In late 2017, as Bitcoin’s price surged toward $10,000, I developed a trading algorithm that executed approximately $50 million in market making transactions over several months. This article shares the story of how I entered market making, the theory behind the algorithm, and the intensifying competition among automated traders.
Understanding the Basics: Bid-Ask Spreads
When most people think of trading assets like stocks, they assume there’s a single fixed price. In reality, every traded asset has two distinct prices:
- The bid price: the highest price a buyer is willing to pay.
- The ask price: the lowest price a seller is willing to accept.
The difference between these two prices is known as the spread.
For highly liquid stocks like Apple, the spread is often just a few cents. But in December 2017, Bitcoin’s spread could be as high as $100 while trading around $10,000—a 1% spread compared to Apple’s 0.02%.
Large spreads may seem problematic, but they also present profit opportunities. If you could buy Bitcoin at the bid price and immediately sell at the ask price, you would capture the spread as profit. But there’s a catch: transaction fees and order types.
The Role of Maker and Taker Orders
To understand how market makers profit, you must understand two types of orders:
- Maker orders: These are limit orders that provide liquidity to the market. They sit on the order book until matched with a taker order.
- Taker orders: These are market orders that remove liquidity by immediately matching with existing maker orders.
Exchanges typically charge higher fees for taker orders to incentivize liquidity provision. Some exchanges even offer zero fees for maker orders.
This fee structure is crucial. If you use a taker order to buy and another to sell, you’d likely lose money due to fees and the spread. But if you place maker orders on both sides, you can potentially capture the spread risk-free—assuming you can execute at favorable prices.
How Market Making Bots Work
The core idea is simple:
- Place a buy order slightly above the current bid price.
- Place a sell order slightly below the current ask price.
- Capture the spread when both orders are filled.
But execution is challenging. Between querying the current bid price and submitting your order, the market may have moved. To ensure your buy order remains competitive, you must place it a few cents above the current bid.
Similarly, on the sell side, you must place your order just below the current ask price. This ensures that incoming taker orders match with your maker orders.
Profit per trade is roughly:
[
\text{Profit} = (\text{Ask Price} - \text{Bid Price}) \times \text{Quantity Traded}
]
This profit accumulates over hundreds or thousands of trades daily.
Growing Competition Among Bots
Whenever a profitable opportunity arises, traders flock to it. By December 2017, competition among market-making bots was fierce. Bid prices changed multiple times per second, often by just one or two cents, as algorithms competed to offer the best prices.
Initially, my bot placed buy orders one cent above the current bid. But by the time the order was processed, the bid price had often increased, leaving my order uncompetitive.
To solve this, I measured how much the bid price changed every 100 milliseconds. This allowed the bot to predict market movement and place orders sufficiently above the bid to stay competitive.
I also incorporated spread size into the algorithm. When spreads were large, the bot would place orders farther above the bid to ensure execution. When spreads were narrow, it would be more conservative.
Surprisingly, competition was fiercer on the buy side than the sell side during this period. This was likely due to Bitcoin’s supply shortage and overwhelming buyer demand.
Market Manipulation and Spoofing
Cryptocurrency markets are largely unregulated, making them ripe for manipulation. One common tactic is spoofing: placing fake orders to create misleading market conditions.
For example, a manipulator might place a buy order for 0.00001 BTC at $9,999.99 when the ask price is $10,000. This narrows the spread to just one cent, making market making unprofitable. Other bots might shut down, assuming no opportunity exists.
But these spoofing orders are usually small. When a large sell taker order arrives, the first 0.00001 BTC is sold to the spoofer. The remaining quantity is sold to the next highest bid, which could be much lower.
To counter this, I modified my bot to ignore spoofing orders when calculating the spread. If the real spread was still profitable, the bot would compete for genuine orders behind the spoofed ones.
👉 Explore advanced trading strategies
The Boom and Bust of Early 2018
January and February 2018 were highly volatile. After Bitcoin peaked near $20,000 in December, prices began falling. But high trading volumes created massive opportunities for market makers.
However, this window was short-lived. By March, volumes dried up, and profits dwindled. Increased competition narrowed spreads, and more bots meant shorter windows for order placement.
I shut down my bots in March 2018 and never restarted them. Many exchanges raised fees for market makers, further reducing profitability. But for those who understand the mechanics, opportunities still arise during volatile periods.
Over three months, my algorithm traded $50 million worth of Bitcoin and achieved a 20.2% return on investment.
Frequently Asked Questions
What is crypto market making?
Market making involves placing both buy and sell orders for an asset to profit from the bid-ask spread. Market makers provide liquidity and earn small profits on each trade.
How do market makers make money?
They capture the difference between the bid and ask prices. By placing orders on both sides of the market, they earn the spread repeatedly throughout the day.
Is market making risk-free?
No. Market makers face risks like price volatility, order execution delays, and increasing competition. Proper risk management and algorithm adjustments are essential.
Can individuals become market makers?
Yes, but it requires programming skills, market knowledge, and access to exchange APIs. Competition from institutional players is intense.
What are maker and taker fees?
Maker fees are charged for providing liquidity (often low or zero). Taker fees are charged for removing liquidity (usually higher).
How has market making evolved since 2018?
Competition has increased, spreads have narrowed, and many exchanges now offer incentives for professional market makers. Algorithmic sophistication is now critical.