Market making strategy

Automated Market Maker

Market making strategy is an automated investment algorithm that is used to provide liquidity, by filling up the order book with buy and sell orders, so that other market participants, buyers and sellers alike, could execute their orders whenever they need to.

So market maker plays a special role in the financial ecosystem providing price discovery and building trust in the market.

From a technical point of view, market making is like arbitrage or triangular arbitrage, a high frequency trading (HFT) strategy.

Goals of a market maker

Market makers have duties on providing liquidity, which is understood as:

  • Being present with their orders no matter the market conditions, allowing other market participants to trade 
  • Making the depth of the market, 
  • Making the bid-ask spread tight
  • Reduce the costs of other investors to trade

Market making algorithm

The market making strategies constantly quotes buy and sell orders on both sides of the order book with a defined bid-ask spread. Being always present on the market creates many risks that impose many additional requirements on market making strategies that distinguish them from other strategies:

  • They are super fast to react quickly to changes in the market before other professional traders will.
  • They are fully automated.
  • They need to minimize the risk of adverse price movement.

Different types of marketing strategies:

  • Remarketing, or liquidity mirroring
  • Market making with or without hedge
  • Spot vs futures market making – benchmarking and hedging to a futures market, like perpetuals for the same instrument

Features, parameters

For market making strategy to function in the market, it usually needs parameters that define its behavior in the market. Some basic examples of parameters include:

  • Benchmark instrument
  • Hedging instrument
  • Quoted spread, along with minimum and maximum values
  • Order size
  • Order levels – how many bid and ask orders on different price levels will be populated in the order book
  • Price intervals – between order levels – another parameter defining the order ladder
  • Position limit – risk management
  • Minimum and maximum quoted price levels – safety checks for bad ticks or extreme volatility possible in the events like large liquidations. Any safety parameters are always welcome in the market making operations
  • Maximum drawdown – another safety check defining the maximum loss that can be incurred by a strategy

What else is needed to perform market making operations?

So we have a basic algorithm, now what about its execution? 

Software and hardware infrastructure is a big topic due to the requirements layout at the beginning of the article. Market making is basically a low-latency strategy. If you want to be 100% time on the market, you have to be extremely fast to not be exploited by other market participants that have information or time advantage. 

We invested a lot in building a software platform to execute algorithms, basically an operating system, that also covers connectivity to different exchanges, and translates all the messages and events into a common format understandable to all our algorithms.

Now all our infrastructure runs on over 80 production servers (and an additional 50 for test and staging purposes), and we are not the biggest ones in the industry. So yes, we could say that market making is mainly a software business and we are all engineers.

Can I use ready-made market making bots?

To play or test, yes. But for production operations of providing liquidity? For all the reasons explained in the last point, definitely not. They are too slow and too immature from a technological perspective, and you will just slowly bleed out of capital. 

Is market making risk free?

Definitely not. They must sell when people want to buy and prices rise. They must buy when people want to sell and prices decline. Market makers carry therefore an inventory risk. 

How do market makers make money?

Basically on the quoted spread, minus costs of hedging and adverse price moments on the inventory.

Can you beat the market maker strategy?

Yes, but rather not in low latency game. Market makers have to be present all the time on the market. You don’t. Market makers have duties on providing liquidity, making the depth of the market, and making the spread tight which carries some risks, described before. If beating a market maker in the order book for some reason would be a goal, I would try to exploit these properties.

Crypto market making strategy

Does it differ? Yes, Market making on crypto exchanges is a bit different. Markets are fragmented, there are many exchanges, where the price is created, and there is no one dominant exchange. And what’s new is the whole DeFi space. Decentralized exchanges do not have order books, they operate with the Automated Market Maker and Liquidity Pools concepts, which adds one more level of complexity to this topic.

Disclaimer

We are market makers. We have built software for market making for over ten years on both traditional financial markets as well as in the crypto space, which is now our focus. Take a look here to see more about how we provide liquidity 

Related resources:

For those feeling hungry for some theoretical background, a short list of books on market making.

Learn more on High Frequency Trading here

Check similar strategies to market making like arbitrage, and triangular arbitrage, or check our strategy catalog for a thorough description of different execution strategies applied by hedge funds