Trend Trading

Statistically Valid

Ease of Use

Simple to Master

Robustness

Durability
Summary
We are told that “the trend is your friend.” But is it really? Have you ever seen statistically valid evidence that either proves or disproves this wellworn market maxim?
Does trend trading reveal an intraday market edge? Should you be trading with the trend, with your intraday trading?
The answer is yes. And we would like to present evidence that suggests that trend trading is the best option for new traders. Included is a practical strategy that disproves the efficient market hypotheses and serves as a launch point in which to develop your own custom trading strategies.
When it comes to intraday trading strategies, is the trend really your friend?
We often hear this oftrepeated maxim about how we should always trade with the trend. That we should, “go with the flow” and that “trend trading works.” We are told that trading against the trend is a surefire path to financial ruin.
And, we also hear and read about magical trading indicators and supposed “order flow” techniques that can supposedly predict market tops and bottoms.
As usual, the truth lies somewhere in the middle. This space in which market falsehoods occupy and perpetuate can seem as long, lonely and barren as a drive across the state of Texas. Those of you that have driven across Texas know exactly what I am talking about.
Anyway, I wanted to take a moment today and talk about intraday trends. And a simple trading indicator that you can use that is both elegant, statistically valid, robust, and might just save your ass from blowing up yet another trading account.
The MidPoint of the day
The midpoint of the daily range is a simple indicator. Suppose that are you are trading the Emini SP500 futures contract. And the high of the day is 2350, and the low of the day is 2300. What is the midpoint? No fancy calculator needed. The midpoint would be 2325.
And so, if you are trading intraday, a long position above 2325, then you are quite simply trading with the trend.
Conversely, if the price of the EminiSP500 is below the midpoint of 2325, then we would consider the immediate trend to be down. If you are short below 2325, then you would also be trading with the trend.
The concept is ridiculously simple. And it also conforms with the theory of Occam’s Razor, that which is most easily observable, and most obvious, is usually the best answer. For instance, if you observe 10 chickens in a henhouse, then you are most likely going to see eggs. If you see a wolf dressed like a chicken, and he is selling a trading indicator, then he is most likely trying to steal your eggs (or your chickens). What is most easily observable will usually yield the most statistically valid answer.
So let’s jump back into our super simple trading theory that if the current price of the Emini SP500 is above the midpoint of the trading day, then we want to only be a buyer. Nothing complicated here. We want to use this simple observation to hopefully predict the future.
Conversely, if the price of the Emini SP500 is below the midpoint of the trading day, then we only want to be a seller.
This is the quintessential meaning, and most easily defined and elegant approach to trend trading. Going with the flow. Let’s test our theory with actual trading data.
Our MidPoint Strategy Defined
With the following test, we are going to use 5minute bars of the Emini SP500 futures contract.
Before we take any trades, we are going to wait 2 hours. We want to simply observe the first 2 hours of the trading day. We want to let some sort of market structure develop.
After 2 hours, we then calculate and plot the middle point of the intraday range. As the market moves to new highs or new lows, then the middle point of the day will continue to adjust higher or lower.
We only want to take trades in the direction of the trend. So if the price is above the middle point of the day, then we only want to take long trades. If the price is below the middle point, then we only want to take short trades.
Our entry point for buy signals is exactly as follows:
 Wait 2 hours.
 Calculate the middle point of the intraday range.
 If the low of a 5minute bar crosses above the midpoint, then we want to buy at the market.
 Exit the trade for whatever profit at the end of the trading day.
 Exit the trade for a loss if the high of the 5minute bar crosses below the midpoint of the day.
The following example is a winning trade:
But what if we are wrong? How do we exit the trade? We will use the exact same logic to exit the trade.
 If we are long, we exit the position if the high of the 5minute bar crosses below the midpoint of the intraday range.
 The exit conforms to our entry logic, that if the price is above the midpoint we want to be long, and if the price is below the midpoint then we want to exit our long position.
The example below is what a losing trading looks like:
The entry and the exit are simple, congruent, easy to identify and keep with the Theory of Occam’s Razer that simplicity and easily observable events tend to yield the most robust answers.
The following equity curve highlights the stupid simple concept of always “trading with the primary trend” by only using the midpoint of the intraday range.
With an astounding sample size of 3,654 trades over over the past 17.5 years, we can see the robustness of the approach.
But many readers are probably already screaming foul! And they are saying, “hey smart ass, know nothing Emmett, this supposed genius (con artist, jail bird, felon, hustler, scoundrel, etc) …these trades were in the stock market and are only LONG trades. And the stock market has basically only gone in one direction. Up.” Correct! So now we have to look at how our system performed on the Short trades.
Shorting the Stock Market: A suckers game?
Now we are going to reverse our strategy, and only take short trades, trading the Emini SP500 futures contract. Which since 1929, going short the SP500 has basically been a fool’s errand.
Once again, the exact rules are as follows:
 Wait 2 hours.
 Calculate the middle point of the intraday range.
 If the high of a 5minute bar crosses below the midpoint, then we want to sell at the market.
 Exit the trade for whatever profit at the end of the trading day.
 Exit the trade for a loss if the low of the 5minute bar crosses above the midpoint of the day.
The following is what a winning trade looks like:
But what if we are wrong? How do we exit the trade? We will use the exact same logic to exit the trade.
 If we are short, we exit the position if the low of the 5minute bar crosses above the midpoint of the intraday range.
 The exit conforms to our entry logic, that if the price is below the midpoint we want to be short, and if the price is above the midpoint then we want to exit our short position.
The example below is what a losing trading looks like:
The following equity curve highlights the stupid simple concept of always “trading with the primary trend” by only using the midpoint of the intraday range.
With an astounding sample size of 3,457 trades over the past 17.5 years, we can see the robustness of the approach.
So now that we have investigated both the long side and the short side, by using only the midpoint of the intraday range, let’s combine the results. The following are the combined results of trading both long and short, using only the midpoint of the intraday range.
Is this trend trading strategy robust?
Absolutely, for just the Emini SP500 futures contract, the total sample size is over 7000 trades. This is a massive amount of data. If the market were truly random, then the equity curve would be flat, and the trade expectancy would be close to $0.
Sometimes you just have to trust the Law Of Large Numbers, which simply states that the larger the sample size, the more reliable the statistical output.
But is this strategy ready to be deployed and traded ASIS? Absolutely not. The truth is that the average trade size is only $16 per trade. Not enough to outpace slippage and commission.
However, the sample size of 7,000 gives us a huge amount of space in which to introduce and test different scenarios in which we should be taking trades and not taking trades. And these different scenarios will be the launch point of many new articles that I will be writing that find, and expose these market biases.
Trading Strategy development should be like making soup.
This might sound strange. But when making a soup, you have to start with a stock. You have to boil out large quantities of bones of vegetables. And you are ultimately left with a rich and nourishing broth. A base in which to add different ingredients, to find the right taste, texture, and nutrition.
Same goes with trading strategy development. You need to find these large biases that disprove the efficient market hypothesis. You need to find and isolate these broad and wide occurrences, like the MidPoint Trading Strategy that we have just tested. Something that initially yields a large broth of statistical edge. And then you need to start working on the data, introducing different scenarios that whittle down the sample size and yield an average trade that financially makes sense. That outpaces slippage and commissions.
In future blog posts, we are going to be taking the MidPoint Trading Strategy, and treat it like a soup broth. We will then be adding and subtracting interconnected data points from other markets, in the hope that we can find something that launches your own imagination into unknown territory.
Thanks for reading.
This was a pretty long blog post. And the topic might have been difficult for many to follow. Many readers are unfamiliar with trading strategy development. So we are going to take it very slowly. And release trading strategies that “stairstep” from a basic concept with a large amount of “broth” into more refined concepts that yield a full blown delicious soup. Something that will yield actual nourishment and inspire your imagination towards places from where your genius is hidden.
And if you have any trading ideas that you simply cannot program, and are curious if any legitimate market edge exists…let me know. I can program and test just about any idea that you can imagine.
Thanks for reading.
Hello Emmett,
When you say to wait two hours, I’m assuming you are talking about regular trading hours. My question is: Do you ignore the price action during Globex (overnight trading)? I’m talking about futures only, RTY, NQ, YM, ES, CL, etc.
Thank you.
Yes, totally ignore the overnight session.
However, this does not mean that overnight data is not useful. As a matter of fact, you can create a big bunch of trading systems just based upon the overnight session. In particular, if the overnight range is greater than a % threshold, then the market will tend to perform in a certain way during the day session.
Hello Emmett,
How do you define a % threshold?
Also, regular trading hours are from 9:30 a.m. to 4 p.m. New York time. 6 hours 30 minutes total. The middle would be 3 hours 15 minutes or 12:45 pm EST. If we wait 2 hours from the open, it would be 11:30 am EST.
So when do we look at the midpoint of an intraday range, at 11:30 am EST or at 12:45 EST? I would appreciate if you could clarify that.
Thank you.
Hi John,
Can you do me a favor and ask this question and at my new Q and A website? Its located at Trader.Help.
This new site was specifically created so that I can answer these questions and will make it easier for others to contribute.
Thanks,
Emmett
Hi Emmett, you got any good reviews coming up?
“We want to simply observe the first 2 hours of the trading day.”
What does this mean? When does the trading day start? Obviously not at midnight Eastern (although that’s when the algos reset). 7 AM? 8 AM? 9:30 AM? The examples you show appear to be picking times to start looking for trades at different times. Please advise – thanks!
Hi Ed,
My technical writing skill is sorely lacking.
With the ES or Emini SP500, the morning session begins at 8:30 AM in Chicago.
Hi Emmet, this seems promising, allow me to ask a few questions 🙂
1. Can you better define what you mean when you say mid point of daily range? When it’s 2 hours past open ( 11:30AM) and I’m looking at the chart, the day as barely just begun, how do I establish my mid point? Is it the range between 8:30 AM and 11:30 AM and I split that in 2 to find the mid point? Is there an indicator available?
2. Can you please point me toward a platform / website that provides a proper VIX indicator? I trade the SP500 through my forex broker MT4 .
Thank you!
Hi Emmett,
Really like your blog and want to say thank you for your review on TopStep Trader. Count me as one of the hamsters. Been doing well ever since going with OneUp Trader and have built up enough profits to soon be going out on my own.
Regarding the midpoint / median daysession price idea. I have had good success with this concept using VWAP. Much of the time VWAP and the median daysession price plot pretty close to each other in the morning and will widen the gap as the day progresses. A really great way to catch some large moves.
I’m also a huge fan of Connors work. Been trading his ideas for over 10 years.
Thanks again for all you do.
Sincerely,
Gomez
I don’t get it. How can you wait 2 hours to establish what the middle point is. Don’t you have to wait an entire day to find out what the low and high was of that day?
e.g as you mention “2350, and the low of the day is 2300”
This is where I get confused, wouldn’t we have to wait a whole day to draw the middle point, meaning we can’t trade that day as we are waiting to find what the actual middle point is first …. then we could trade the next day off that previous days middle point as a indicator?
I’m probably not understanding it correctly. Perhaps if you could clarify it would be greatly appreciated.
Very interesting article. I’m wondering how stop loss strategy will be e.g where to place the stop loss, and ideas for targets.
Cheers
Forgot to mention in my previous comment. Is there any way we can code an indicator for this that constantly shows the mid point ?
If you could let me know the basics required I could get one made.
Sure, its pretty simple. All you need is the following:
((Highest High of intraday) + (Lowest Low of Intraday)/2)
I don’t get it. Your examples show other bars crossing the midpoint before your expample bar.