I'm sure some HFTs engage in event-driven strategies like this, but it's not their bread and butter. Most HFT strategies can be thought of almost like a lubrica

I'm sure some HFTs engage in event-driven strategies like this, but it's not their bread and butter. Most HFT strategies can be thought of almost like a lubricant that makes markets operate more smoothly and efficiently by moving liquidity to where it's needed. Market makers move liquidity across time, matching up buyers and sellers who arrive at different rates and possibly on different exchanges. They do this by posting two-way prices and managing their inventory risk. Since placing limit orders gives a free option to other more informed traders, HFT MMs must be very skilled at predicting short-term price movements and adjusting their quotes quickly based on these predictions. They attempt to profit on the difference in their buy and sell prices net of fees. The market maker is buying when potentially informed sellers enter the market and selling when buyers enter the market, and prices often move against them. In reality if they are good at predicting prices and managing risk they will capture a fraction of the spread. This activity dampens volatility since there are more ready buyers and sellers in the market which keeps prices close to fair value rather than moving abruptly simply because of a short-term imbalance in the number of buyers/sellers. For a counter example of this look at the Bitcoin market which is relatively illiquid and large orders can move the price substantially. Arbitrage is another common HFT strategy where they are moving liquidity and price discovery from one product to another. An example would be trading an ETF vs. the basket of stocks it contains. If the basket is cheap relative to the ETF, the HFT firm will buy the basket of stocks and sell the ETF and realize a profit by either unwinding the trade or creating/redeeming the ETF/basket. This activity keeps prices fair so when you buy shares in an ETF they have very little tracking error relative to the underlying basket of stocks. Because computers can trade for almost no margin, trades like this are extremely competitive to the point where most major ETFs essentially never trade at a discount or premium. The reason market-making HFT is especially beneficial is because it can operate at such low margins. Human market makers had to charge wide bid-ask spreads because one floor trader could only manage a handful of products at once so they need to make more money in each product. In today's world a small team could run automated market-making on a thousand symbols or more. They're happy making a penny spread or even less since they cast such a wide net. Financial intermediaries have always existed since buyers and sellers rarely arrive at the exact same time and place. Automating this process is far more efficient than doing it by hand. Was there so much moral outrage and misunderstanding when manned telephone switchboards were replaced by computers? There are real issues in the US market structure (internalization/payment for order flow, poor tick size choices that encourage penny jumping in high-priced stocks and jockeying for queue position in cheap ones, excessive number of exchanges some created just to offer different fee structures, tiered rebates, etc.) but HFT generally has positive effects on market quality.