Sorry if I came across as abrasive, I just need my coffee.... In your ACD quest search for everything that Mav has ever said about OR%. He says some pretty telling things outside of this ACD thread. He spoke about the inverse relationship of OR% size and the correlation to trend days. He also talked about scanning for 50% OR% fades and he told the poster he does not do that anymore. Based on this conversation from years ago that I read I decided to see if the stats backed up what Mav said. Sure enough it holds up very well. Mav always says you should not hang your hat on "magic levels" but his observations and research on OR% and trend days are real. This ties back directly to what Mark Fisher originally said. Spend an equal amount of time on Number Lines if you can. If you need data or if I can be of any assistance in your ACD quest then let me know.
Hello Double M, Thank you sir. I did a poor job of reading your post (quick scan) and thought you were referring to the creation of A levels. Had I read it with any depth ... Duh! Anyway, I pulled out my well worn "Mav's Underground ACD e-book" and sure enough, under OR/ATR Ratios, I found the following: "5/13/11 I created a ratio that I use to put on my charts that simply shows the opening range divided by the ATR. That was very helpful. But wide opening ranges give you great fade trades especially into long term trends. 11/25/11 OK, I'll throw out a bone here. For stocks, the best way to look for good confirmed A ups and A downs is to look for narrow open ranges. You want to program this into a ratio. Take the OR into say a 30 day ATR. Then run the scans during the day to find the tightest ratios, for example all stocks under 20%. From this list, you look for confirmed A ups and A downs. Combine that with stocks with narrow pivots or narrowing pivots, and you will find your breakouts. 11/26/11 SINA...Good example of a stock today with a tight opening range (17% OR/ATR) and dropped 10 pts after it broke through the A down. 7/12/13 I use to have a ratio that automatically calculated the size of the OR divided into the ATR and if that ratio was greater then 30% I think, I avoided breakout trades. Then on the extremes, if say the ratio was greater then 50% then I actively looked for fades and with size. And if the ratio was under 20% then I actively pursued breakouts and with size. So there was a lot of conditionality. For the record, I don't do this currently, but that is how I used that data." Thanks for reminding me of this ratio. Could pay dividends when, say you have a strong sector NL, then particular stocks with a strong NL, then perhaps nice pullbacks to a failed A level, then perhaps a nice Ratio, then ...... I also wondered if it would have any significance with weekly. So, I've started some data gathering with daily and weekly material. I have a number of them so far. Here's a daily from today and a weekly from February 13 thru the 17th. The daily ATR is a 10 day with extra weight for the most current 5 days. The weekly is a 20 day with more weight for the most current 10 days. Thanks again.
No problem Robert, I actually found some of my info from Mav by searching online and using the wayback machine to search cached pages that are now possibly offline/deleted? It looks like that underground E-Book does contain some of the info that I found. Too bad Maverick74 does not get royalties from that ebook..... Honestly, I have not done any research outside of applying this to daytrading. I plan on taking one more shot at daytrading for a living via the ACD method. I am putting the final touches on my own custom version. If it does not workout then I will have to work a day job and I will probably trade on a weekly to monthly time frame. The heart of this method is around daytrading. The idea that markets are closed overnight and when they open they in the morning that a spike in volume as well as range will happen early on and that levels to trade off of can be established. Let's say that every market around became a 24 hour market and that there was no longer any real rush of volume and range at a predictable time of day (like the open), well this method is still very compelling because of the Number Lines. I like many other ACD newbies spent way too much time trying to find magic levels. In my opinion the gift of Number Lines is not only an initial bias but also product selection. Some people look at way too many names when trading and some narrow their world severely with "I only trade Apple!", "I only look at ES and CL!". Why not let the best ranking Number Lines tell you where you need to be? A consistent approach and structure to product selection, a neglected skill and art no-doubt. Randomly being persuaded by news headlines and CNBC top stories or being married to a single product seems to be a way of life for most.
Here are some charts to look at regarding Trend days vs Non trend days and the Opening range as a percentage of the 10 Day Average range in both of these cases. I ran 2 runs over around 30 etfs roughly 1.5 years of data which resulted in 7955 observations. I defined an up trend day as a market having an open within either 10% or 25% of the daily low or high and a close in either the top 10% or 25% of the low or high. Here are the resulting histograms for Trend Days and Non Trend Days using 10% and 25% values. I defined the opening range being first 5 min high-low from New York open. Using 10% value out of 7955obs 7475(94%) were non trend days 480(6%) were trend days. Using 25% value out of 7955obs 5856(74%) were non trend days 2099(26%) were trend days. Would be interesting to see what you guys make of these.
I think your definition of a trend day is not strict enough. A trend is structural, higher highs, closes at the highs of the day ect..ect.. You could be picking up late day reversals that look nothing like a clear trend. I am not clear on your % filters either. What Mav originally said in an extremely strict sense was that trend days tend to happen with an Opening Range LESS THAN 20% of the 5 Day ATR. In my study I looked at everything under 30%. Again I am not clear on how you put this study together, you want to find out what percentage the Opening Range is of the 5 Day ATR. Then filter those results for OR%<30.
First rule about Mav's Underground ACD e-book...don't talk about Mav's Underground ACD e-book. Second rule about Mav's Underground ACD e-book...don't talk about Mav's Underground ACD e-book!!!!