Intraday stock trading systems

Intraday Stock Trading.

Intraday Stock Trading.

Intraday Stock Trading does NOT mean just day trading.

To understand how I use the term Intraday Stock Trading it is first necessary to define some terms. The conventional definition of “day trading”, means buying and selling a stock within the same day. Day traders typically carry very large positions in order to make profits by leveraging large positions that will make a lot of money with just small price movements. Day traders need a combination of high liquidity and volatility for all this to work. Volatility is measured by the daily range and the larger the range the larger the potential profit or loss.

Intraday Stock Data.

However over 20 years ago I became interested in intraday data , not for purposes of developing day trading systems, but as a way for developing strategies for short term stock trading. Intraday data became an indispensable tool making it possible to make 2 and 3 day trading systems far more sophisticated than ever before.

When most traders hear the phrase Intraday Stock Trading they think of day trading. I have never liked day trading because I have found that day trading systems that I have developed, although marginally profitable, make more money if they are held onto for an extra day or two.

There was a time when day trading had some appeal because brokers allowed a day trader to trade more positions with less money because, or so the reasoning went, there was much less risk if no trades were held overnight. Hence a trader with limited funds could still trade significant position size as long as he was out of his positions at the end of the day. The bottom line was that a day trader could leverage his positions more than could a position trader.

But alas in 2015 this is no longer the case. After the 2008 crash Day traders became like the perceived communists of the 1950s and got blamed for all the stock market melt downs of the first decade of the 21 st century. Of course day traders had nothing to do with those melt downs, but the government that has never understood the dynamics of market behavior anyway, nevertheless instituted many restrictions designed to restrict day trading.

First and foremost among these restrictions is a rule that requires “pattern day traders” to have trading accounts of at least $25,000. If the account of a “pattern day trader” falls below $25,000 the account must, by law, be closed. Of course the rule makes no sense at all and to my knowledge nobody has ever asked the logical question, “How is it that people with less than $25,000 to trade could have caused, in 2008 a world wide melt down of virtually all equity markets?” Nevertheless day traders were in reality “little guys” with no political clout and consequently they made easy scapegoats for the government.

Regardless, in 2015, there is no good reason to day trade. Today day trading will cause you to need more money to make less.

This pattern day trading rule never has really affected me because I do not day trade. But I did have to eliminate most of my day of entry stops so that neither my broker nor the government would ever perceive me as a day trader. Surprisingly the elimination of day of entry stops has not degraded the performance of my two and three day short term trading systems. (For more interesting details on trading without stops (please see my article, Stop Loss Order ).

So if I am not a day trader why am I interested in intraday stock trading?

The reason is very simple. Intraday data gives the trader the opportunity to better control his positions even if he holds his positions, as I do, for two or three days. In order to understand what I am talking about let me first share with you just one general observation about markets.

I have said many times that I believe markets are predominantly random. HOWEVER, it is also my observation that the markets are more random at the beginning of the day and less random at the end of the day.

To work with this observation it is very important to be able to observe intraday data and plan entries and exits based on time frames other than daily bars. My trading system, Jordi’s Intraday 2 (please see automated stock trading software ), does in fact use two time frames, daily bars AND 15 minute bars.

Using intraday data, 15 minutes bars, is very important for my trading system even though I do not day trade. On Friday July 9, 2010 when the market opened my open trade equity was negative $5,000 but in only 10 minutes it went from minus 5k to positive 20k! Like I say the market is very random at the beginning of the day and I have programmed my system never to enter or exit a trade in the first 15 minutes. On that same Friday I closed out all those positions with a nice profit of $24,000. Can you imagine how upset I might have been had I exited all those good trades when they were negative five thousand dollars?

Furthermore if you read my article , Stop Loss Order , you will note that I prefer TIME STOPS over PRICE STOPS. Obviously Price stops are based on price only whereas my time stops combine price AND time, which of course requires intraday stock data for calculation. If I was using only price stops on Friday I would have been stopped out of my positions at the worst possible time.

But by writing time into my stops I avoided the wild random price swings at the beginning of the market day and took $24,000 in profits towards the end of the day when the markets are far less random. My intraday trading systems are unique because they combine two data streams, daily AND 15 minute bars, and this allows me to trade a vastly superior trading strategy and still hold trades beyond a single day.

But the real point I wish to make here is that Intraday Stock Trading does NOT mean day trading. I am a short term stock trader, but I am NOT a day trader or a swing trader and even though I hold positions for as long as three days I AM an Intraday Stock Trader and I am glued to 15 minute bar data throughout the market day. Using intraday data in this manner gives me an enormous advantage over most other traders.

Day Trading Systems.

Basic System Components.

With regard to day trading systems, whether you plan to be a discretionary day trader or completely automate your stock day trading, most intraday trading systems must consider the following components:

The ENTRY which originates from your setup and trigger to get you into the trade.

The STOP which controls your risk on the trade.

The EXIT which is the result of the initial stop being moved in the desired direction of the trade or a target price being placed at the time of entry.

The old saying that "amateurs concentrate on ENTRIES", is probably true.

After all, that's the interesting part -- looking at all those charts and figuring out how you're going to get in all those money making trades. That's what I did when I was a newbie twenty years ago, and if you're a newbie right now, that's probably what you're doing too. Right?

All of the above components of a day trading system are important, but if the term position size is foreign to you, do yourself a favor and learn something about it. I promise, it's an easy concept to learn and I'll keep the explanation as simple as I can.

For those of you saying to yourself "Hey, I thought expectancy was part of a trading system". don't worry, I've got a separate page for that and we'll go over that later.

WHAT'S THE DIFFERENCE BETWEEN DAY TRADING SYSTEMS, TRADING STRATEGIES AND SET-UPS?

I consider day trading strategies to consist of three basic steps:

A Set-up that puts the trader into 'alert mode' or 'watchlist mode' for a particular stock. This might include scanning stocks for particular patterns to trade or it might have to do with certain stocks meeting certain price gain and/or volume requirements. It all depends on the methods used by the trader. But whatever the strategy, it serves to put individual stocks on notice as having "potential" for a trade that day. One might post these stocks up in a tile fashion on their screen once found, with buy stop orders in place, so that they might watch for orders to be triggered.

A Trigger or Signal that initiates a Long or Short position in a stock. A simple example is a price move above a particular line of resistance.

A Stop that is either placed manually by the trader or automatically by the trading software.

As you can see by comparing to the list at the top of the page, trading strategies (under my definition) contain the work of finding or locating individual stocks to trade that day, setting them up for a Long or Short trade and immediately placing a stop order on that trade.

It requires possibly the most time consuming and interesting parts of the trade, but does not contain ALL, nor the most important components of a trade.

HOW DO TRADERS USE DAY TRADING SYSTEMS?

There are three basic ways that traders go about using day trading systems:

You can trade them 100% automatically using automated trading software. All of the above components of a system are initiated by the program's algorithm without the trader's assistance. The trader does not normally interfere with the programs decisions, unless the drawdown of the trading system seems completely out of normal system limits.

You can trade them semi-automatically. Which means software plays the role of assisting a trader to locate stocks automatically that meet his programmed criteria, and may even set up an order for him, but the trader decides using individual judgment on each stock, whether or not to hit the trade. Once in the trade, the semi-automatic day trading system would then take over and manage all exits.

A discretionary trader would use a system to trade off of, but would have very little, if any automation in his process. An example of this is a trader that uses a list of stocks such as a gainers or gap list to 'cull' or sort stocks for potential trades. Or, a trader that uses a third party stock scanner to find stocks for trades that meet his system's requirements. Another example is the trader that has a complete and backtested day trading system, that he trusts, but prefers to trade with 100% discretion as opposed to 100% automation.

As you can see there's some different roads you can go down during your trading journey and quite a bit of decision making that you'll need to do before actually starting to trade. Some of these decisions will be based on your individual background.

For instance, how deep is your computer programming knowledge? Strong programming and math skills will certainly make it easier to go 100% automated.

But, newer more user friendly software is coming on the market all the time, that make it easier for non-programmers to get involved in automated trading.

If that's the direction you want to take and don't currently have the skills, then you might want to check out my list of books on trading system design => trading books [choose the "trading system design" category] .

So, don't necessarily rule automated day trading systems out if programming isn't part of your education.

Stocks Intraday Trading.

Mode of education – Prerecorded videos.

What is Intraday Trading? – Where max. holding period is one day.

Segment covered – Stocks (Cash Market)

Introduction to Intraday Trading. Risk management in Intraday Trading. Money management in Intraday Trading. Bracket Order in Intraday Trading. Leverage & Margin calculations. Brokerage, Taxes and Charges calculations. Example of an Intraday Trading Strategy.

Description.

You will know how to integrate all the above components of trading in Intraday Trading System with the example of a Intraday Trading Strategy. This will help you to become a profession trader. This exemplified Intraday Trading Strategy requires 15 minutes per day.

There are no reviews yet.

Only logged in customers who have purchased this product may leave a review.

Intraday Trading Systems With End Of Day Data: Pivot Points Study.

Share this post:

This post looks at a couple of intraday trading system ideas using End-of-Day price data and pivot points.

Readers will be aware that I use Amibroker and Norgate Premium Data to construct trading systems and test trading ideas. This data is EOD (end-of-day) which means it contains just the open, high, low, and close prices for each full trading day.

In other words, the data does not include any intraday price quotes, not even at the hourly level. This data is thus perfect for medium-long term trading systems but it is perfectly useless for short-term trading.

You see, even though we don’t have any intraday price quotes, it is still possible to test some interesting short-term trading ideas. Let’s see how.

Intraday Trading Systems With EOD Data.

For quite a long time, I failed to realise that just because we don’t have intraday quotes doesn’t mean we can’t test some short-term ideas.

We have the open, high, low, and close prices available. We are therefore able to buy on the open and sell on the close; a short-term trading idea in itself. We are also able to trade on the close and sell on the open, an overnight trading idea.

So let’s look at some more creative ideas for intraday trading.

Pivot Points.

Pivot points are pre-determined levels that many day traders look at to help make trade decisions. They are calculated using the previous day’s price range and consist of the main pivot level, then three support and three resistance levels, placed below and above the pivot respectively.

You can read more about pivots and how they are calculated here. But essentially, pivots are key intraday levels that many traders take notice of. Since pivots are calculated using the previous day’s price action we can use them to test some interesting ideas.

In test one, we take EOD data for the E-mini S&P 500 futures contract (referred to as ‘&ES’ by Premium Data) and we buy the market whenever it opens below it’s pivot. We sell on the close.

To keep things simple we will use a fixed position size of 1 contract. Point value is set at $50, commission is set at $15 per trade and starting capital is $100,000. (For more information about setting up back-tests for futures using Amibroker see here).

Test one rules:

• If ES opens below it’s pivot, buy on the open & sell on the close.

• Position size = 1 contract, fixed.

Following is the code used to calculate the pivot levels:

P = Pivot = Ref((High+Low+Close)/3,-1);

Test one results:

As you can see, running this system between 1/1/2002 and 1/1/2014 produced a compounded annual return of 1.98% with a maximum drawdown of -18.87% over 1329 trades.

The win ratio is adequate but the CAR/MDD ratio is not particularly attractive. I then decided to test some other variations of pivots using the same settings. The results are shown below:

Buying the market when the market opens under the pivot and selling on the close was the best performing variable. There may be some potential for development by introducing longer holding times, a market filter, or adding extra rules and complexity.

Using (optimistic) commissions of $15 per trade, it is hard to find a simple trading system that works well for the E-Mini future. The E-Mini is a particularly tight market to trade and traders might have more success with more volatile instruments such as the Dow Jones Average, Nasdaq, FTSE 100 or DAX.

In test two, we are going to use the same settings as test one and stick with the E-Mini.

This time we are going to buy the index on the open, so long as the open is lower than the pivot AND higher than the second support. If the market touches the third support we will sell at that level, if not, we will sell on the close.

Test two rules:

• If ES opens between the pivot and second support buy on the open.

• If the market trades at or below S3, exit at S3. Otherwise exit on the close.

• Position size = 1 contract, fixed.

To set this up correctly I used the below formula. The S3 sell price was rounded to the nearest integer since the E-Mini has a tick size of 0.25. There is probably a better way to do it.

Test two results.

Running the test on the E-Mini between 1/1/2002 and 1/1/2014 returned a CAR of 1.63% with a maximum drawdown of -21.79%% from 1318 trades.

The system was run out-of-sample and produced a CAR of 3.98% with a maximum drawdown of -6.84% from 122 trades. The equity curve for the OOS period is shown below:

Test two performed better in the out-of-sample period than in the in-sample period albeit with fewer trades. There may be some potential for development by adding extra rules and experimenting with different levels of risk. Again, the E-Mini might not be the best contract to test. Maybe a commodity such as gold or crude oil would do better.

Test Three.

In our final test, we are going to return to the rules shown in test one but we are going to test them on stocks from the S&P 100 universe and we will not be using any margin. Instead, we will use a starting capital of $100,000 and allocate 20% of capital to each trade.

This time, whenever a stock opens below the first support, we will buy on the open with 20% of our capital and sell on the close. Commissions are set at $0.01 per share and only 1 trade can be made at a time. Some extra conditions are used for liquidity purposes.

Test three rules:

• If a stock opens below it’s pivot buy on the open and sell on the close.

• Universe: S&P 100 including historical constituents.

• Position size = 20%

• Starting capital = $100,000.

• Portfolio size = 1.

• Liquidity rule = open price is greater than $1.

• Liquidity rule = 25-day average volume is greater than 100000.

• Commissions = $0.01 per share.

Test three results:

Running the test between 1/1/2002 and 1/1/2014, returned the following results and equity curve:

Here is the out-of-sample equity curve for 1/1/2014 – 5/1/2015:

The best performing trade in-sample was Amazon $AMZN which was bought on the 23rd October 2008 (trade shown by the green arrow):

The results from test three show potential for further development, which could come in the form of longer holding periods, improved ranking mechanism, adding a market timing filter, additional complexity and different position sizing.

The results suggest that trading individual stocks might be easier than trading futures contracts such as the E-Mini and this could be because individual stocks are likely to be less efficient than high volume futures markets.

Conclusion & Criticisms.

So far, there has been limited stress-testing applied to these trading ideas. Monte-carlo analysis would be beneficial and we should also test more stringent slippage.

By making some adjustments to position sizing on strategy three it is possible to get some extraordinary returns (50%-70% CAR with only minor drawdown) but be warned.

The problem with test three is that it relies on getting accurate price fills on the open, which is not always an easy task. Bid/Ask spreads often widen on the open so (depending on the liquidity of the stock) it is necessary to increase slippage estimates. In order to do so, I recommend setting commissions from 0.2% or 0.5% up to 1% per trade and re-running the system.

Unfortunately, estimating the correct amount of slippage to use is not an easy task since bid/ask spreads change throughout the day, as liquidity increases and decreases. For some high volume stocks, the spread can be as low as 0.05% each way but for others (such as penny stocks) it can be as high as 20%!

So when you test strategies like this, make a note of the bid/ask spreads and make sure you have realistic assumptions about the stocks you plan to trade and the likely slippage you will face.

My advice would be to be conservative with slippage estimates and also, use some discretion when taking signals. If a stock opens below S2 on some very bad news, then you’re probably better off ignoring the signal. On the upper hand, if it opens lower than S2 for no apparent reason, that might be a good signal to take.

Also, where test three is concerned, you need to be able to scan for entries from a broad number of stocks, right on the open. Looking for setups in the pre-market and executing them with a MOO order is a consideration which might be worth further investigation and this is something that may be possible with the scanning tools at Interactive Brokers.

Overall, the main point of this article was to introduce some ideas about how to trade intraday using EOD data and I hope I have stirred your imagination a little.

There are hundreds more ideas that could be tested. Such as using the previous day’s bottom Bollinger Band as a level to buy. Or perhaps the previous day’s moving average level.

Or, we could try a trend following system by specifying our intraday buy price as the high of x number of bars ago. If you happen to try any such ideas yourself, please let me know in the comments.

Disclaimer: Past performance is no guarantee of future returns, you should do your own due diligence. No responsibility shall be taken for errors, miscalculations or trading losses. Please read the full disclaimer.

See More Posts Like This One.

Share this post:

2 opinions.

in the line above, what is the meaning of ” Ref() ”

what does it mean — Ref()

Ref is used to ‘reference’ previous or future bars. For example Ref(C,-1) returns the previous bar’s close.

Leave a Reply Cancel reply.

Recommended Educational Resources:

Remember: financial trading is risky and you can lose money. Nothing on this site is to be regarded as personalised investment advice. Past performance is not indicative of future results. Please see the full disclaimer.

Search.

JB Marwood.

Independent trader, analyst & writer.

JB Marwood is an independent trader and writer specialising in mechanical trading systems. He began his career trading the FTSE 100 and German Bund for a trading house in London and now works through his own company. He also writes for Seeking Alpha and other financial publications. Google+

Please remember financial trading is risky and you could incur significant loss of capital. Nothing on this site is to be construed as personalised investment advice. Please see the full disclaimer.