Reader Question: Best Way to Trade Intraday

Reader Andrew asks (paraphrased question):

Can I first of all say that I recently discovered this website and your book and I think they are fantastic! I would like to congratulate you on a great job with both of them. I have only read the first few pages of your book but I am looking forward to working my way through the book and website together. So congratulations on a job well done.
I recently completed the […] course. I was wondering if you trade intraday trading using the tape to enter and exit positions, trading move to move and then occasionally looking for trades to hold longer. I was hoping to hear your thoughts on what you believe to be the best approach to intraday trading. How important are volume and order flow and action on the tape?

Hi Andrew. Thank you for your kind words on my book and this blog; I’m glad you’ve found something here that is useful (or at least throught-provoking).

I think the basic question you’re asking is “what is the best way to trade intraday?”, and there’s no good answer to that. There’s no one way to trade in any timeframe, but there are many approaches that might work. We need to be very careful of broad rules and generalizations. Anytime we say words like “must, always, never”, we might be in trouble. Even some rules that make perfect sense in one context, like “never add to a losing trade” might be completely wrong in another. Some options traders, some pullback traders, some intraday “fade traders”—all of these traders might depend adding to a loser as a key part of their trading strategy, but, for other traders, doing so could lead to ruin. Be very careful anytime anyone tells you, or you are tempted to think, that something is absolutely true for all traders.

So, there’s no way I can give you an answer as to the best way to trade intraday. I can, however, tell you my personal experience, which is that I made consistent money trading intraday in a few different contexts. In the mid/late 90’s, I made money basically scalping, but this was a complicated game in a different world: Two phone lines (yes, using a modem for datafeed), on the phone with a clerk in the pit, often executing offsetting orders before I got fills back on the previous order—like I said, it was a different time and markets were quite a bit different. I also traded on various timeframes (1-30 min), primarily in index futures and bond futures, using very simple chart patterns. Both of those approaches worked for me.

To your specific question about trading intraday using the tape, I’m very suspicious of order flow/volume/tape-based methodologies. There are some interesting approaches out there, and they are based on crunching large amounts of data with the idea that orders that hit the bid are “seller motivated” trades, and those on the offer are “buyer motivated” and that you can get some insight into imbalances of buyers and sellers by statistical analysis of this data. For instance, if a stock is trading up but orders are consistently hitting the bids, that’s a little unusual—in a typical situation we should expect people to be driving the stock higher by paying offers, or at least that’s the thinking. This work is not based on visually watching trades, but depends on crunching a lot of numbers in real time. In reality, there’s a lot of noise and many conflicting factors in this analysis.

As for visual inspection of the tape, I have trouble understanding where the edge would be. In the early electronic markets, there was an edge. People would put gigantic buy orders in, you could see the order in the book, and the order would likely have to step higher to be filled, so it was relatively easy to trade in front of the security of those orders. Today, everything is done with algorithmic execution that breaks up and hides large orders, to say nothing of the kind of deceptive games that are played “on the tape”. (I remember executing large size in thin stocks and doing things like putting large sell orders in above the market when I was actually a buyer. Why would someone show their hand if they don’t have to?) There is a lot of legitimate concern about HFT’s and algorithmic trading, but this type of order execution is ultimately better for the customer. Most of those large orders are customer orders, index rebalancing, pension funds, etc.—”Joe Public” is the customer by proxy for a much of the algorithmic trading and the algos get him better execution for lower costs. It would be very difficult to maintain the number of exchange traded products with low fees we see today without clean algorithmic execution, so I would actually argue this is a very good thing for the market and for the individual investor.

It is, however, bad for the daytrader, who was essentially a friction in the system. Today, much of the volume doesn’t even hit the tape. So much of the volume (upwards of 40% in some stocks) doesn’t even hit the tape, and any tape you see is filtered. If you were looking at raw transactional data, it would quickly exceed your ability to process and recognize numbers, so you are only, in the best case, seeing a fraction of the market activity. Much of the dark pool volume is delayed for minutes, so you aren’t seeing that within a relevant timeframe for many daytraders anyway. A large sell order hit the tape? Is someone bearish, taking profits on a long position, or is it part of a multi-leg strategy involving options (you never see the whole transaction) that is very bullish? Also, consider that there is simply a limit to human processing speed. People pay a lot of money to rent a space for a computer that is a few meters closer to the exchange server. The fact they pay those rents tells you that it is financially worthwhile, which means they derive an edge. When you think about the speed of electricity through a copper wire (or fiber), that should pretty much tell you the speed game is over for the individual trader. There’s a lot of vitriol written by struggling intraday traders, but much of this comes from putting your big, wet brain up against a dispassionate silicon brain that can make 1,000 decisions (literally) before you are even aware that anything has happened. There are things your brain does very well, better than any computer, so why not develop a trading strategy that accentuates those strengths?

I’m revealing my personal bias here, but it’s based on my experience, and the fact I have never seen a demonstrated edge (either quantitatively or in the form of a trader pulling consistent profits from the tape). I’ve looked high and low for trading edges, and simply was unable to verify that one exists here. Maybe there’s an edge I’m missing—there probably is. Intraday trading is not currently something I do, but if I do wade back into that arena I would not use the tape or order flow to any significant degree.


Adam Grimes has over two decades of experience in the industry as a trader, analyst and system developer. The author of a best-selling trading book, he has traded for his own account, for a top prop firm, and spent several years at the New York Mercantile Exchange. He focuses on the intersection of quantitative analysis and discretionary trading, and has a talent for teaching and helping traders find their own way in the market.