## Tuesday, August 21, 2012

### Day trading and why you shouldn't do it

Day trading is when you try to take advantage of the intraday fluctuations in prices of financial instruments like stocks and bonds. The idea is that even though the overall daily price may decrease, the price will increase at some points throughout the day. If a trader can time his buys right before the increases and sell right before it starts decreasing, he can make a profit.

Consider the following intraday time series for a stock:

Overall, the stock price dropped from $10 to$8, a 20% loss. However, an omniscient investor could in theory buy the stock at 10 AM, sell at 11 AM, buy again at 1 PM, sell at 2 PM, buy at 3 PM, and finally sell at 4 PM. The net result is a $3/share, a 30% gain. This kind of arbitrage can almost always occur, regardless of market conditions or time scale. This is because while the overall trends are consistent and can be predictable, the small fluctuations caused by people buying and selling shares is inevitable. There will almost always be both buyers and sellers throughout the day and if at any given point the number of supply exceeds the demand, the price will go up temporarily. The potential for profits are huge as our previous example illustrates, but how does one go about realizing that potential? There are a number of techniques that traders use to try and predict these upswings and downturns. One simple one is to monitor the news. When good news comes out, the stock price can be expected to go up in the near future as other traders are enticed into buying. Similarly when bad news comes out, the price is expected to go down. Important news reports like quarterly earnings can result in huge changes in stock price like when Chipotle Mexican Grill (CMG) plummeted over 20% on July 19th when it's second quarter earnings of 2012 failed to meet analyst expctations. Analysts publish their own "buy/sell" ratings which in turn influence the public's perception of a company. Companies also announce things like new products, price cuts, etc. which affect the stock price. You can also enter into the realm of technical analysis, which involves looking at charts and time series of a stock to predict it's future behavior. Proponents of technical analysis try to take advantage of history repeating itself and looking for patterns of stock price movements. For example, I'm sure most stock traders have thought about or tried buying a stock simply because it has reached a 52 week low. The thought process is that the price must go up afterwards, or at least one can reasonably have a positive expected value by gambling on that. ### Criticism I am of the opinion that one should not attempt day trading. My reasoning is as follows: #### Trade Commissions Whenever you buy or sell a stock, you typically pay a fee to your broker for completing the trade on your behalf. The fee is usually on the order of$8 per trade, or $16 total for both buying and selling a stock. If we look at the intraday swings of a stock, they are typically on the order of 1% of the price. This means that if we make the optimal trade decisions, we would have to invest over$1,600 in a single stock to simply break even.

More reasonably speaking, even the most well-informed traders cannot hope to accurately pinpoint the optimal time to buy and sell, reducing the profit margin even more. I would conservatively estimate the average profit margin for each day trade to be about 0.5% at best.

Also by investing so much money into a single stock, you are losing a lot in the way of diversity. An individual investor simply cannot afford the risk of dumping all his money into one stock. For sufficient diversity and margin of safety, an investor would likely need over $500,000 in trading capital to potentially make any semblance of a profit. Note that most stock brokers require a minimum account balance of only around$25,000 to day trade.

#### Opportunity Cost

Suppose I felt lucky and decided to spend $100,000 today on 10 day trades, making a 0.5% gross profit. This would amount to$500. The 10 day trades would result in a $320 commission fee, leaving me with a$180 net profit for the day or a little over $20/hour. Well if I had$100,000 of spare cash lying around I was probably already making over \$50/hour at a regular day job with almost no risk, additional benefits, and probably a more fulfilling position. With the my spare cash, I could invested that in a less risky and simpler manner for a significant additional source of income.

#### You Can't Beat the Machine

Many financial firms are participating in high frequency trading (HFT) which use computers to day trade. They use various algorithms to let computers analyze stocks and make trades faster than a human can blink. It is likely that these computers already beat a human day trader to the potential arbitrage opportunity when the information or signs become available.

I don't think day trading is necessarily bad, but I believe that it is now outside the scope of human traders. In fact, it is absolutely essential by providing liquidity to the market and making the price reflect the value of a company more quickly.

I would associate day trading with arithmetic and long term investing with mathematics. One is better left to computers while the other to humans.