Blackboxes/Algorithms Nothing to Fear

BellaGeneral Comments, Mike Bellafiore's (Bella's) BlogsLeave a Comment

I received a GChat message from a future trader, Carter, who was interested in the impact of blackboxes or algorithms on intraday trading. I have heard Carter’s concerns from other future traders. Some believe that blackboxes will rule trading and eliminate the intraday trader. A few thoughts:

1) I have heard this argument for the past 5 years and I am still trading profitably. And our desk just had its best week ever.

2) Blackboxes have made intraday trading more difficult, which is part of the reason why it is imperative in this market to be trained well (think SMB Training,

3) You must be more selective with your trading. But if you are then there are still many plays where you will have a trading edge, for example a momentum trade in a Stock in Play. If you are not selective then the blackboxes will certainly take your money.

4) A higher percentage of blackboxes fail than traders.

5) Too many blackboxes are written by programmers who have not learned how to trade and thus are destined to fail. You can’t just crunch numbers and expect to be a profitable trader. It is not that simple. You must always be able to sense the human emotion in the market. And understand that past data may not help you in a current market.

6) You must be able to identify the different algorithmic programs so that you can trade against them. This takes some experience, good mentoring, and practice.

7) I have seen many blackboxes get destroyed and many that certainly will. The most disappointing blackbox program today is what we call the New Low Buy Program. Actually, I have a fundamental problem with this blackbox as it disobeys fundamental trading principles (don’t fade a weak stock!). A stock makes a new intraday low, so then many day traders get short and play the downside momentum. The program buys the shorts from the daytraders, then pushes the stock higher causing all of the daytraders to cover. The problem is that this program was turned on often in FNM and LEH and FRE during their recent collapse. The stocks were broken technically. And not only did daytraders hit the new intraday low but so did huge institutional sellers. Thus these programs were obliterated.

8) Programs should be respected but not feared.

9) Anyone who has created a profitable blackbox is a genius. It is really that difficult to build one.

10) Blackboxes have to be turned on and off just like bids need to be placed and withdrawn. And what works today may not work tomorrow. And because of this blackboxes can never completely rule trading. There will always be the need for the trader who understands when he has an edge to buy or sell. And the market is always changing such that it is impossible to program all of the different variables that eliminates the need for trader discretion.

Blackboxes will continue to become more prevalent. I have the utmost respect for those who have created successful ones. And there is a fortune to be made in creating a blackbox. But there isn’t a program that can be developed to trade against the well trained and disciplined trader. There are just too many variables. And I spend my day only making trades where I have an edge. If I recognize there are too many programs in a stock in order to gain an edge then I move on. And I find a trade where I will have an edge. But as some on my desk joke, I guess I’m just a human blackbox.

Leave a Reply