capt j ... a little elaboration ...
I know Newton's laws inside and out ... nuff said on that. I am also proficient at computer programming. I also have a great deal of self interest in including any new ideas in my trading models. So why not include EW in my models if I can see a reason? I would be stupid not to.
When I trade, I use one of two methods: a trend following system; or an overbought/oversold system. Either way, the computer does the figuring because I'm too biased. Both systems are based on closing prices, so they are end-of-day systems, and changes of position are made rather infrequently. When the market has a general direction, I use the trend system; when it doesn't (i.e. the market is flat on a short-term, usually weekly, basis), I use the other system, which pays no attention to trends. My interest in EW was to find out if there is something there worth incorporating in either model.
A trend system does not just use the rearview mirror. Trends are channels which are usually wide enough to allow for price movement without triggering trades; so as I will explain, one can also look out the side windows and the front window. The channel I use (as well as most channels) are almost always more than two ATR wide, and it is uncommon that price moves more than one ATR in one day. When it does, the most common result is that the channel is automatically widened by the formula defining the channel. Also, channel boundaries are usually determined by trading highs and lows, not just closes. So, depending on where price is within the channel, it is very easy to predict what will happen during the day -- trade or no trade -- and also accurately predict what the channel will look like on the following day. Most indicator formulas, such as for instance that for RSI, can be easily rewritten so that realistic estimates for the following day’s prices can be plugged in for purposes of determining a likely range for the next day’s RSI. One can use this method to even look two days ahead, though with significantly less certainty.
I don’t have any statistics on this, but my impression is that when average traders have prediction schemes, they typically invest too early in the hope that they are catching a better price. If they have any sophistication at all, they set tight stops, or are at least glued to the monitor all day with fingers on keyboard. The even more sophisticated traders hedge with options strategies and walk away.
I have a prediction scheme on my trend system. I use it to set an alarm to warn me that I need to pay attention to the market on the following day because there is a possibility of a pending trade. On most days, the alarm isn’t set, so I know that I can ignore the market. The system is currently monitoring QID/QLD, SDS/SSO, and UYG/SKF. There is no warning alarm set for any of those, so I know I can ignore the market tomorrow. I am currently only active in UYG/SKF. Since March 19 the system has had me change position from UYG to SKF only once, on May 15 to SKF from UYG. Since March 19 the paper gain is 61.9%.
So far, all that I have seen regarding EW is that it is in the same category as astrology or religion and requires more faith than knowledge. I believe I am totally unbiased on this. Most passionate practitioners of EW think it’s pretty darned good; but I see it as merely a mother’s love for her own child.
