I have been developing trading systems for almost 25 years now. When I started in this business, the two methods everyone was using were channel breakout for long-term trading and volatility/opening range breakout for shorter term swing trading. Volatility breakout was amazing from the 1980s through early 2000s. You may have even argued that back then it was a license to print money. During this time, the open outcry pit system was king. Electronic markets started in the early 2000s and did not have much volume. Everyone traded the day pit sessions, using an opening time that became hardwired into traders’ brains. During this golden age of trading, both volatility and opening range breakout took on legendary status. For example, during the period of 1986-1987, Larry Williams and Sheldon Knight both took small accounts ($10k and $50k, respectively) to over $1 million in less than a year. Larry did so in the Robbin’s World Cup and Sheldon did so in his own private account. I have seen proof of both these achievements. Traders may also recognize opening range breakout from Toby Crabel’s famous book “Day Trading and Short Term Price Patterns and Opening Range Breakouts”. This book had one printing and is really rare to find now. It has a cult-like following and because of this, Crabel’s book now sells for $600 for a “like-new” copy on Amazon. Back then, there were also a bunch of commercial systems that used to sell in the thousands. Then, in 2007-2010, the open outcry pits closed. In these new electronic markets which were 24 hours, they would close trading for anywhere from 30 minutes to 4-5 hours and then re-open them. So they are open from 5 PM and 1 AM (for example) when there is not any volume on the open and we don’t have the effect of overnight news. That rendered opening range breakout (including all those many systems based on opening range breakout) useless for futures and commodities. One of the greatest trading methods of all time was dead. This really bothered me. I was determined to bring back opening range breakout. I spent a few thousand hours researching these breakouts off of the open during the 20 years before the pits were closed. I even produced a best selling video that was translated into several languages. It featured my work on opening range breakout and the methods of Crabel, Williams, and Knight. For the past 7 years, on and off, I have been thinking that there must be a way to resurrect this method. In order to do this you either need a “new time” which holds the same predictive value as the old pit open or a way to calculate breakout points based on the previous 2-5 days price action. A few months ago, I finally discovered a brand-new methodology that used price and volume patterns on 24 hour markets to discover a new statistically-valid predictive opening time. What is amazing about what this technology does is that the open time shifts +/- up to 2 hours based on my analysis. It actually automatically shifts based on reports coming out. I have called this breakthrough idea “Dynamic Open™“. You may be wondering if this code is “trivial”. It is not. I do not just average the open of the past few days or something minor like that. This idea was, in fact, so complex that I had to call one of Using EasyLanguage’s experts and co-creator of EasyLanguage, Sam Tennis, to help me code this idea. We worked on it for almost a month and now have coded Dynamic Open™ and all the supporting functions necessary to develop trading systems on it. The original opening range breakout methods work well on stocks, My Dynamic open solves the problem that futures now have because they are now 24 hr. See my thread in the announcement section. Let's discuss classic opening range breakout for stock here and I will answer questions on how you can use my dynamic open to update your opening range breakout systems currently in moth balls.
Thanks Murray. I gave you a like cause you don't have enough. I have a vague concept of this based on my own trading and L.R.'s work. Now in currencies so I see firsthand what a market with limited time boundaries is like to trade. --Swimr
Still works on commodities in high volatility environment, what have you discovered about "volatility" surrounding OR times ? The largest component is volatility... / No ?
I used Opening range breakout/Volatility breakout in many systems before it stopped working in the mid/late 2000's on futures. I been looking to find a methodology which will give me a opening point which works like the old pit open. That a product of mine. I also discovered that classic opening range breakout still works well in stocks. I have TradeStation code for that which I sell for stocks. But this is just classic opening range breakout, I will discuss the rules and issues in this thread openly. I have code and tools, system shells for all three methods. 1) Crabel Stretch 2) Knight Stretch 3) Williams Volatility offset If you want to code it your self. I will share enough info you can but for $99.00, many people will just buy it. The Open for Futures markets I call Dynamic Open(tm) This is something I am selling and I even require a NDA to be signed. So I guess it's a bit of both.
I use relationship between price and volume on a 24 hr market over some recent window and use that to find a time to define as the open for the next day. This time changes by +/- 2 hrs longer term. This is what I call Dynamic open(tm).
Toby Crabel in his book defined a formula to use as a offset to the open. We call this the Crabel Stretch, The formula is below in easylanguage, based on intra-day data in Data1 and daily data in Data2. average(minlist(open data2-low data2,high data2-open data2),SMA_Lenght); SMA_Length=10 as default. That what he used in this book. In my code I created functions which work on intra-day data only without needed to use 2 data streams , the daily data, that one nice feature of my simple library for stocks. Next Williams Volatility offset is simply Open+ some fraction of the 3 day average range for buys and Open -some fraction of the three 3 day average range for sells. Finally the Knight Stretch is as follows: We calculate the size of the channel breakout channel. Highest(High,X)-Lowest(Low,X) We then use a percentage of this added the the open for buy and belowe the open for sells . The number is often small 15%-30%.
So it sounds like you're using bump in volume around the time of an open to actually define The open for that day
If you just look for a bump , you get nothing useful. You need to do more statistical analysis on the price and volume to figure out where the dynamic open should be tomorrow.