MY STYLE - BLOG
Next generation charting
Most trading approaches use the immediate price movement patterns for trading decisions. While thats a good approach, such approaches can be difficult to manage if one does not look at multiple timeframes (for time based) and coarser views for non time based charts.
One of our powerful trading methods which has been making trading more efficient in the recent days is an application of Kagi chart principles along with core Renko/Point and figure methods, resulting in a completely new chart that is different from both PnF, Renko and Kagi, though it looks like the latter. Combined with a trend recognition method thats amazingly simple but effective, results in extreme noise cancellation and a significant reduction in ineffective trades. This results in charts that effectively get closer to mechanical trading. An example is shown in the chart along side, where we have plotted the trend for NIFTY futures for intraday and positional trades. And the best part is that the core data for these charts is 1 minute time frame data demonstrating again that intraday and positional trades can be initiated from the same charts. This effort is part of the ongoing effort to simpler and more effective trading at TradeWithMe. Notice that each vertical bar in the chart may hold a few minutes, few hours or few days of data seamlessly. The bar patterns themselves have a story to tell and can be used in conjunction with the trend line. There are no ranging zones on the charts as these have been effectively filtered out by the charting technology. There is also zero lag in pattern formation as the data source is 1 minute data and no price derivatives used. One may argue that you dont need 1 minute data, however, when event based price movements occur, these are invaluable. I dont yet have access to tick data to experiment, so cannot comment on how the charts woud perform with that data source. The addon sophistication of swing logic, fib retracements etc. are other tools that one can use here without worrying about chart responsiveness. (click images for a larger view)
Normally, no, as the visual area of the chart does not allow charts in smaller time frames to show the bigger picture and these are cluttered with noise associated with ranging periods as well. At best traders look at higher and lower time frame charts. side by side. The picture changes, when one experiments with charts that are not dependent on time but just on price movements. Consider the convenience when intraday and positional trades are actually part of an overlapping trend. Exits and entries can then become precise and the two trading styles can overlap seamlessly.
Check this unusual proprietary trading chart that uses our version of the Kagi method that is noise cancelling like Point and Figure and Renko, but visually shows opportunities for both intraday and positional trades, as the chart is capable of showing several days data in the same viewing range. Addition of moving averages provides resistance and support information, that obviates the need for any other indicator for trading. Traditional Kagi charts use closing prices and %age reversals or absolute reversal values. In this chart, the reversals are based on point and figure principles and so is the data. Our charts show all the information that a chart should - namely date and time, ability to enter buy and short signals, even on Kagi style charts and consequently the profit/loss.
In the two charts shown, the slower moving average is a turning point for positional trades, while faster average suggests (but not neccessarily completely) additional trades on an intraday basis.
This chart is made available as one of the trading options for traders who are part of our mentored trading programs.
I had promised to show the impact of using a higher time frame chart during ranging periods. But instead of juggling between charts on different views, here is a method of showing the higher time frame chart on the time line of a lower time frame chart. The example is of 5 minute BankNifty futures shown along with a 1 minute chart of the same chart. Look at what happens on the higher time frame chart during ranging periods. Mini trends can be observed in the higher time frame that are not visible in the lower time frame. Now think of combining such charts so that you switch on and off higher time frame at will. Imagine the possibilities. Thats innovation at work at TradeWithMe. (click on any image to see a larger view).
Now, if you thought that was neat, here is the combined chart...
You can clearly now see a small ripple with two troughs, which then moves on to be part of the larger uptrend. This innovation may be small, but look at charts when in a trend. Its possible to stay stickily within the trend. See below the chart for October 9 2013
Technical analysts have used various methods to handle ranging periods. However, these dont work most of the time as a ranging period still appears as a range.
Here are some different approaches to handling ranging. Range bars for example are a part solution where noisy movement within a given amplitude is suppressed. Here are some other options :
1 - Time based compression : In this approach, one would see ranging segments within an existing time frame by changing the time frame of that period from a current level to a higher period dynamically. What this means is that one would see a 5 minute chart having a user selectable period when that part of the chart is displayed with a higher time frame. What will happen is that the ranging period will drop or disappear in the overall context of the chart. I will show the conceptual implementation of the same in a few days.
2 - Amplitude based compression - beyond range bars - where variable amplitude thresholds are implemented based on the trend. So that trends stay sticky. What that means that if a chart shows a trend, the system will automatically start increasing the amplitude of the reversal threshold so that no mid trend whipsaws affect the trades. This is perhaps one of the best kept secrets of smart technical analysts, as it is not easy to implement such a scheme.
3 - Noise less charts - Point and figure, Renko and Kagi charts eliminate noise. Change the key threshold parameters in these charts to get the same impact of amplitude based compression. This is current part of TradeWithMe's cutting edge research, where initial experiments demonstrate very good potential. See samples of what is achieved. Same data - two different logic:
Much has been written by many writers and traders about ranging periods.
As a technical trader, I prefer to allow the trading system to manage such days so that trading remains non discretionary and enable automation. What are the ways to handle ranging ? These are some ideas:
1 - Stay away from the market - easiest and most prone to opportunity profit losses.
2 - Put a limit on stop loss loss limit for the day and exit trading for the day.
3 - Enable a well thought out strategy to play itself out. One example of that is to take trades initiated following a cross of a threshold. For example, if RSI or Stochastics does not cross 30/70 or 20/80, do not initiate a new trade.
Option 3 is the most promising and can be used by both discretionary and non discretionary traders for consistent trading. The choice of threshold logic is wide open, and can include what the trader prefers and uses normally.
A fundamental input, that may be used is to stay away from periods that include ranging before a market news event. Thats optional.
Fast Reversal Trading
A good idea to manage volatile markets is to trade more frequently than even intraday strategies. This is something which I call "Fast Reversal" where trend direction is estimated at every price move and a trade taken. It can result in amazing accuracy as our own tests on BankNifty, which is very volatile, have shown. Accuracy levels are usually in excess of 80% for a well configured approach. As usual, no indicators - price is king!
Several posts ago, I indicated that you can make money by coin tossing, which has a 50% success ratio by statistical probability. How does that happen? Smart traders always try to keep stop losses manageable and smaller than their profit targets. Effectively, your average win profit divided by average loss should be a factor of 1.5 or more, so that you make money. What this is means that for every losing trade your winning trades will make at least 1.5 times what you lost. On a win ratio of 0.5 or 50%, this will effectively translate to a win ratio of 1.25 in money terms across all trades.
(see the definitions below for the ratios)
So the question is, what is good ? Just a great absolute profit ratio in real money or also the trade count win ratio.
Here is where risk comes in again. Any trading system with a low win ratio will have a higher risk exposure, as the laws of statistics will work in the long run, but in the short run may expose you to significant draw down in a low trade count success ratio system versus one that has a high success ratio.
Traders are sometimes impatient people. And that is their failing. By trying to search for trades in every market move, one is going to add up to loss overheads without any significant gain. And yours truly also admits having fallen for that sometimes.
But here is what we have done now. We had a trade win ratio of 60% and a profit win ratio of better than 90%, but we were concerned with the draw downs on sharp market moves. So here are two things that we did, which others can also benefit from :
- Tweaked the systems that we use to push up trade win ratio to better than 80%. Fewer trades, but successful.
- Move to fixed reverse stop loss (if the trade does not make money, exit it at a predefined level).
Important observation was that none of this reduced the overall net profit capabili
Definitions : Trade win ratio or trade count win ratio = Winning trades / sum of winning and losing trades or all trades
Profit win ratio = Profit from winning trades/ sum of profit and loss from all trades.
During the last three months we have had our downs where range bound periods tended to make mince meat of our trades. Conventional wisdom suggests that one stays out of ranging periods. But there lays a story. Ranging periods like trends tell us that there is indecision. Most professional traders give up and ignore ranging periods. We dont. As ranging periods tell us about changing market sentiments. If we can capture that sentiment, trades can be more effective.
And we have done exactly that! We have refined our methods and done immense original automation around our core trading methods, that trading time is focused on observing market sentiment alone. We are working on a sentiment index and will share information around that soon. And all of that is simply around price action.
We have published indicator less trading concepts, which are again cutting edge approaches which approximate the extremely effective "tukka" or chance/gut feel traders. Again price is king in this approach.
We have dissected the point and figure trading approach to superimpose that on normal time charts and then reverse engineer the original concept to add features that combine the best of both approaches, without making a mess of them. Check the Point and Figure knowledgebase in the technical analysis section.
How are we performing now?
This is a question I keep asking myself every week. And the answer relies on just point. Are we trading consistently? Which means is our trading method giving consistent profits - sum of winning trades over losing trades remains consistently positive. And the answer is a resounding YES.
If I asked this question a month ago, I had the niggling doubt, that we were getting exposed to whipsaws now and then and losing good profits in a few bad trades. Now that has changed. How?
First our trading systems use small stop losses for all classes of trades. For intraday trades for Nifty the stop losses are as small as 4-10 points and for Bank nifty 12-25 points. This implies that we can survive more stop loss trades (not that we need to have them more often!) in our risk management approach. The second and more significant improvement is being able to make right interpretations of the bigger picture during market hours so trade trends are correctly identified and trades are entered at the right levels.
That is where our proprietary research into identifying trends and supports and resistances has come to play. A visual display of this information is displayed by a conventional bar chart and is superimposed onto point and figure precision. Infact, the two systems now help double check against one another like a closed loop. The impact was felt today -- expiry volatility in intraday trades was completely managed! See the example of BankNifty trades below. The light green and reddish brown lines are the resistances and supports, which another proprietary program generates, trades are generated using a P&F system and the trade monitoring robot generates the stop losses based on these levels. The result is amazing precision! And thanks to pretty original research that generates these charts.
The original swing theory of higher highs and lows for long trades and lower lows for short trades remains king. The issue where traders stumble is to filter the noise out and identify the highs/lows. Thats where technical traders come in with their arsenal. If you can master this, and I am not yet fully there, but slowly moving in that direction, trading takes the level of an art! Because you try to project trader psychology onto the charts, and that can be subjective and disastrous at times. The upside is that you trade only with price action, the purest form of market direction. Think about this!
Abnash Singh, Am a Trader helping small traders to realize their dreams.