Has anyone programmed and backtested the anaSuperTrendM1 system. I am curious. If not can someone help me do it? I am looking for a simple stop and reverse sytem using the signal arrows. Is this easy to do in Ninja and run a backtest to see how it would perform? Tks
Yes, it is relatively easy to create a strategy with the Supertrend indicators. However you would need to add some filters to make it profitable. As it is a breakout strategy, it would require a time filter, as volatility breakouts will not follow through during all times of the day.
The following user says Thank You to Fat Tails for this post:
Ok great. Tks fats. I assume its easy to set up a basic strategy first without filters and then can add them later? I would like to see where the results are without filters first. Also how difficult is it to run a backtest on it as far back as i want to see the historical results just based simply on market stop and reversal orders at the signals? Can NinjaTrader perform the analysis that Tradestation can on stuff like this? Not sure this matters but would like to run it on my 2 range Renko-Hybrid chart if possible. Tks
2 nice filters for a simple and good working SuperTrend strategy you can realize by using 2 of my free on elite published tools (trade only if equal SuperTrends for Symbol,CumUD and one most time correlated market):
I cannot hear the word Renko anymore. Renko bars are really bullsh... A few remarks:
(1) The SuperTrend is designed to adapt to volatility. If volatility is low, the stop is closer to price action, and if volatility is high it is further away from price action. Volatility is measured via the average true range. If you use the Supertrend on Renko bars, you have eliminated volatility from the equation, as all bars have the same size. You don't need a SuperTrend on a Renko chart, this is simply nonsense.
(2) With 2 range Renko-Hybrid bars, you enter the realm of high frequency trading. It is not realistic to trade such small bars. You will only produce losses, as the commissions, the slippage and the spread - if you enter with market orders - will eat up your inexistent edge. During a news release about 50 or 100 2-Renko bars can be produced per second. The strategy will no way be able to execute signals generated that fast, because you will get 20 new bars, before an order is even filled. If you want to do HFT trading you need an exchange seat for low commissions and a fast line to the exchange without passing by the credit control of the broker.
(3) That said Renko bars will produce nice profits on replay and simulation, but may produce considerable losses if you use them with real money. Renko bars are the favourite of simulation traders.
(4) Renko bars cannot be backtested, as they show a false open, a false high, a false low and a false close. With BetterRenko backtest is possible, I do not know whether Renko-Hybrid bars are backtestable.
(5) Code a strategy for Renko bars and backtest it, is probably the most difficult thing ever you can do with NinjaTrader. If you are inexperienced in coding strategies, coding a Renko strategy is definitely the last thing to do.
The following 8 users say Thank You to Fat Tails for this post:
Fat Tails has say all to Renko or HybridRenko with SuperTrend and his vola logic!
=> "SuperTrend" work very fine and good backtestable with the free "MomentumBars" (dont use the "NoGap" mode!)
=> VolumeBars work nice
=> dont use range bars, while the same ATR vola problem
-> you can try any more bartypes, if its give different resulting barsizes
The following user says Thank You to TimeTrade for this post:
Yes, its is a little bit tricky, The standard MomentumType work better with higher PeriodValues, when its give more candles with true shadows... The best is any bartype with different body and shadow sizes on the most bars.
(my problem, i use only my own types and my experience with standard types is limited)
Hi Booker, I'm not trying to put you off or anything but I've spent considerable time with this indicator in various strategies on a multitude of timeframes on another PC running 24 x7. The problem with nearly all indicators is some do well in trends and some do well in balanced markets. The problem is it's very difficult to determine the state of the market in a backtest and then apply the right indicator to the right market conditions. For this reason 95% of indicators will therefore result in overall unprofitable trading in my opinion.
Another approach is to focus more on market theory, researching volatility and understanding where inventory is resting at various price levels. As you become more familiar these concepts you will understand and see the imbalances in a market and be in the position to take advantage of them.
To put it in perspective, I used to go to car auctions years ago with a car dealer friend of mine. He was selling bigger cars, and surprisingly when I was there he bid on a small Ford and won. I asked him why he bid on such a small car and he said that the market was changing to more fuel efficient cars and that he was happy because some of those Fords sold a few weeks at the same auction at much higher prices. He thought he could sell the little Ford at a good profit.
In other words he recognized the market was changing and bought the car below value and intended to sell at or above value. When I asked him the question why he bought the small car, notice he didn't say "Well price crossed over the MACD zero line and the EMA 8 was above the SMA 10, Price was at the Fib 127% level and is also above the R1 pivot point etc" etc.
Having said that if you wish to learn more about strategies etc I found @Ducman's strategy extremely good and the thread is very interesting. He has a version there you can run and modify with the supertrend.