As of 01/17/2025
Indus: 43,488 +334.70 +0.8%
Trans: 16,431 -169.75 -1.0%
Utils: 1,013 +1.53 +0.2%
Nasdaq: 19,630 +291.91 +1.5%
S&P 500: 5,997 +59.32 +1.0%
|
YTD
+2.2%
+3.4%
+3.0%
+1.7%
+2.0%
|
44,700 or 41,600 by 02/01/2025
17,200 or 15,700 by 02/01/2025
1,050 or 950 by 02/01/2025
20,500 or 18,670 by 02/01/2025
6,100 or 5,700 by 02/01/2025
|
As of 01/17/2025
Indus: 43,488 +334.70 +0.8%
Trans: 16,431 -169.75 -1.0%
Utils: 1,013 +1.53 +0.2%
Nasdaq: 19,630 +291.91 +1.5%
S&P 500: 5,997 +59.32 +1.0%
|
YTD
+2.2%
+3.4%
+3.0%
+1.7%
+2.0%
| |
44,700 or 41,600 by 02/01/2025
17,200 or 15,700 by 02/01/2025
1,050 or 950 by 02/01/2025
20,500 or 18,670 by 02/01/2025
6,100 or 5,700 by 02/01/2025
| ||
Updated with new performance information on 1/17/25.
To determine the configuration of the best performing bullish fakey, I ran tests. This article describes what I found.
I was introduced to a fakey pattern in an email in May 2018, so I did some investigating. The version of the bullish fakey, as I have implemented it, has been tuned for performance. By that, I mean I tested the last two bars of the pattern (leaving the inside day alone) and tried various combinations of high and low to see which variation worked best.
The improved bullish fakey shown in the picture below is not the best performing, but it is the best performing with a high occurrence rate (the best performing is too rare and the performance difference is minor anyway).
Improved Bullish Fakey
|
The above numbers are based on 4,219 perfect trades. See the glossary for definitions.
The picture on the right shows a version of the traditional bullish flakey pattern.
I couldn't find any specific configuration for the pattern, just general pictures. I show all candles in shades of gray because of the lack of details. The candles could be white or black.
The size of these candles, their color, their position in relation to each other may have been defined by someone, somewhere, but I couldn't find it. So the figure is a generic representation of the traditional bullish fakey.
Candles 1 and 2 make up the inside day pattern. That is, candle 2 fits inside the high-low range of candle 1. You'll often see candle 1 as white and candle 2 as black, but they can be either color.
Candle 3 shows a strong downward thrust. Often this candle is portrayed as black (bearish, with a close below the day's open) and tall.
Candle 4 is the reversal of the downward thrust. I write "thrust" for both candles 3 and 4 (instead of breakout) because of the lack of configuration details on the pattern. Often this candle is portrayed as white and tall.
Do NOT use this picture to define where the open-high-low-close prices should be in relation to the other candles. Again, I found no rules for that.
I used daily price data in 519 stocks from January 1, 2010 to 6/5/2018 in the test. Not all stocks covered the entire period. I set a minimum price of $3 for each trade.
I found 4,219 trades during that bull market run.
I tested the configuration of the last two candles in the pattern and ignored the inside day pattern (the first two candles). As long as price on day 2 was inside the high-low range of day 1, that was fine with me (ties not allowed).
I tested 20 variations of price, moving the last two candles up and down versus the high-low range of candles 1 and 2. I did not care about candle color nor the height of any price bar.
Rather, I focused on the high to high and low to low relationship of the candles.
For example, I asked was candle 3's high above candle 1 and it's low below the low of candle 1?
I measured performance from the top of the pattern to the ultimate high. The ultimate high is the highest price the stock reaches before tumbling either 20% or below the bottom of the chart pattern, measured high to close.
Thus, the measure is a representation of perfect trades. However, it does allow a trader to compare one chart pattern to another, regardless of stop loss, exit strategies, and so on.
If you wish to review the data, you can download the Excel spreadsheet here (~ 2.2 mb). The tests assume you've tested for the first two bars, an inside day.
The table refers to the picture on the right. Here is how to identify an improved bullish fakey. In essence, it's a four-day downward price trend that ends with the last candle in the pattern.
If price continues lower, then it's not an improved bullish fakey. Price has to rise above the top of the pattern to be valid. I'll explain this in trading tactics.
Characteristic | Discussion |
Inside day | The high of day 1 is above the high of day 2. Low of day 1 is below the low of day 2. Price ties not allowed (not tested, actually). |
Candle 3 | The high of day 3 is below the high of day 1. The low of day 3 is below the low of day 1. |
Candle 4 | The low of day 4 is below the low of day 3. The high of day 4 is below the high of day 3. |
Confirmation | If price drops below the low of day 4, it's not an improved bullish fakey pattern. Price must first rise above the high of day 1 to be a valid fakey pattern. |
Candle color | Not tested |
Candle height | Not tested |
Inside day | Configuration not tested, only that day 2 fits inside the high-low range of day 1. |
Place a buy stop a penny above the top (high) of candle 1. For example, if candle 1 has a high price of $10.00, then place a buy stop at $10.01.
I show the buy stop in the figure on the right as a thin green line (the confirmation line, but entry should be a penny above candle 1).
If price drops below the low of day 4, cancel the buy stop. You're not looking at a bullish fakey.
In my review of this pattern, too many pattern see price continue lower. That's why price must reverse (move higher) immediately after candle 4.
I show failure of the potential pattern as the thin red line in the figure.
This pattern has a very high failure rate: 47%. That means almost half of all patterns will see price rise less than 5% above the top of the bullish fakey before reversing and closing below the bottom of the pattern.
Also, consider looking for bullish fakey patterns which rest on support, or look at the price trend leading to the start of the pattern (that is, a rising or falling inbound trend). I haven't evaluated any of these situations in my analysis of the bullish fakey.
This is a busy chart, so let's go through it.
There are several bullish fakey patterns shown. From left to right, numbers 3 and 4 show the start and end of a fakey. The rise leads up to 2 before price reverses, posting a gain of 4%.
Pattern 2-5 (the bar to the right of 5) is another bullish fakey. This one tops out at the same ultimate high but entry is made when price climbs a penny above 2, leaving a rise of just 9%, and that's without commissions and fees.
Fakey ending at 1 is the one we'll be discussing in a moment.
The last bullish fakey on the chart begins at 6, which is the first candle of the inside day. It's four price bars long. If you were to trade this one perfectly, you'd make 2%
The inset shows the fakey ending at 1. It's easy to spot. Look for an inside day followed by two days of consecutively lower prices. After the fourth bar in the pattern, price has to make a higher low. If it doesn't, then you don't have a bullish fakey.
In this example, price doesn't drop below the low at 1 before zipping higher and confirming the pattern when it rises above the pattern's high.
I show the confirmation price and entry price on the chart (the same price). Place a buy stop a penny above the high of the pattern to get you into the trade.
Notice that this bullish fakey rests on support, which I show as a horizontal red line. Also notice that this fakey forms the right shoulder of a head-and-shoulders bottom chart pattern. The left shoulder is the low at 4, head is at 5, and right shoulder is at 1.
Using the down-sloping neckline of the head-and-shoulders might have allowed you to enter this trade at a slightly lower price.
Price rises to the ultimate high. If you bought into the stock at the entry price, sold at the ultimate high, you would have made 12% if you could trade for free, less after commissions and fees.
The setup for the following tests is different from the ones previously described.
Trading using a target exit is simple to explain. Look at the adjacent chart.
I boxed the fakey pattern in red. Entry happens when the stock climbs above a buy stop placed a penny above the top of the 4-bar pattern. The target would be twice the height of the fakey added to the top of the pattern. In this case, the stock does not rise to meet the target. Instead, the stock trips the stop loss order, which is placed a penny below the bottom of fakey.
For a more detailed explanation of the method I used to test the Fakey, see the link.
As explained in the example above, I used a target exit placed twice as high as the height of the fakey pattern (from highest price bar to the lowest one in the fakey). I placed a stop loss a penny below the bottom of the pattern and waited for one or the other to trigger.
The following table shows results for bull markets with upward breakouts and an inbound price trend either up or down. I used 490 stocks in the test.
Metric | Fakey In Uptrend | Uptrend Benchmark | Fakey in Downtrend | Downtrend Benchmark |
Trades | 5,025 | 5,794 | 4,189 | 5,187 |
Average profit/loss per trade | $110.88 | $88.58 | $119.23 | $100.27 |
Win/loss ratio | 42% | 41% | 43% | 42% |
Average hold time (days) | 29 | 26 | 29 | 26 |
Winning trades | 2,109 | 2,368 | 1,782 | 2,154 |
Average gain of winners | 10% | 9% | 10% | 10% |
Average hold time of winners (days) | 36 | 36 | 38 | 37 |
Losing trades | 2,916 | 3,426 | 2,407 | 3,033 |
Average loss | -5% | -5% | -6% | -5% |
Average hold time of losers (days) | 23 | 22 | 23 | 23 |
The Fakey pattern in stocks for both up and down trend directions beats the benchmark. The best is in an uptrend with the average profit per trade of $110.88 versus $88.58, for a 4-bar pattern.
The associated chart shows an example of how I tested the fakey pattern in exchange traded funds (ETFs).
The red box highlights fakey. Two days after the pattern ends, we have an upward breakout and buy the ETF.
A stop loss order placed a penny below the bottom of the fakey helps limit the effects of adverse moves.
Computing the height of the fakey (times two), added to the top of the pattern, gives a target price at which to exit. As you can see, the ETF climbed far enough to reach the sale price (Target Exit).
This is the same test as the prior one except I used 94 exchange traded funds (ETFs) instead of common stocks.
Metric | Fakey In Uptrend | Uptrend Benchmark | Fakey in Downtrend | Downtrend Benchmark |
Trades | 896 | 6,926 | 723 | 5,676 |
Average profit/loss per trade | $132.26 | $85.22 | $72.38 | $83.76 |
Win/loss ratio | 45% | 44% | 41% | 43% |
Average hold time (days) | 33 | 28 | 33 | 28 |
Winning trades | 405 | 3,018 | 298 | 2,441 |
Average gain of winners | 7% | 7% | 8% | 7% |
Average hold time of winners (days) | 41 | 38 | 41 | 44 |
Losing trades | 491 | 3,908 | 425 | 3,235 |
Average loss | -4% | -4% | -4% | -4% |
Average hold time of losers (days) | 23 | 22 | 30 | 26 |
Fakey in an uptrend does well against the benchmark but should be avoided in a downtrend.
This chart of a cryptocurrency is similar to the other charts for this test. I placed a buy stop a penny above the top of the Fakey and a stop loss order a penny below the bottom of the pattern.
In this example, the crypto broke out upward and continued higher in a stair-step move upward until reaching the sell target.
This is the same test as the prior one except I used 29 cryptocurrencies.
Metric | Fakey In Uptrend | Uptrend Benchmark | Fakey in Downtrend | Downtrend Benchmark |
Trades | 76 | 311 | 67 | 298 |
Average profit/loss per trade | $392.26 | $358.38 | $334.91 | $250.23 |
Win/loss ratio | 55% | 53% | 51% | 47% |
Average hold time (days) | 12 | 11 | 12 | 11 |
Winning trades | 42 | 166 | 34 | 141 |
Average gain of winners | 13% | 13% | 13% | 13% |
Average hold time of winners (days) | 14 | 11 | 13 | 14 |
Losing trades | 34 | 145 | 33 | 157 |
Average loss | -7% | -7% | -7% | -6% |
Average hold time of losers (days) | 13 | 10 | 8 | 9 |
Fakey in beats the benchmark in both up and down trends. I'd be cautious of the results because of the low sample counts.
-- Thomas Bulkowski
Support this site! Clicking any of the books (below) takes you to
Amazon.com If you buy ANYTHING while there, they pay for the referral.
Legal notice for paid links: "As an Amazon Associate I earn from qualifying purchases."
My Stock Market Books
|
My Novels
|
The secret of success is sincerity. Once you can fake that, you've got it made.