Lucas Liew Founder at AlgoTrading101

4 Quantitative Trading Strategies that Work in 2022

5 min read

Get 10-day Free Algo Trading Course


Last Updated on January 3, 2022

Poster: Quantitative Trading Strategies

The markets are getting more sophisticated and efficient. It is almost impossible to be profitable in the long-run by running strategies that used to work 10 years ago.

So, what trading strategies work today? Here are 4 categories of strategies that work (to some extent):

  1. Alternative Data
  2. Obscure and Small Markets
  3. High-Frequency Trading
  4. Machine Learning

Alternative Data Trading Strategies

Alternative data is non-traditional (not price or volume) data that has predictive value in the financial markets.

Examples of Alternative Data Strategies

  • Satellite images of Walmart parking lots allow hedge fund to analyse the change in the number of people going to Walmart. This enables them to predict Walmart’s sales figures.
  • Surveyors record the number of trucks leaving Company A’s factories. This data allows traders to predict Company A’s revenue.
  • Having social media foot traffic data (such as Foursquare’s check-in data) around Restaurant X’s outlets allows traders to predict Restaurant X’s sales figures.

The 5 popular types of alternative data are:

  • Location Data
  • Consumer Expenditure Data
  • Satellite/Drone Imagery
  • Weather Data
  • Web-scrapped Data

As tradition trading opportunities decreases, traders need information that can put them one step ahead of the competition.

They need to be creative with their information sources. Not only do they need quality alternative data, they need quality alternative data they other traders don’t have access to.

Traders and hedge funds either buy these data from data providers or collect them themselves.

Here is a list of alternative data vendors.

Price of data + Workarounds for retail traders

Note that these data can be expensive. Fresh and exclusive data with good predictive value are even pricier.

It is more feasible for retail traders to collect or scrape data off the web themselves.

They can then use these datasets on their own or mix them with some other bought/scrapped data to create a synthetic index.

Example of a Synthetic index: MongoDB’s popularity – Historical Trend

Obscure and Small Markets

Obscure markets refer to markets which are less popular and regulated.

Small markets refer to markets that can only absorb a small amount of trading volume without a large price movement.

Big funds can only put their capital in certain regulated and large markets.

We can trade any products:

  • Cryptocurrencies? Go for it.
  • Stock market in developing countries? Go for it.
  • Unregulated derivatives? Go for it.
  • Price discrepancy due to geopolitical reasons but you can trade them because you know a trusted local guy? Go for it.
  • Penny stock that can only absorb $50K a day without its price spiking? Go for it.
  • Weird commodity markets in Asia? Yup… go for it!
  • Horse racing quant trading? Someone made a billion dollars doing that! (Bloomberg article)

The reason to trade less regulated and small markets is that those markets are less efficient. There are more opportunities to be made.

I’ve personally seen obvious opportunities on less popular markets. Those opportunities provide consistent profits (almost every month was profitable) for years.

But once those markets get more popular and other big players come in, the market behaviour changes and opportunities get eroded significantly.

» Calendar spreading is an example of a semi-obscure strategy. Learn how to execute it from our “5 Futures Trading Strategies Guide“.

High-Frequency Trading (HFT)

High-frequency trading describes trading that require high computing and communication speeds.

HFT is characterized by high communication and computing speed, large number of trades, low profit per trade and expensive software infrastructure.

High-frequency traders use communication speed to profit and outwit other traders.

High Frequency Strategy Types


Main article: Arbitrage

Arbitrage trades happen when an asset is priced differently on 2 exchanges and a trader buys the cheaper one while shorting the pricier one.

Reaction to news

When a major news is released, the trader who reacts the fastest wins. In this case, the high-frequency trader needs to analyse the news and fire the trade before everyone else

Latency Arbitrage

When a traditional (slower) hedge fund buys a large amount of Stock A, a HFT hedge fund will detect that.

The HFT hedge fund will then buy all the Stock A on the other exchanges and sell it back to the slower hedge fund for a small profit.

The HFT hedge fund might do this millions of times over a day.

Statistical Arbitrage

A large number of similar stocks might move in a similar manner. When any of the stocks diverge, the high-frequency trader will buy the cheaper one and/or short the pricier one.

Index Arbitrage

An index or exchange-traded fund is designed to track the returns of an index such as the S&P500.

Other strategies

HFT is a secretive field. Once a strategy is revealed and the other funds join in, the profit opportunity disappears fast.

Thus, many new innovative strategies are created everyday and are not known to the general public.

Investment in infrastructure

HFT is usually a winner-take-all industry. If you are faster than your competition (even by a slight amount), you get all the profits.

Since relative speed is more important than absolute speed, HFT funds constantly try to be faster than their rivals.

HFT funds spend hundreds of millions on hardware and software infrastructure to reduce their computing and communication speed by the milliseconds.

Investments in infrastructure includes building a straight tunnel to lay communication lines and putting their servers right beside the financial exchange’s servers.

Machine Learning

Machine learning techniques enable computers to do things without being told explicitly how to do them.

The essence of machine learning is the ability for computers to learn by analysing data or through its own experience.

Traditional Computing Rules

If an image has 4 legs, fur, pointy ears and whiskers, label it as a cat.

Machine Learning Rules

We give the computer 1000 cat pictures and 1000 pictures that are not cats. After analyzing these 2000 pictures, the computer will be able to tell if a picture contains a cat.

Advantages of Machine Learning

  • Being able to analyse large quantities of data without being explicitly told what to look for
  • Being able to understand texts (in large quantities and different languages)
  • Being able to interpret images
  • Being able to come up with creative solutions
  • Being able to analyse and output a prediction fast

Examples of Machine Learning Trading Strategies

  • Reading texts fast. So that we can quickly know how a newly published news article affects the market.
  • Reading huge chunks of texts. So that we can get summaries effectively.
  • Looking at many drone and satellite images. So that we can know what the images are telling us. Are the soybean crops dying or booming? Then we’ll buy or short soybeans!
  • Scanning the many orders coming into the market. We are looking for patterns to see if someone is trying to buy or sell a large quantity of Apple stock.

More about machine learning here: Machine Learning Simplified

Does it mean we will be profitable running those strategies?

Unfortunately, most probably not. Those strategies work, but executing them is not straight forward.

Just because many burger restaurants are successful, doesn’t mean that you will be able run a successful burger joint with ease.

Similarly, just because there are top traders and funds running the above trading strategies successfully doesn’t mean that we can run those strategies with ease.

To run those strategies well, you need to put in the hard work.

Try, fail, improve, fail again, improve, repeat until successful.

What trading strategies are suitable for beginners

Start with trading strategies involving 1) alternative data that can be obtained via web scraping or cheaply from vendors and 2) obscure and small markets.

High-frequency trading involves millions of dollars of infrastructure and a team of PhDs so that’s out of the question.

Machine learning is a tool to analyse information, it is not a starting point.

Related Questions

Does price action work? If you are just analysing the price of one asset without any information from other assets or external variables, it is difficult to be profitable in the long run. It is worse if you are trading an efficient market like Forex. Any positive returns in the short term is likely luck. Drawing 20 trendlines and overlaying 10 indicators will not save you.

Does trend following work? The opportunities in trend following has greatly diminished since the days of the Turtle Traders in the 1980s. However, trend following could still work if, in addition to just being a price breakout strategy, it is complemented by good money management, risk reduction (by having opposing trades hedge one another), and quality information sources (quantitative and qualitative research).

Lucas Liew
Lucas Liew Founder at AlgoTrading101