The ultimate criterion for whether automated trading is socially useful is the profit-and-loss test.
If the financial institutions relying on these programs blow up in the long run, we’ll have our answer — if only the government and Fed would stay out of it.
Archive for March, 2011
Breakout trades are much less likely to be successful when there isn’t a strong trend in the direction of the breakout (relative to the timeframe on which the trade would occur.)
So what’s a trend? The basic definition is that an uptrend is a progression of higher highs and higher lows, and a downtrend is a progression of lower lows and lower highs. More precisely, in an uptrend, each swing high is higher than the previous one, and each swing low is also higher than the previous one. In a downtrend, each swing high is lower than the previous one, and each swing low is also lower than the previous one.
In a strong trend, price moves persistently in the same direction, and retracements are usually relatively small in price and/or time—preferably both. In a really strong uptrend, the high of each bar is higher than that of the previous bar, and the low of each bar is also higher than that of the previous one. In a really strong downtrend, the low of each bar is lower than that of the previous bar, and the high of each bar is also lower than that of the previous one.
By contrast, when the there is a lot of overlap between successive price bars, when retracements are deep and/or long-lasting, and when there is not much net forward movement (up or down) from swing high to swing low and from swing low to swing low, then either there is no trend, or else it’s relatively weak.
In addition to the persistence and strength of the trend, the probability of success of a breakout trade also depends on the “gravitas” (or importance placed by the market) on the support or resistance price past which the breakout occurs.
The “gravitas” of a support or resistance price is a function of the following factors:
- Age: How long has the price been there to be used by traders as a reference price for placing stop loss and entry stop orders? The longer the price has been an obvious point of resistance or support, the more exit and entry stops will be near the price, and so the more price-moving market orders will be issued automatically by computers (or fast-acting, disciplined human traders) to propel the price in the direction of the breakout. Don’t judge the “age” simply as clock or calendar time (although that’s valid), consider also the highest and lowest prices within significant time periods: The extreme high and low price of the past year is more significant than the high and low of the past quarter, which in turn is more significant than the high and low of the past month, which in turn is more significant then the high and low of the past week, which in turn is more significant than the high and low of the past day, and so on down the hierarchy of the time periods traders commonly use to analyze markets, make trading decisions and place exit and entry stops..
- Size and strength of the trend before and after the point of trend reversal that establishes the support or resistance price: When a trend is reversed by a high momentum move, or by one that travels relatively far, that fact indicates that there was very significant breadth and depth of participation (many traders, many shares/contracts traded) that propelled price in a new direction. When a trend is preceded by a high momentum move, or by one that travels relatively far, that fact indicates that the buying or selling pressure that propelled the move finally exhausted itself—and so the move should be over and done with for a significant amount of time. Either one by itself, but especially both together, inspire very high confidence in that support or resistance price as one that should, with high probability, repel any future attempt to break past it. That means that many traders, and traders with very substantial assets with which to trade, will place entry and exit stops at or near such support or resistance prices. There will be great surprise and panic if such support and resistance prices fail to hold.
Be judicious in choosing which support and resistance prices to use when placing breakout trades. Avoid the weaker ones. Use profit targets whose distance from the point of entry is commensurate with the scale of the panic that should reasonably be expected to ensue if the support or resistance fails.
Professional traders generally “fade” the first attempt to break past resistance or support, but usually follow the move on subsequent attempts. Each attempt to break past resistance or support uses up the finite amount of support or resistance at the price, so each attempt is more likely to succeed. The supply of support and resistance is comprised of the professional traders who are fading the move, those who have profitable positions they want to exit at a profit, and those who have losing positions they want to exit as near to break-even as they can.
Breakout trades should work immediately, result in strong moves in the direction of the trade, and not exhibit any significant retracement initially. If price moves back to the other side of the breakout price for any significant period of time, or by any significant extent, exit the trade with as small a loss as possible. Breakout trades sometimes don’t work, and when they don’t, they signal their failure quickly and unambiguously, Listen to the market. It’s always right.
Believe what you see, and take appropriate action.
From Safehaven.com (Andre Gratian):
Could we make new lows? We discussed earlier that the SPX has a valid count to about 1240 and, based on its P&F chart I could make a case for the QQQQ to reach 52.50-53.00 before ending its decline. If these objectives are reached, it would put both indices outside their intermediate channel.
Still, I have reasons to think that this decline will be of limited extent and duration. This is, in part, based on the current position of the SentimenTrader which is shown below on the left (courtesy of same), and which is very bullish.
There are several legal wavecounts. The chart shows one of the relatively bearish ones, a five-wave downward impulse that bottomed on Wednesday, 16 Mar 2011:
It’s also possible–and notably easier!–to count the same waves as either a double or triple zig zag. There are strong, non-Elliott Wave reasons to prefer such corrective wavecounts. Those corrective counts (not shown) will be proven correct if the ES makes a new high above 1338 before it falls below 1197.75.
On the other hand, if the 5-wave downward impulse wavecount (as shown) is correct, then what should follow now is a corrective (non-impulsive upward or sideways move that stays below 1338 (preferably well below,) followed by another motive wave down below 1241.25. Such a downward motive wave, it if happens, could either be yet another 5-subwave downward impulse, or an ending diagonal wave. That dowwave may, or may not, break below 1197.75. If it does fall below 1197.75, the implications could be severely bearish. If it can hold above that price, the implications could be rather bullish.
At this point, there is no Elliott Wave way to disambiguate. There are other reasons that make me prefer the bullish interpretation at this time. But I have identified the evidence that would prove me either right or wrong, and will let the market be the final judge. It’s always right.
Update: Here’s the analogous wavecount of the S&P 500 Index itself [60-minute bars]
Physicist Guannan Zhao, Ph.D. student at the University of Miami, and his collaborators have developed a mathematical model to describe the timing of price changes of currencies and the overall dynamics of the Foreign Exchange (FX) market. Zhao presented his findings in a keynote address on Saturday, March 12, at the second International Conference on Financial Theory and Engineering (ICFTE 2011), in Shanghai, China.
Last Friday, the SPX pulled back one more time to the level of its former low of 1195 and, by rallying and closing 10 points above, has essentially made a double-bottom.
It is possible that the index has seen the low of its correction, but in order to confirm this, it will have to start rallying past all the down-trend lines that are in its way.