The Study of Fibonacci Patterns
Simone Johnson | Posted: January 2, 2011 at 5:26 am | Updated: April 11, 2016 at 12:04 pm
Categories: Pattern Trading Performance
There’s been a lot of work done on Fibonacci Patterns and although there has been some great work by some wonderful people, we decided to do our own study on a large number of trades and their potential results. I also wanted to shed some light on the aspects of the research we did and give some clarity to the study.
The “Rules” To Be Marked As “Successful”
We looked at approx 100,000 of the main patterns in the study. In order to mark a pattern as “successful” we looked for the following conditions:
These patterns were on a variety of different time frames (5M, 15M, 30M, 1H, 4H, D, M).
A reversal had to occur once price entered the PRZ, but not before price exceeded the furthest point of the PRZ. This would indicate a loss and not a successful pattern.
A successful reversal must reverse 15% from when price entered the PRZ to where it exited the PRZ. If it did not reverse 15% and stayed inside the PRZ, again this would be a failed pattern.
The patterns would not be filtered in any way, so no matter the size of the PRZ or the structure of the pattern, the pattern would be included in the study.
After a trade moves 15% we automatically move stops to preserve our capital. The breakeven trades are counted as successful for a few reasons: A) Any professional trader would agree that a free ride in the market without losing money is a success! B) With the way the forex brokers work, you get rebates based off the volume you trade. Thus because a trade is breakeven you probably still get a 1 pip rebate from your broker. If you don’t, I would consider finding one that would give you a rebate.
The results were on track with what we were expecting. Let’s talk about the patterns first and then the success rates associated with those patterns.
Gartley: This pattern had the highest NUMBER of patterns produced from the 100,000 pattern study. This is not really a surprise. The first “fibonacci pattern” discovered in the 30’s was the Gartley so it makes sense that it was the highest producing pattern if it’s the most common one to be found. However, the Gartley also had the lowest success rate.
Butterfly: The second most common pattern found in the market is the butterfly, with an 88.44% success rate (remember these numbers take into account breakeven trades).
Bat: This is one of my favorites just because of the small PRZ or Potential Reversal Zone). This of course is your window of opportunity to get into the trade. It’s a common pattern and is always a nice one to trade but just make sure the risk isn’t too big; some of these patterns based off their study have a large PRZ.
Crab: This is one of the most rare patterns and it’s amazing how few times a month they will show themselves. But you will also notice with a 90.89% success rate, it’s one of the few patterns that hit that 90% level.
One Important Question Left to Answer
One of the most popular questions we get after reading a study like this is: ‘What is the % of profitable trades where the trades hit their target?’
Easy answer: Anywhere between 70-75%. When we ran the numbers with targets in mind, the success rate with JUST the pattern alone (not using SONAR to bump up the statistics) gave us a 70-75% success rate. We were using the B point as our first target in that example.
What can you expect from a 70-75% win ratio?
The managed account we ran performed to around a 65% return by the end of the year and that was with a win ratio of 68.5%. So we actually did less than the 70-75% average. This is probably because December was a frustrating month with 19 losing trades in a row, however we still came up with a handsome profit. How? It all has to do with how much you risk. More on that another time, but until then, Happy Trading.
Ps. If you have any questions, I’m happy to answer them. Please post them in the comments below.