The 30-Minute Stock Trader
Page 9
Objectives
Trade long only, a big index, mainly blue-chip stocks that rise in value.
Only execute trades once a week (during the week you don’t need to check the markets at all).
Jump on the train of big trending stocks, with the expectation that they will continue to rise.
Strongly outperform the S&P 500 (at least double the CAGR), with lower drawdowns in bear markets.
Beliefs
Markets trade sideways, and they trend. My belief is that if we jump on the strongest performing stocks, we will have outsized returns trading them. We will stay in those positions as long as they continue to be the strongest performers in the markets. History has shown us over and over again that stocks can trend for a long time. Perfect examples are Microsoft, Apple, Dell, Netflix, and so on.
The biggest edge in this strategy is the unexpected, outsized, favorable returns.
As long as the trend is up and the stock is within the best ten performers of the universe, we remain in those stocks. Why would we sell a winning stock?
Trading Universe
Only trade stocks from the S&P 500 index.
For testing, this means we take into account all listed and delisted S&P 500 stocks, and their joiners and leavers data.
Filters
Minimum Average Volume of the last twenty days is above 1,000,000 shares (so that we have enough liquidity).
Minimum price is 1 USD.
Position Sizing
We trade a maximum of ten positions, and we divide our equity by ten to calculate our position size. This is a very simple position-sizing strategy. Of course, depending on your objectives, we can trade the same strategy using a different algorithm.
Example:Total equity: 100,000 USD
Size per position: 100,000 / 10 = 10,000 USD
If stock price is trading at 40.00, we buy: 10,000 / 40.00 = 250 shares.
Entry Rules
Today is the last trading day of the week (typically a Friday afternoon or evening, after the market has closed). We take end-of-day data.
The close of the SPY (exchange-traded fund, or ETF, of S&P 500 index) is above the two-hundred-daily simple moving average (SMA) band.There is a band set at 2 percent below the two-hundred-daily SMA.The price can go below the two-hundred-daily SMA, but not by more than 2 percent.
The two-hundred-daily SMA is a powerful and simple tool. Many institutional traders look at this moving average, but there is some noise involved. Many times the SPY closes below this average, only to go up again the following day, so we include a 2 percent buffer.
As long as the stock is above the two-hundred-daily SMA band, we are allowed to enter positions.
The SPY tracks completely the price of the S&P 500.
We trade up to a maximum of ten positions, and we select them as follows:The three-day RSI (Relative Strength Index) of the stock is below 50.RSI is an oscillator that measures how overbought or oversold a stock is. The higher the RSI, the more overbought the stock is.
Our method ensures the stock isn’t completely overbought. We like to enter trending stocks, but the statistical edge diminishes if they are in extremely overbought conditions.
If rules 1–3 are true, then we select the ten stocks with the highest rate of change (ROC—percentage increase) over the last two hundred trading days.We want to be in the biggest-moving stocks.We’re measuring momentum.
We enter the first day of the next week: market on open.
Exit Rules
Today is the last trading day of the week.
We stay in the same position as long as the stock is in the top ten of highest ROC over the last two hundred days. This means the stock is still on an uptrend. There is no reason to sell it.
We rotate into a new position as soon as the stock is not in the top ten anymore. This means we sell market at open on the next trading day and we replace the stock with a new stock that is in the top ten of the highest ROC (percentage return) over the last two hundred days.
In this strategy there are no hard stop losses.The reason for this is because of the exit rotation character. Since we rotate and are only in the strongest-moving stocks, a stop loss is not helpful, nor necessary. Also, setting stop losses on a weekly basis doesn’t make sense, and our objective is only to trade once a week.
The testing results are almost identical with or without stops. That is because the stop is rarely reached—when the stock starts to lose value, it gets replaced automatically.
If you feel uncomfortable trading without stops, and would like to better define your risk, just place a 20 percent stop loss, which equals 2 percent risk of your account (20% stop loss x 10% position).
This is an example of the 200-SMA band. The dark grey line is the two-hundred-daily SMA, and the light grey line is the SMA band, which is 2 percent lower.
For my testing, the software generated the following statistics: MAR, Sharpe ratio, R-squared, R-cubed, Sortino ratio, and Ulcer index. Each has its pros and cons. The most important thing is to clearly define your objectives before testing, so that you can identify which statistics will be most illustrative for your situation.
None of the statistics is a “magic number.” There is no such thing as hitting a certain mark on one and being set for life. All of them are based on past data.
Institutional traders often use the MAR ratio and Sharpe ratio. The MAR (which stands for “Managed Accounts Report,” the newsletter that developed this metric) is simply the relation between the CAGR and the maximum drawdown, a simple and clear g) ain-to-pain ratio. It’s great to use, but it doesn’t tell you anything about the duration of the drawdown. Also, once you’ve had a larger drawdown, the MAR will be punished for the duration of your time trading the strategy.
The Sharpe ratio is also used often by institutional traders—it’s the most widely used method for calculating risk-adjusted returns. If you want to trade for institutions, a high Sharpe ratio will be useful to impress them. It measures the average period return, in excess of the risk-free rate, by the standard deviation of the return-generating process.
A variation of the Sharpe ratio is the Sortino ratio, which does not punish upside volatility and can be useful for trend-following strategies, where the edge lies in a small amount of outsized gains.
R-squared measures the slope of the equity curve, and R-cubed measures the relation of the CAGR, max drawdown, and the duration of the drawdown.
The Ulcer index measures the downside risk, and takes both the maximum drawdown and duration into account. The lower the Ulcer index, the better.
There is no magic indicator. Define your objectives and see which statistic or combination of statistics is most useful for you.
The benchmark is the S&P 500 in this example. As you can see in the back-tested results above, the CAGR is almost three times higher than the benchmark, the max drawdown is almost half, and the MAR (CAGR divided by maximum drawdown) is far superior. The Sharpe ratio is almost three times as good, and the strategy has a low correlation.
The win rate is not high, but a win/loss ratio of two to one means the average winning trade is twice the average losing trade. With trend following, that’s your edge.
From 1997 to early 2000, there was a huge increase. Then, it was flat for two years. That’s exactly the time where the S&P 500 dipped below the two-hundred-daily SMA band, meaning we stopped trading. During the bear market, we were flat. With the market going down, that’s a great place to be.
After that, things kick off again, but go flat after the big bear market of 2008. By not doing anything with our money, we limit drawdowns, staying safe in cash. When the market recovered, we saw outsized returns.
It’s crucial to understand that this strategy can be boring—in bear markets, you don’t trade at all. That’s a good thing—you’re flat while others are losing.
The years 1998 and 1999 saw insane results, as you can see. Even in bear markets, we were still positive. Our benchmark here is the SPY, and as
you can see, we outperformed it by a lot.
Basically, this strategy outperforms in bull markets, and it limits drawdowns—going flat—in bear markets. When the markets are going down, we don’t have many trade setups. In sideways markets, it does OK—there isn’t a clear up or down trend.
When you see huge, outsized returns one year, there may be a time when the market turns and you’ll pay a price, with a big drawdown. That is to be expected when you’re a trend follower.
As I’ve explained, you don’t need to trade this strategy with stop losses. However, some people are so used to trading with stop losses that they feel uncomfortable when acting otherwise. The following chart explains how to trade the Weekly Rotation with stop losses.
We can clearly see that if we want to use a stop loss, small stop losses do not work well. That’s because with these strong-moving stocks, we must allow them room. They are often quite volatile, so if we set small stop losses, they’ll get stopped out. When we place the stop loss around 20 percent, the results are close to the original strategy’s.
Now, we’ll show some trade examples.
Example 1, Entry
After a long rise, DELL has entered the top ten stocks with the highest ROC over the last two hundred days, and it has an three-day RSI less than 50, so we enter on the open at the first day of the week.
Example 2, Exit
This was a dream trade, which we stayed in for over two years. The stock kept rising and was the whole time in its best performers.
Example 3
Entry: 85.04
Exit: 131.08
The Weekly Rotation strategy shows how easy it can be to make high double-digit returns, while working just thirty minutes a week. The simplicity of the strategy makes it easily tradable, yet you wind up with almost three times the annual return (19.7% CAGR) of the S&P 500, with almost half the drawdown.
You have one task each week. Check if the trend of the overall market is up, and if so, select the ten strongest-performing stocks (highest ROC) over the last two hundred days. Anyone can do it.
This strategy is great for people who like to be a bit more contrarian and who are comfortable trading against the herd. That means people who won’t have trouble ignoring the news, and won’t be susceptible to being affected by outside noise. Also, it is good for people with IRA accounts.
Objectives
Trade long only on a large universe of stocks, taking advantage of oversold conditions by buying the best stocks and selling each when it reverts to its mean.
Execute trade in less than thirty minutes a day.
Take advantage of stocks that are in a long-term uptrend, that have significant volatility and are oversold on the shorter term. By buying these stocks even lower and selling them when they snap back, we have a consistent edge, and aim for 60 to 70 percent winning trades.
Beat the overall market in both bull and sideways markets.
Beliefs
There are always times in which the markets will show such irrational, fear-based behavior that there is a statistically larger than normal probability that a reaction will cause a significant, opposite move.
Markets are driven by fear and greed. If we want to take advantage of this on the long side, we need to look for candidates that have shown a lot of fear. These stocks are mostly out of favor and have shown large sell-offs, mostly accompanied with larger than normal volatility.
There will come a time when the panic selling of these stocks will stop, and often this is when professional traders jump in and look for bargains. Statistically, there is an edge in “buying fear” and selling it when it has reverted, or shown some upward movement.
This kind of trading goes against human nature; it is trading against the herd. It is not easy to be a buyer when everybody is selling and all news messages show panic, but that is exactly what makes it a profitable strategy.
Since we are looking for a shorter-term trade (a couple of days), we need a large trade frequency. This is done by both scanning over a large stock universe, and creating exit rules so that we get out quickly.
Trading Universe
We trade all US stocks from AMEX, NASDAQ, and NYSE (around seven thousand listed stocks).
We trade a large basket of stocks, because we want a high number of trades (mean reversion trades are short term—therefore we need a high number of trades for significant profit).
We do not trade ETFs, pink sheets, or bulletin board stocks.
For testing, this means that we take into account all listed and delisted stocks. Since 1995, this list consists of around forty thousand stocks.
Filters
Minimum Average Volume of the last fifty days is above 500,000 shares.
Minimum price is 1 USD.
Dollar volume is at least 2,500,000 USD.This is to make sure that, if we’re trading low-priced stocks, there’s enough dollar volume for it to be worth trading.
Position Sizing
- Fixed fractional risk: 2 percent This means we take into account the dollar volatility of the stock.The higher the volatility of the share prices, the lower our position will be, and vice versa.
We trade a maximum of ten positions in this strategy.
Each position risks 2 percent of our equity. The calculation of this is as follows.Equity: 100,000 USD
Risk per trade: 2 percent
Dollar risk per position: 2,000 USD (2% x 100,000)
The risk is calculated by the difference between the entry price and stop loss price.
If the entry is at 20.00 USD and our stop loss is 17.00 USD, then we have a dollar risk per share of 3.00 USD.
How we calculate our desired position:If we are risking a total of 2,000 USD per position, and this equals 3 USD per share, our size is: 2,000 USD / 3 USD = 666 shares.
We use this type of position sizing so that we know exactly what percentage of our equity we can lose on each position. Also, we define risk by the difference between the entry and stop loss. Since we use a wide stop loss, few trades get stopped out. If we close a position with a loss, we typically lose less than the total amount risked.
However, we have not yet defined our total size, just our risk per trade. There could be a time where a position gaps against us. In the previous example, we entered at 20.00 USD, and our stop loss was set at 17.00 USD, but there’s a chance that an overnight news event could cause the stock to open at 10.00 USD. If that happens, our stop is irrelevant; we will suffer a large loss at the open price of 10.00 USD.
Also, in times of low volatility, the stop can limit us, because the stop is set based on the volatility. The lower the volatility, the smaller the stop size, and the smaller the stop size, the larger the total position.
To combat these two issues, we limit the total position size to a percentage of total equity.
Max size: 10 percent.In this example, we had been allocated 666 shares, 13,320 USD, or 13.32 percent of our equity.
This is will be adapted to 10 percent of our equity. So, the final size is: 100,000 / 10% = 10,000 USD / 20 USD (STOCK PRICE) = 500 SHARES
This second rule only applies in certain times, mainly times of low volatility, depending on the stop size.
When we only use fixed fractional risk, we define our position sizing based on past volatility. Of course, it’s good to base our sizing on volatility—however, we’re using the past, and we don’t know if the future will be the same. That’s the drawback of this method. There are many ways to deal with this, with advanced sizing algorithms, but this is just a simple example that limits our position sizing to a percentage of our total equity.
Entry Rules
Close of the stock is above the 150-daily SMA.We want stocks to be in a long-term uptrend.
Seven-day average directional index (ADX) is above forty-five.ADX measures the strength of the (shorter term) trend. The higher it is, the stronger the trend is.
Here, we’re looking for quick, short-term profits. For that, we need stocks that move significantly. (Hig
h ADX.)
The average true range percent (ATR%, measures volatility) of the last ten days is above 4 percent.It’s important to use ATR percent, to adjust for share prices.If not, the higher-priced stocks would show a much larger ATR versus low-priced stocks, but that’s not an objective comparison. By expressing ATR as a percentage of the closing price (ATR%), we objectively compare the volatility of all stocks.
For perspective: The S&P 500 has a long-term ATR% of 1.6 percent. So, at greater than 4 percent, we’re only trading high-volatility stocks.
The three-day RSI is below thirty.RSI is an oscillator that measures overbought and oversold conditions.
The lower it is, the more oversold the stock.
With this rule, we make sure we select the stocks that are oversold on a shorter-term basis.This is our first method of quantifying fear.
When rules 1–4 apply, we place a maximum of ten orders.
We rank the orders by the lowest three-day RSI (these are the most-oversold stocks).It’s important to rank them, because we are only trading the ten most-oversold stocks. Without rankings, we could have fifty to two hundred setups to choose from.
In this rule, we ensure that we’re selecting the stocks which have the most fear.