As of 06/25/2019
Industrials: 26,548 179.32 0.7%
Transports: 10,110 87.48 0.9%
Utilities: 821 5.99 0.7%
Nasdaq: 7,885 120.98 1.5%
S&P 500: 2,917 27.97 0.9%

YTD
+13.8%
+10.2%
+15.2%
+18.8%
+16.4%

26,800 or 25,400 by 07/01/2019
10,700 or 9,900 by 07/01/2019
830 or 780 by 07/01/2019
8,150 or 7,575 by 07/01/2019
3,000 or 2,875 by 07/01/2019

As of 06/25/2019
Industrials: 26,548 179.32 0.7%
Transports: 10,110 87.48 0.9%
Utilities: 821 5.99 0.7%
Nasdaq: 7,885 120.98 1.5%
S&P 500: 2,917 27.97 0.9%

YTD
+13.8%
+10.2%
+15.2%
+18.8%
+16.4%

26,800 or 25,400 by 07/01/2019
10,700 or 9,900 by 07/01/2019
830 or 780 by 07/01/2019
8,150 or 7,575 by 07/01/2019
3,000 or 2,875 by 07/01/2019
 
This article discusses a trading setup by Thomas Bierovic for ascending and descending triangles. Tests results for ascending triangles are presented.
The Bierovic setup uses ascending triangles to ride upward breakouts and capture profit from trending moves. He also has a similar setup for descending triangles with downward breakouts, but I didn't test it.
Performance results for ascending triangles show that the setup makes 0.8% per trade. If you look at trades excluded because the pattern was too short or price was below the exponential moving average(s), then the setup makes more money: 1.7%. Regardless, the amount of profit on a $10,000 trade is just $80 (included trades) or $170 (for excluded trades) with an average hold time of 15 days for both.
The maximum drawdown was 21% with a hold time loss of 30%. The average values for both were small, less than 4% for each.
Analysis of the setup suggests that the entry rules do more harm than good. Taking all trades would improve profits. Taking only excluded trades also increases profits.
The September 2002 Active Trader article titled, "The ups and downs of triangles," by Thomas Bierovic, discusses trading setups for ascending and descending triangles.
Here are the rules he used for ascending triangles with upward breakouts.
Here are the rules for descending triangles with downward breakouts.
I did not test this setup, but I include it here for completeness.
I used ascending triangles with upward breakouts starting from 1991 to 2011 in 627 stocks (I actually used more stocks to find those that held triangles). That gave me 1,061 triangles. I found each triangle manually, over the years.
I programmed my computer and used daily price data to test the setup for ascending triangles with upward breakouts only. Bierovic says that this setup works with intraday data but instead of a dime, use two cents away from the stop values.
The test used $10 commissions ($20 round trip), but did not compensate for slippage or other factors. Each trade bought 100 shares of the stock and did not combine profits from prior trades. That meant all trades were taken (meaning the portfolio never suffered from a lack of cash to buy a position).
I tested his ascending triangle setup for upward breakouts only, and the following table shows the results.
Test  Completed Trades  Excluded Trades 
Average gain per trade  0.8%  1.7% 
Average gain on $10,000  $80  $170 
Maximum drawdown  21.1%  30.7% 
Average drawdown  3.6%  4.6% 
Maximum hold time loss  29.6%  27.8% 
Average hold time loss  3.5%  4.1% 
Win/loss ratio  60%  63% 
Average hold time (days)  15  15 
Notes:
Of the 1,061 trades, 19 of them were excluded because they were less than 15 price bars long. Had they been included, their gains averaged 1.3%, which is above the 0.8% reported by the setup. Thus, it makes more sense to include short patterns than exclude them.
233 trades were excluded because the start of the triangle was below the 13 dayEMA. Those trades had gains averaging 2.6%. Similarly, 288 trades were excluded because the start was below the 55day EMA. Those gains averaged 2.1%.
Just 5 trades at buy time were below the 13day EMA but those excluded gained 6%. 87 trades were below the 55day EMA at buy time. They showed gains averaging 2.9%.
Since each trade can be excluded for any of the five setup rules, a total of 469 trades were excluded, some because the EMA did not have enough data in the stock to calculate it (that means the triangle was too close to data start).
Based on this analysis, it appears that the pattern length and use of two moving averages do more harm than good.
If all excluded trades were included, the gain would average 1.1% with a 61% win/loss ratio.
Next, I dissected the stops. Where was the trade being stopped out most?
The stock hit the initial stop 130 times (22% of trades) and lost 3.4%, on average. 206 trades, or 35%, were stopped out at break even (break even includes the cost of round trip commissions: $20).
The third type of stop is one the stock reaches the full height target and the setup tightens the stop to a dime below the prior candle's low. That stop gets hit 23% of the time but by then the stock averages a gain of 9.1%. The final stop is the swing low stop. That gets hit 20% of the time for a loss of 0.4%.
The following table shows this information in tabular form.
Stop Type  Hit  Gain/Loss 
Initial stop  22%  3.4% 
break even  35%  0.0% 
Profit target reached  23%  9.1% 
Swing low  20%  0.4% 
The figure is of XL Group (XL) on the daily scale. It shows an ascending triangle spanning 20102011. The top of the pattern is at 22.30 so a buy stop at 22.40 is the entry price.
Entry occurs on the day shown, then price makes a quick move up. The initial stop would be at 22.08, which is 10 cents below the low the day before the breakout (point C in the figure).
The height of the ascending triangle is 22.30  21.53 or 77 cents. The full height target is at 22.30 + 0.77 or 23.07. The halfheight target would be 22.30 + 0.77/2 or 22.69. The stock exceeds the halfheight target on the entry day, so the stop gets raised to break even, or 22.50 (assumes 100 shares with $20 round trip commission).
The next day, price makes another strong move up, exceeding the full height target. Thus, the stop uses the prior candle's low minus a dime. In this case, that's 22.24  0.10, or 22.14. A day later, D, the stop gets raised to 22.76. On the sell day, the stop is raised again to 23.31, a dime below D, before trading begins.
The stock closes lower and hits the stop, closing out the trade for a gain of 4.1%
 Thomas Bulkowski
See Also

George Westinghouse invented the electric chair. Think about that the next time you get in an elevator.