Somehow I thought the above statement was made to trade and predict a gap before its happening, potentially. Now it appears you may refer to trading a gap afer its happening.
Good posting. Where did you take the graph from? Where can one get such data about the success of intraday patterns e.g. certain candlestick formations for NQ? Thankful for any hint.
It depends on whether a gap of lesser than (say) 1% is a Real gap, as comparing to the daily (or else) volatility. I think it would be nice if other charts for 0.5%, 0.25 %, etc. are available for reference.
The data was collected from Yahoo each night into an Excel spreadsheet and then the graph was created. The part that is a bit unusual is the percentage info. I don't have that computer running right now, but will post the spreadsheet in my journal later this weekend for anyone who wants to download it. I can't give you any clues to successful patterns as I haven't found that info on the net. I posted this link awhile back maybe you didn't see it. http://www.smallinvestors.com/SP500/GapSP500.htm
I hadn't thought of that. That might be interesting. Basically, it would require making a few more columns for the spreadsheet. I was just thinking that shifts in volatility are difficult to detect using conventional means. You have to compare the volatility around a changing mean which makes the changes hard to detect. I don't think I could modify the spreadsheet that much. Probably, I could include the columns for .50% changes, though.
Hi Heilbronner, You can create your own via whatever trade methodology your using and via whatever stats you consider to be important. I have such data but only on certain patterns I trade and only on the dates I was reveiwing... Thus, the info isn't accumulated. More importantly, you can easily do your own research or database on your favorite candlestick patterns. I recently did one on both ES and NQ via a particular type of Bullish Harami candlestick pattern. (Note: I have a list of about 7 different types of Bullish Harami's and currently only trade one type) You should start printing charts of every trading day and going back as far as your charting program has historical intraday data... http://www.elitetrader.com/vb/showthread.php?s=&threadid=28482 Your database will be more valuable and more accurate than the database of another trader that maintains info on intraday Japanese Candlestick patterns... this will help you to know your level of risk prior to entry. Why? Their backtesting...regardless if its mechanical or manually is based upon their trade management rules for that particular pattern which more often than not is different than your trade management rules. For example...that particular Bullish Harami pattern I mentioned above... I know the following for the past 3 years of the ES and NQ (will be doing YM soon)...data current til February 20th Friday and will ceased on the last trading day of next month: __________________ 1. What specific dates this particular type of Bullish Harami appeared. 2. Statistically...how many times after the trade signal for entry did it reached profit target 1, profit target 2 and profit target 3. 3. Statistically...how many times the initial stop/loss was threaten (moving within 2 ticks of the mental stop). 4. Statistically...how many times the initial stop/loss was threaten (moving within 2 ticks of the actual hard stop). 5. Statistically...how many times the hard stop was taken out and then the contingency plan gets activated (re-entry via a particular pattern)... did it reach its new profit target 1, profit target 2 and profit target 3. 6. Statistically...what's the average profit for profit target 1, profit target 2 and profit target 3... as in points. 7. Statistically...how many times the trailing stop (not the initial hard stop) of my remainders would have been picked off after profit target 1 is reached. __________________ Via the info above (my most detailed type of candlestick research)... I'm able to easily adapt the trade signal to the current market environment... For example...profit target 3 is only reached 7% of the time, profit target 2 is reached 23% of the time and profit target 1 is reached 67% of the time... Contingency Plan profit target 1 is reached 79% of the time. Therefore, the last 5 times I've actually traded this specific type of Bullish Harami candlestick pattern in the past week or so... I've been exiting 1/3 at pt 1, another 1/3 at pt 2 and another 3rd at pt 3. Starting next week...I'll be exiting 2/3 at pt 1 and exiting all remainders at pt 2 and won't even bother holding longer to see if profit target 3 is reached. Thus, I have a better understanding of my specific trade management rules and what to expect when I put on a trade via that pattern... allowing me to better trade the pattern. Like I said...if you did your own database, your own research...it'll be via your own trade methodology (not somebody else). Thus, the stats or results will be much more important than what somebody else saids and/or has backtested. Last of all...more importantly...you'll have a better understanding of how to tweak something when tweaking is needed. To some this sounds like hard work...for me (not a stats guy)...it is hard work. Yet...the payoff is worth all the research. Heilbronner...how did your TKTX trade turn out? http://www.elitetrader.com/vb/showthread.php?s=&postid=408901 P.S. My type of stats may not be the type of stats another trader would look for but its what I feel is important to me to be prepared for what to do when the pattern appears... helps me to know my level of risk prior to entry. All my studies are via manaually reveiwing past printed charts of historical intraday data. Thus, I use no backtesting software via my experience that such cannot even come close to producing accurate results to reflect my trade methodology. NihabaAshi
Hello NihabAshi: I learned something new from your post. It seemed as though I was the only one left who manually reviewed intraday price charts, but I see that you do the same. I appreciate the advantages that a skilled backtester has when using good software, but as you point out, it can be difficult to obtain statistics that accurately reflect a specific trading methodology. An additional benefit that one gets from manually reviewing data is "market feel". After going back through many charts over a period of years, I notice that I have developed "feel" for the way that price moves across a number of markets. This market "sense" seems to manifest itself in a kind of subconscious "anticipation" of movement, that I can call on when I am trying to find a place to enter or exit a market. Periodically, I see software that I might like to try that could automate the process, but ultimately, I think it is best to continue to do the job by hand. Good luck in the Markets, Steve46