Indicators of Nothing?

Greetings, I'm a 'long-term' trader who is looking to expand his horizons with shorter term trading.

I've noticed while browsing various forums that many day-traders appear to make frequent use of technical indicators in their trades. I've personally done extensive backtests on roughly forty of the more popular technical indicators using end of day data. Stocks were held for one week, and the indicators were used on the basis of whether to skip a trade or not. The results were dismal to say the least. From the stock universe I used, not a single indicator raised the probability of receiving a winning trade over the time frame (6+ years.) Some did outperform the underlying stocks during the bear market, and a few outperformed the underlying stocks during the bull market. And a very select few even managed to beat the basic buy and hold strategy slightly over the time frame -- but once transaction costs of more frequent trading were factored in, those select few too became losers.

The results, as you might guess, suggested to me that technical indicators in general are next to worthless, at least on longer time frames. Unfortunately, I have no (cheap) resource for historical intraday data, so testing any further is out of my reach.

What do you make of these results? Do you feel that technical indicators used intraday are somehow different than when used end of day in a way which causes them to work? Are their failings merely a result of them being public now, wherein the returns that might have once been gleaned from them have gradually been arbitraged out? Or are the successful results some day traders have had using technical indicators merely a result of random outliers that are unlikely to continue into the future?

In my view any single indicator on its own would not be better than a random approach. Have you tried combining indicators ? I do know that some intra-day trading approaches cannot be back tested and do give better than random approaches as has been continually demonstrated on the Technical Trader forum.


Paul

Hi MarkSA,
I've not conducted a poll or done anything to establish definite facts about the use of indicators by traders, but my general impression is that they are used increasingly as a filter to add / remove stocks from watchlists or as an alert to highlight changing conditions which might affect open positions. A good example of this is the well known Swing Trader Marc Rivalland who will not enter new positions if the RSI (Welles Wilder's) is overbought or oversold. Additionally, if he has an open trade, he will use RSI to take partial profits or close the trade all together. Some might argue that this is a rather negative use of an indicator, but if it stops you buying at the very top of a move (something I'm expert at!) - or helps you to take some or all of the profit at the top of a move, then it's a very useful and positive tool indeed. It seems to me that relatively few traders will use an indicator like RSI for set ups and entries, but I may be wrong - this is just my general impression.
Tim.

Hi All,

I have to agree with both Tim and Paul.

Indicators on their own have been proven to achieve no more a reliable entry than flipping a coin. Indicators are better used when combined together or when confirmed by price action.


Thanks

Damian

Price action must confirm indicator signal

Hello

I agree with everything said above, indicators give an idea about how oversold or overbought the market is, but this don´t mean that the market can´t keep oversold (for example) for long time, as it happened last weeks. In this cases, the strategy called "pop corn" gives better results than other, this means buy when the indicator shows oversold and vice versa.

Using more than single indicator is usually better, for example, Bollinger Bands and RSI, or Volatility bands with CCI. Always trying to keep the chart clean enough so that we can keep our eyes on the price level and volume, very important.

However, my singular point of view about this, I´d rather use more time thinking of my money management strategy and risk management, I think this is the key to survive at the market.

You can get in using a coin, but if you manage your position (bear or bull) wisely, you won´t get naked.

Happy trading.

mark

Mark -
For short term trades I use Bollinger bands an stochastics a lot plus RSI and MACD some what

In this example is the Dow 15-min chart. I trade the Dow futures and generally buy when the candle is under the Bollinger band and stochastics under 20 - better to wait until it goes back above 20. Then I see above the top band and stochastics over 80 (or falling under 80) this is tempered by if it is at resistance or support etc.

It is pretty simple and very accurate. In this chart the RSI is set to 14 but I often use 9 for day trades.

On this last buy below the bottom band and stochastics under 20 and sell above 80 and above top band today was worth 140 points or $700 per e-mini contract. Each contract ties up about 1900 so the gain is 37%. Watch it and you will see how often it is a good money maker

Yes, I did do a little work with combining indicators. I ran a few tests with a few different types of indicators initially testing instances where they confirmed each other. I also ran a backtest on the top three 'best' performing indicators of the initial backtest, but there were still no significant improvements over the buy and hold strategy unfortunately.

To clarify, stocks weren't chosen based solely on what the indicators stated. I began with a criteria which chooses stocks which have had a large price change recently. Historically, based on the backtest, this resulted in decent returns, but I noticed that there were often stocks in the mix which would take a hefty plunge shortly after buying in. I was hoping to eliminate even a few of those with the help of the indicators in order to boost returns higher.

Perhaps O'Neils CANSLIM appoach may suit you better or Weinsteins MA approach to stocks?