Location

In an earlier post, I outlined how you could trade both directions as long as you exit on the close of the entry bar. So the question becomes, why would I ever want to only trade with-trend? Why do I insist on taking only a few good trades and let most trades go by?

What is the point in taking only a small number of trades? Isn't it better to also add those little trades to pad my profit? Won't all those small trades add up?

Today's action illustrates why all those little trades do little or nothing to your bottom line for most traders. For example, today I added five counter-trend trades at reasonable setups:

#2: Multi-leg fBO with strong signal bar
#7: TCL OS
#10: Strong signal bar
#11: small trend entry bar (occasionally these are followed by BO bar)
#12 possible W

Even with the express intention of exiting on the entry bar, three of the five trades were losers.

In contrast, only one with-trend trade, #3 was a loser. This was a news related experimental trade. Trade #9 was a scratch and was an accident.

The important take away here is that if I only took #1 and #5, my cumulative profit would have been larger.

A second observation is that the longer you trade on a given day, the worse your performance gets. For example, My peak profit was after trade #6. I gained nothing by trading for an additional hour.

So what's special about #1 and #5? They were clearly at the trendline. Their superior location is obvious even without drawing an explicit trendline.

Building blocks

Price action trading has so many angles and components that it can be overwhelming. An average trader is unlikely to be able to recall every single aspect of price action and make the correct call most of the time. In contrast to simpler indicator based approach where everything boils down to a buy/sell, price action concepts reflect the complexity of the market and the trader needs to weigh opposing readings and make the right call.

To be able to absorb and build up one's knowledge and expertise and then practice and apply it in daily trading is no simple matter. To do it with every single bit of price action knowledge is next to impossible.

I believe that the layers of knowledge and expertise need to be built up one by one. The first and foremost is directional correctness. The next is the right entry and the last is a target estimate.

Directional Correctness: A new trader should trade not with profit in mind but with the desire to hone his directional correctness. If a trader is able to take trades that eventually move in the direction of your trade, then they could theoretically be profitable with larger stops. A simple way to do this is to not bet on any reversal attempts. Instead let the reversal attempt actually succeed (take out a swing point) and then take the next pullback. With-trend entries enable directional correctness. The market needs to be trending  (higher highs/lows or lower lows/highs) and not in a chop. A simple way to ensure that you are entering with-trend is to always draw a trendline and ensure there has never been a trendline break (a sustained move beyond the trendline) or the trend has otherwise terminated (no HH/HL or LL/LH).

The right entry: A second requirement is to only enter at support. A test of the trendline or ema or HLC of the prior day are usually good choices for entry. A good location ensures that many more traders enter with you increasing the probability of success.  Two or three legged pullbacks have higher probability than a one legged pullback. The earlier the pullback, the weaker it can be, i.e., the first pullback can be one-legged with a poor bar if its deep. The second and subsequent pullbacks should be two legs and deep and should preferably have a good signal bar. Do not be tempted by one-legged shallow pullbacks after the first pullback. They are likely to fail and give a two-legged deep pullback. If you find that you are often stopped out and the price eventually moves in your direction, you need to develop patience and let the market develop better entries. Once you have mastered this for normal trends, you can make adjustment for hard and soft-trends. You would still need to be cautious after an extended move or third pushes since a retrace would need to be two or three legs.

A good way to distill the points above is to reduce them to location, pattern and bar. A good location is a pullback to a support in a trend move. A good pattern is a deep pullback, preferably two or three legged. A good bar is a strong close, non-overlapped signal bar (with adjustments for 2BR, oio, etc., where you would treat the entire pattern as one bar).

Estimating targets: Once you have mastered directional correctness and are generally successful taking the right entries, you can work on maximizing your profits. The most obvious choices are measured moves of a trading range and overshoot of TCL on third push. You should also look to exit near the recent extreme once a trend is broken.

Remember the order of mastery and take them up one by one. Until you have mastered directional correctness, your mind does not have the bandwidth to absorb the intricacies of the right pattern and entry. Until you can confidently take the right entry without fear of being stopped out, you will have no conviction to hold until your estimated target.

Once you have all the three, you have developed a framework into which other observations of the market fall naturally into place. After this point, your trading style develops sophistication and absorbs information from the market and other participants naturally.

The first few weeks

Beginner's discipline

One of the enduring problems of a trader is the laxity in discipline that creeps in with success. My initial success rate of over 65% has dropped to below 50%. Some of this can be attributed to necessary experimentation, but not all of it. At least half should be attributed to overconfidence and a misplaced self-assurance that I can overcome any losses easily. In general, I try to design my experiments such that losses are minimal. However, for many goals such as figuring out where to fade an overlap, there is no practical way to do it.

Many of the big down days were wednesdays, trying to figure out how to extract maximum value from news events. I have given this up as an unreasonable and expensive goal.

My current goal is to redesign experiments such that even with losing trades, I wont have any losing days.

Plateau

Its possible that I have hit a plateau. The big losing week is especially bothersome and has reminded me to be more disciplined. I need to take care not to let my uptrend in the equity curve turn into a down-trend.

I will post with increasing frequency in the coming days now that I have a baseline to compare my performance.

Alternative RSS Reader

As many of you are aware, google is shutting down its blog reader. An alternative reader similar to Google Reader is Old Reader. You can use it to read this (or any other) blog. You can also export your google reader settings and import it in old reader.

Late 12-2010 and 01-2011 posts restored

My trading style has evolved over the time I have maintained this blog and have leaned towards new setups and new ways of seeing the market. However, the early articles are still relevant and extremely educational. These articles had lost their images due to deletion on the server.

Thanks to the contribution from readers worldwide, most of the early articles have been restored from images submitted by various readers.

I have regenerated images for posts dated 2011-01-10 to 2011-01-14. Those charts do not have their original markings. If anyone has these charts, please submit them for restoration.

Thanks to everyone for their continuing support. I hope to be able to continue posting with regularity and end the dearth of posts over the last several months.

Update: All images have been restored. Thanks to my readers.

Targets and turning points

A large first bar is usually an opening range, especially if it has large tails on both ends. Any breakout beyond it is likely to fail (b2, b5) and test the other end of the trading range. On tight trading days, price could bounce between two ends for the entire trading session but occasionally, a breakout will succeed and such breakouts usually lead to a measured move of the opening range (b14).

Trading the fBO of the opening range should be attempted with caution since the profit potential, defined by the range size needs to be large enough to justify the risk. A weak bar such as b2 or b7 are not very high probability entries and your chances of a successful trade are lower compared to a strong signal bar for an fBO setup.

A successful BO that leads to a measured move may turn and retrace most of the breakout (b24) or even most or all of the opening range (b49). Sometimes breakouts continue to trend up for the rest of the session, but a strong prior move (b3-5) suggest a turn at the measured move is likely.

Therefore targets such as the other end of the range such as b1L for b2 short, b2H for b7 long and MM target of b1 at b14H are potential turning points and you should watch carefully for signal bars. A strong signal(b15) bar may mean good chances of a reversal or retrace.

Structure before pattern

A tick or bar scalper only needs to find setups that move at least as many ticks as his stop. A swing trader's job is harder, since he has to correctly identify setups with a high probability of a large move. But the advantage of the swing trader is that he comes out ahead even with a 50% win rate.

One of the classic big move scenarios is a failure. A failure of a W trade (#1) sets up a measured move of the size of the W in the opposing direction. If the W setup was weak to begin with (b19,20), this strengthens the failure trade (#2).

Failures usually have a second leg and after the move down to b26, you should never take a reversal until a new low is reached (b52) or 3 pushes to a HL.

A second leg trade (#4) may only take out the prior leg or it could give a large breakout. A channel type move such as today is more likely to reverse while a breakout is more likely to result in a sustained trend.