（转帖）Indicators are overhyped

	说明：      这是一篇关于指标应用方面的文章。      有英文比较好的朋友，帮忙翻译一下，供大家看看     Indicators are overhyped... The truth about them First of all, I have at times spent days/weeks/months backtesting various indicators from morning to evening and strategies through backtesting tool on my charting software. I have also read MANY books seen MANY videos and read a lot of indepth look at the codes behind indicators. I have also coded my own indicators. The more I read and test the more the bitter truth sets in... What are indicators? They are nothing more than some PRICE of a currency graphed in a certain way. For example a candle is nothing more than OHLC of a certain length of tick data ploted in a visually appealing way. RSI measures the ration of up and down candles, thats IT. All these overbough/oversold rubbish is not what the indicator is showing! If you use RSI as an overbought oversold indicator than you are misapplying it. For example if lets say you have 10 up candles each 2 pips and one 100 pip candle down then RSI (11) would tell you that the currency is overbought!!! Yet it is clearly not.... Stochastic shows where the candle is closing relative to its body, if the candle is closing higher then stochastic rises and vice versa.. Yet what if the stock/equity/whatever closes to the bottom of the shadow (such as gravestone doji) yet the price rises, the stochastic could be 0 yet the price is rising!!! So much for it accurately predicting anything... Fact: Indicators (such as RSI, roc, ma's, candles) NEVER predict, at best they show a CERTAIN aspect of price's action. An unexpected NFP (or any other major news) data can easily disregard ANY technical.... No technical indicator that can predict how people will feel "tomorrow at 10:45 am" or how much fresh meat will join the market today tomorrow or next year + 5 days... This applies to fundamental analysis as well... Fact: Many indicators are almost identical, and since they ALL look at the same data you could make a case that they all are virtually identical. Fact: In order for a market to remain a market there MUST be a seller for each buyer. There must be a person that will be willing to take an opposite side of the trade. If everyone knows that "Oh the indicator XXX says that stock is over bought/sold so I only short/long it" then that pattern will not work out. Decades ago people used to make a KILLING using moving averages. Try that today... I automaticly backtested DOZENS of moving average combination of 3+ years of data (the max my charting software allows for short term candles) and they all bombed if spread is taken into account. Even today the so called pros are still making a buck from indicators, by selling them.... Today with the computers and signals being fully automated anyone can use indicators thus any edge that was found in indicators is now gone. There is no need to count them by hand. Everyone can instantly see what "Triple smoothed cant fail redon stochastic XYZ" is showing. Try calculating DMI or something like that BY hand for many years worth of data. If in the old days those people who were good at math (or who had the money to hire people) did have an advantage, but today it is gone.... Too bad that the computers brought positives and well as negative changes to trading... [ 本帖最后由 xiujian 于 2006-3-14 23:39 编辑 ]

	嘿嘿，技术指标无用论啊，市场其实就是斗心理预期的，可是没有技术指标就没有那些专家了。

	作者把技术指标看作垃圾。但是他并没有提出不依靠技术指标，又能如何赚钱？跟着数据跑？ 他说对了一点，技术指标不是万能的。 There are should be some cases that the indicators will fail. If not, everybody will win. 市场博得是几率，只要指标给我们的启示盈利大于亏损就好了。若是反过来的话，恐怕没人用技术指标了。经过市场验证的毕竟是合理的。尽信书则不如无书。 可能作者不会用指标，输了好多钱。。。