Many different algorithms have been developed to implement different trading strategies. Much early algo trading was developed for the buy side in order to reduce transactions costs. Recently, high frequency trading, which is generally a type of market making by sell side traders, has become more prominent and controversial. These algorithms or techniques are commonly given names such as "Iceberg", "Dagger", "Guerrilla", "Sniper" and "Sniffer".

Transaction cost reduction

Large orders are broken down into several smaller orders and entered into the market over time. This basic strategy is called "iceberging". The success of this strategy may be measured by the average purchase price against the VWAP for the market over that time period. One algorithm designed to find hidden orders or icebergs is called "Guerrilla".

Market making and high frequency trading

Market making involves placing a limit order to sell (or offer) above the current market price or a buy limit order (or bid) below the current price in order to benefit from the bid-ask spread. Automated Trading Desk, which was bought by Citigroup in July 2007, has been an active market maker, accounting for about 6% of total volume on both NASDAQ and the New York Stock Exchange. High frequency traders use computers that execute trades within milliseconds, or "with extremely low latency" in the jargon of the trade. In the U.S., high-frequency trading firms represent 2.0% of the approximately 20,000 firms operating today, but account for 73.0% of all equity trading volume. As of the first quarter in 2009, total assets under management for hedge funds with high frequency trading strategies were $141 billion, down about 21% from their high. The high frequency strategy was first made successful by Renaissance Technologies. High frequency funds started to become especially popular in 2007 and 2008.Many high frequency firms say they are market makers and that the liquidity they add to the market has lowered volatility and helped narrow spreads, but unlike traditional market makers, such as specialists on the New York Stock Exchange, they have few or no regulatory requirements.

These funds are highly dependent on ultra-low latency networks. They profit by providing information, such as competing bids and offers, to their algorithms microseconds faster than their competitors. The revolutionary advance in speed has led to the need for firms to have a real-time, co-located trading platform in order to benefit from implementing high frequency strategies. Strategies are constantly altered to reflect the subtle changes in the market as well as to combat the threat of the strategy being reverse engineered by competitors. There is also a very strong pressure to continuously add features or improvements to a particular algorithm, such as client specific modifications and various performance enhancing changes (regarding benchmark trading performance, cost reduction for the trading firm or a range of other implementations). This is due to the evolutionary nature of algorithmic trading strategies - they must be able to adapt and trade intelligently, regardless of market conditions, which involves being flexible enough to withstand a vast array of market scenarios. As a result, a significant proportion of net revenue from firms is spent on the R&D of these autonomous trading systems.

Arbitrage

A classical arbitrage strategy might involve three or four securities such as covered interest rate parity in the foreign exchange market which gives a relation between the prices of a domestic bond, a bond denominated in a foreign currency, the spot price of the currency, and the price of a forward contract on the currency. If the market prices are sufficiently different from those implied in the model to cover transactions cost then four transactions can be made to guarantee a risk-free profit. Algorithmic trading allows similar arbitrages using models of greater complexity involving many more than 4 securities. The TABB Group estimates that annual aggregate profits of low latency arbitrage strategies currently exceed US$21 billion.

More complicated strategies

A "benchmarking" algorithm is used by traders attempting to mimic an index's return.
Any type of algo trading which depends on the programming skills of other algo traders is called "gaming". Dark pools are alternative electronic stock exchanges where trading takes place anonymously, with most orders hidden or "iceberged." Gamers or "sharks" sniff out large orders by "pinging" small market orders to buy and sell. When several small orders are filled the sharks may have discovered the presence of a large iceberged order. They then front run the order. "They look for big, dumb elephants leaving big footprints," said Joe Saluzzi, head of equity trading at Themis Trading.... "If you're getting tapped by odd lots, if it happens 40 times...you're being gamed."

Any sort of pattern recognition or predictive model can be used to initiate algo trading. Neural networks and genetic programming have been used to create these models. “Now it’s an arms race,” said Andrew Lo, director of the Massachusetts Institute of Technology’s Laboratory for Financial Engineering. “Everyone is building more sophisticated algorithms, and the more competition exists, the smaller the profits."

The arms race has allegedly included stealing computer code. UBS has sued three of its former traders and Jefferies & Company for stealing algorithmic trading programs.