Because the system is based on strict mathematical variance (volatility), it is blind to your account balance. It doesn't know if you are long or short. It only knows that it must generate a price that adheres to the statistical volatility of that specific index (e.g., 75% volatility).

The "stop hunt" illusion (and the reality of variance)

If the system isn't rigged, why does it feel like it is?

In quantitative finance, we call this Confirmation Bias meeting Market Microstructure.

In real markets (like forex), "stop hunting" is a real phenomenon where big institutional players push price toward liquidity pools (where retail traders pile their stop losses).

In synthetic indices, there are no "institutional banks" pushing the price. However, the math of the indices often simulates real-world market geometry. Prices naturally gravitate toward mean reversion or breakout points.

The Reality: You likely placed your stop loss at a very obvious technical level (e.g., just below a recent low).

The Probability: Because that is a statistical "extreme," the algorithm's variance often tests those boundaries before reverting.

The Result: It feels like a hunt, but it is actually just probability distribution playing out.

Why manipulation is bad business

Finally, let’s look at this from a business logic perspective.

Deriv has been around for 25 years. We process millions of transactions a day. Our business model is built on volume—we earn from the spread and the massive flow of activity.

If we were to manipulate prices:

We would lose our licenses: Regulators monitor our trade execution data. Anomalies are red flags.

We would destroy trust: In the age of social media, one proven instance of manipulation would end our company overnight.

It’s unnecessary: The "House Edge" in trading (the spread) is sufficient for a sustainable business. We don't need to cheat to win; we just need to provide the stadium for you to play in.

Don't just take my word for it

Trust in trading shouldn't be based on blind faith; it should be based on verifiable facts.

If you are interested in the deeper mechanics of how we ensure fairness, I invite you to read our order execution policy. It details exactly how your trade travels from your click to our server and back.

Read the order execution policy here.

Happy (and fair) trading!!