What exactly is order flow? Is it time-and-sales information?

H

helpme_please

Often, I hear from professional traders that trading according to the order flow is critical to their edge. What exactly is order flow? Is it something that only institutional brokers who can see the client orders coming in can execute? Is order flow an impossible strategy to smaller retail investors of Interactive Brokers? Is order flow referring to the time-and-sales information which tape readers use?

Can the professionals here shed some light?

Thank you.
 
helpme_please said:
Often, I hear from professional traders that trading according to the order flow is critical to their edge. What exactly is order flow? Is it something that only institutional brokers who can see the client orders coming in can execute? Is order flow an impossible strategy to smaller retail investors of Interactive Brokers? Is order flow referring to the time-and-sales information which tape readers use?

Can the professionals here shed some light?

Thank you.
More...

The price of a security is determined by demand and supply. Order flow is a way to quantify the direction and the magnitude of the changes in the demand and supply. Order flow can be determined from the exchange limit order book, to which you can have access with the so-called "market depth" data subscription, sometimes also referred to as the L2 (level 2) market data feed.

You don't have to be an institutional trader to have access to market depth. For example, with Interactive Brokers, you can have market depth for as little as $20 per month. However, a high quality, institution-oriented market depth feed would cost thousands of dollars per month.

Once you have the subscription, the challenge is to interpret the complex dynamics taking place in the order book. For some liquid instruments (such as the ES), the order book changes hundreds of times per second, so any kind of eye-balling the book is almost certainly a futile exercise. There are some tools (both open source and commercial) for analyzing the order book with the purpose of predicting the order flow. The most recognizable players in this field are probably Jigsaw and BookMap, so look them up and see what they do.

There is also a challenge of dealing with the historical order book data, if you want to backtest against it. The data sets are massive (think in terms of gigabytes per day, per symbol), and are very expensive to get.

There are people who will opine that the L2 data is useless, and there are people who will say that they make a living from trading the order flow. Just like with anything else, you would need to research, test, formulate your own ideas, and make your own conclusion about it. This would probably take you about 10,000 hours of dedicated work.
 
Last edited:
Thanks for the reply. From your reply, I think using order flow is not suitable for stock investors who have a diversified portfolio of stocks. It is too time-consuming to do this for a portfolio of >30 stocks. It seems to be more suitable for forex and stock market indices as there are fewer to choose from and one can specialise in a particularly security, say, USDEUR or Nasdaq100 futures contract.

nonlinear5 said:
The price of a security is determined by demand and supply. Order flow is a way to quantify the direction and the magnitude of the changes in the demand and supply. Order flow can be determined from the exchange limit order book, to which you can have access with the so-called "market depth" data subscription, sometimes also referred to as the L2 (level 2) market data feed.

You don't have to be an institutional trader to have access to market depth. For example, with Interactive Brokers, you can have market depth for as little as $20 per month. However, a high quality, institution-oriented market depth feed would cost thousands of dollars per month.

Once you have the subscription, the challenge is to interpret the complex dynamics taking place in the order book. For some liquid instruments (such as the ES), the order book changes hundreds of times per second, so any kind of eye-balling the book is almost certainly a futile exercise. There are some tools (both open source and commercial) for analyzing the order book with the purpose of predicting the order flow. The most recognizable players in this field are probably Jigsaw and BookMap, so look them up and see what they do.

There is also a challenge of dealing with the historical order book data, if you want to backtest against it. The data sets are massive (think in terms of gigabytes per day, per symbol), and are very expensive to get.

There are people who will opine that the L2 data is useless, and there are people who will say that they make a living from trading the order flow. Just like with anything else, you would need to research, test, formulate your own ideas, and make your own conclusion about it. This would probably take you about 10,000 hours of dedicated work.
More...
 
Order flow is generally referred to customer orders that a broker can trade against or take the other side of your trade. The reality is that without market makers making two sided markets, there are not enough bids and offers from the "public" all the time to provide depth and liquidity. That is more evident in derivatives. Capturing order flow provides those MM the opportunity to make a profit.
 
Robert Morse said:
Order flow is generally referred to customer orders that a broker can trade against or take the other side of your trade. The reality is that without market makers making two sided markets, there are not enough bids and offers from the "public" all the time to provide depth and liquidity. That is more evident in derivatives. Capturing order flow provides those MM the opportunity to make a profit.
More...

Yet, MMs fail most of the time."Paradox", one may say...
 
You must log in or register to reply here.
Back
Top