If using indicators, for signals and/or backtesting, make sure the chosen one takes account of the full range of price movement for each period and not just the close/current price. The most well known indicator that does this is Wilder's Directional Movement System. The problem with the ADX, however, is that it's two times removed from price, so the lag severely restricts its effectiveness.
If you could have just one indicator, which would it be?
- Thread starter JTrades
- Start date
Hi,
DMI and ADX are not perfect. Unfortunately, I have not been able to find anything better.
I have looked at Vortex Indicator, Choppiness Index, Kaufman's Efficiency Ratio (which is very similar to the Chande Momentum Oscillator), etc.
Just my two cents.
And a good two cents it is too.
I can see that at least one of us is switched on. The DMI is once removed from price and can be used independently of the ADX as an indicator, with less lag. I think the DMI, or a variation of it, is used by the inde benchmark as a starting point. In other words, properly backtest any indicator or combination of indicators against the DMI performance, and if they don't reach the mark then get rid. Find something that outperforms the DMI and ratchet it up from there.
I say 'properly backtest' because most backtesting methods are plain wrong. This is why you first need to find an indicator and backtesting method that accounts for the full range of price movement throughout each period.
I can see that at least one of us is switched on. The DMI is once removed from price and can be used independently of the ADX as an indicator, with less lag. I think the DMI, or a variation of it, is used by the inde benchmark as a starting point. In other words, properly backtest any indicator or combination of indicators against the DMI performance, and if they don't reach the mark then get rid. Find something that outperforms the DMI and ratchet it up from there.
I say 'properly backtest' because most backtesting methods are plain wrong. This is why you first need to find an indicator and backtesting method that accounts for the full range of price movement throughout each period.
What is the use of posing this question?
There is and there will never be 1 indicator that will give good trading results because no a single indicator is able to analyse all market conditions and give you a clear signal. Markets are too complex to fit in 1 indicator. You need a good system, not a good indicator.
There is and there will never be 1 indicator that will give good trading results because no a single indicator is able to analyse all market conditions and give you a clear signal. Markets are too complex to fit in 1 indicator. You need a good system, not a good indicator.
Don't you believe that "indicator" is a tool like a trendline and is not a system? They alert me to possibilities but as far as being an all that you need for entries/exists, but depending on one's knowledge of price/emotions, need far greater than an indicator. I know when many were first made, creators might have thought it was all one would need and where many new traders would like, but in reality, indicator show one bit of information of looking at the forest. Price structure shows many times when an indicator show us when it is not smart to keep taking entry signals. One might get a signal to sell when an upsloping trendline on larger timeframe clearly shows to buy, so in my trading, it is the times when not to use an indicator is as important as when there are signals. I think most of us use strengths of signals to decide whether to take some trades, lower timeframes comes last for me and higher are stronger.
I also prefer slower lagging indicators, too fast means too many false signals, too much trading. By learning the indicator well, often gives us clues of divergences that are not always present under normal circumstances.
And I agree that good systems do well, but keeping good tools in toolbox good to have as well.
An "indicator" is indeed a tool. You need several tools to trade the markets.
If you have to repair cars you need a lot of tools, with 1 tool you cannot repair all problems that cars can have. Trading is the same in this aspect.
So give us a number. How many indicators do you think are needed to adequately analyze all market conditions?
Impossible to answer. It depends of what the indicator is doing, how good he is. I don't know your indicators so...
Some people repair cars with 5 tools others need 12 for the same job because their tools are not so good.
Some people repair cars with 5 tools others need 12 for the same job because their tools are not so good.
Then you can't say with real certainty that one indicator isn't enough. As you say, it depends on what the indicator is doing.
I find your car repair analogy to be inapplicable. Cars can go wrong a million different ways and full-service car repair facilities have dozens if not scores of different tools ... no 5-tool garage tinkerer nonsense for them. Fortunately price movement analysis isn't nearly as complicated as all the things that can go wrong with a car. After all, price can only move in two directions: up and down.
So you can trade very successful with 1 indicator? I am sure it is impossible. I mean to trade VERY successfully, because you can make some money with 1 indicator, but never have very strong performance.
My car analogy is not wrong. A car can be repaired or not be repaired, so also only two possibilities.
If it is so simple and prices go only up or down, why are so many people then losing money?
I know that with 1 indicator you will never make the return that I make. That is for me the ultimate proof that 1 indicator is not enough. But I don't want to discuss my trading performance so I think the discussion stops here. We can discuss but when it comes to close to my private life I will not discuss any further. My privacy is more important for me then becoming a guru or an idiot on this forum.