The setups I include on this blog are used in conjunction with the 3/10macd and the criteria I ascribe to it as a way to alert me to an existing condition of price. The key concept to take away from this blog is that I try to anticipate what will happen on the higher time frame by using a faster time frame to trigger the trade setup. I do not trade a "system" I use two indicators to clue me in to price conditions. Please read the Disclaimer located in the sidebar of this site. I can be contacted via email at toddstrade@gmail.com
I am always open to questions, comments, or suggestions on how to improve this blog.


Showing posts with label indicators. Show all posts
Showing posts with label indicators. Show all posts

Wednesday, March 31, 2010

Andrew's Pitchfork

I've been playing (and having loads of fun) with the Andrew's "Pitchfork" (or Andrews Median Lines as they are also known as) lately. There are plenty of links (well links within the links) to visit at the bottom of this post. For now, start with the basics,
Drawing the Andrews Pitchfork: see this link- Choose two alternating swing high/low pivots, label "B" and "C"- Construct the Median Line (ML) to connect "BC" with a third pivot. "This pivot usually immediately [precedes] price swing BC but it can be any pivot [preceding] BC."
So, essentially, you have two pivots that are bisected with a prior pivot, while extensions off of your "B" & "C" pivots become your Upper & Lower Median Lines (UML/LML) like so:There's also the Schiff Variation:
-Both share the same "BC" line, however the Schiff Meridian Line originates from the midpoint between A & B, like so:
This Schiff variation is good for using with gaps.

There's also the concept of including Upper and Lower Trigger Lines:
- Simply extending lines off of the AB point and AC point, like so:See this link for Trigger Line setups.

There's also the "Sliding Parallel Method" which I'm yet unable to comprehend.

Now some fun with charts: SPY Daily

The Schiff version - the handle (center line) originates at the mid-point between the A and B points:The Modified Schiff version - the handle (center line) originates 50% between the A-B points and directly above the "A" point.
Tradestation also gives you the ability to add parallel bands to the pitchfork (use for the Sliding Parallel Method?) to give midpoints between the BC lines or percentage extensions off of the pivots.Here are 3 forks in the recent SPY price action. I included a stochastic for the sake of strategy development which one might base trades off of, like those indicated by vertical dash lines.
Some more key concepts include:
- Median Lines should be tested before being traded off of.

"How do I know which pivots or swings to choose? The answer sounds simple-minded, yet it is the single best answer I can give: Draw many lines and use the lines that prices respect when you trade. Draw as many lines as your eye sees as pivots because, often, the first or second or third combination will not 'describe' any of the action of price when you first draw the MLs...I try different pivots, starting with major pivots in the weekly or monthly mode...and then draw MLs off of lesser and lesser pivots. I do erase MLs that are not of major pivots and do not 'describe' price."

"
The first simple rule is that downward sloping MLs indicate prices in a downtrend--downtrends are made to be sold and the Lower ML of a downward sloping ML describes prices that will trend lower. Upward sloping MLs describe uptrends."

"Most traders think that MLs are a stand-alone tool, but I was taught to always use them along with many other signs of support and resistance and it is the integration and interaction of these many tools that can make the techniques valuable.""...it would intuitively seem that major swings or pivots make the 'best' points to use for all Median lines. But in practice, the best points to use are the points or pivots that describe or contain the price action. I know there are 'purist' chartists and if that's your preference, then by all means, draw the lines that suit you. But the goal of the Median line is to contain or describe action as it unfolds--and so by all means, be creative. Often, it's the obvious pivots that work the best, but at times, you have to get your hands dirty. I'll give a few creativity hints: If all else fails or if you are looking to use these techniques on 'short term' trading, try using the highs of the three pivot points, or the low. Or use the extreme of a gap. Or use an extremely wide day as the high and low of an A-B-C. Try it. You can always erase it if it doesn't describe or contain action."
"
Don't blindly trust an untested line, unless you have other reasons to expect that support or resistance to hold. But once it has been tested and even better, when you can add the 'confluence' of other indicators like trend lines, fib projections and retracements, chart formations--then you can have much more confidence in the Upper and Lower Median and Center Median lines.""Actually, though it is very seldom taught, you can draw and use a Median line using the open high and close of a bar. Period. Does it work? It can. That depends on what you are trying to use it for. It can be great for setting stops, for example."


Click on all the links on this page, a wealth of information; http://www.trading-naked.com/AndrewsPitchfork.htm

Wednesday, April 22, 2009

Klinger Oscillator

I'm carrying this post over from the blog I kept under a different URL, for the sake of posterity.
I've been using the Klinger Oscillator lately and find it to be very helpful. This oscillator is a construct of volume and is used to determine whether it (volume) is confirming price changes. I just haven't been satisfied trying to read volume bars, OBV, or up/down volume so I trolled through the Tradestation forums and dug this out.
The variables involved in constructing this indicator are Klinger's volume Oscillator with ATR (Average True Range).
The Klinger Oscillator (KO) is intended to show when price changes are confirmed by volume. The computation comes from three types of data:
- the high-low price range (movement)
- volume (force)
- Accumulation/Distribution; where Accumulation is when the sum of the bar's high+low+close is greater than the previous bar. While Distribution is when that sum is less than the previous bar. When the sums are equal the existing trend is maintained.
Volume force is converted into an oscillator that represents the difference between a 34- & 55-period exponential moving average and uses a 13-period trigger.
From what I have read, the KO works well when going with the trend, but not as effective going against it. While it is most useful with price action divergences; especially on new highs/lows in overbought/oversold territory.
So, on the short time-frame (5-min. chart) I'm using an 100-EMA to determine our trend. If price is greater than the 100-EMA the trend is up, vice versa when price is below. If the trend is up you look for the KO to dip deep under the zero line and a cross of the trigger would be a buy signal. If the trend is down, look for the KO to rise up above the zero-line and a crossing of the trigger is a short signal.
This particular oscillator also incorporates an Average True Range length to help in smoothing.
Here are two examples I pulled from price activity in AAPL & FSLR.and more recently from SPY on wednesday April 22

Monday, April 6, 2009

low volume swings

Not much to blog about today; a low volume day in SPY today. The 5-min chart gave us two "First Cross" entries; one short in the morning (red vertical line) and one long later in the afternoon (green vertical line, that would have tested your patience unless you put the order and stop in and walked away, which is what I should have done). There was also a "Slingshot" setup where price made a higher low, momentum made a lower low in momentum and price accelerated after that. In the chart below this setup occurs after the swing low that follows the green vertical line.On a longer time frame (15-min), the most recent 20- & 50-EMA crossovers have been quickly reversed.Looking at the daily chart; where's the momentum? Volume has dropped off, the 3/10 oscillator is drooping, ADX is showing low volatility (a precursor to expansion).

Thursday, April 2, 2009

end, or beginning, of a trend?

We've certainly been in rally mode for nearly a month now. What should transpire from here is figuring out whether or not this is the beginning of future strength or merely a counter-trend move that will soon exhaust itself. Looking at the Daily SPY chart I wanted to highlight what has happened over the course of the last year.
Looking at the ADX for this market we can clearly see what has followed periods of low volatility (indicated by an ADX less than 20). Currently the SPY ADX is below 20, and that's not to say there can't be more upside, but it's curious because a low ADX reading is something that should give us warning of an impending volatile move or trend (in either direction), as volatility contraction precedes expansion.Taking it even further, here we are from Jan. '07 through March '08:
Looking at an hourly chart of the SPY, we gapped up and found resistance at a previous congestion area around the $84.50 area.As this rally threatened a breach of our regression channel, price has recovered, though found resistance at the midline todaySure we have one more day left in this week, but here's where we stand as of today on the S&P500 weekly chart, very much a critical point.

Friday, October 3, 2008

3/10 MacD

An indicator I use on most of my charts is the 3/10 MACD.  The 3/10 line is the difference between a 3 and 10-period Simple Moving Average (Fast Line) while the 16-line (Slow Line) is a 16-period simple moving average of that 3/10 differential.  Below is a chart with price removed with a 3-period simple moving average and a 10-period simple moving average.  Simply a method to help visualize the concept.

This oscillator was made "popular" by Linda Bradford Raschke (LBR), and you can find plenty of information regarding her techniques throughout the internet.  You can also find plenty of information dealing with the 3/10macd, as well as tips in how I use it, within this blog and particularly in the documents listed in the sidebar on the right hand side (just request to view and I will gladly accept).
And, here you can read a Q&A pertaining to the 3/10 macd.

You can use this on any time frame and it helps in highlighting the momentum behind moves; especially tops and bottoms where it can tip off divergences. Be warned however, like most oscillators, it is not very helpful in extremely high or low momentum environments. So, where there are sideways choppy markets and large momentum thrusts (including gaps), you may want to employ a different approach without this oscillator.
To me, this is an invaluable tool for short-term trading as it is very responsive to price, helping to smooth some of the noise which may be distracting.

Here are a few examples of some simple divergences:

Also handy, this oscillator can help alert you to a new higher high or lower low.  Known as the "First Cross" you will notice that when the slow line crosses above or below the zero line for the first time and the fast line corrects into it, we have a First Cross buy or sell trigger.


Another example:

stochastics and cycles

I've moved this post because I want to keep it together with my 3/10MACD post, as these are posts that deal with my trading set-ups.
I've been reading about trading with cycles lately. I've been wanting to read this book, but haven't ordered it, as I've got quite a queue of books stacked up at the moment. If anyone has it, I'd like to hear your reviews. I've also been reading through the Tradestation forums pertaining to cycles, which are a wealth of information. What I've been focusing on lately is the use of multiple double smoothed stochastics, and using them to highlight the alignment of cycles from different time frames to increase the chance of a high probability trade. There are also Long-range cycles.
To give a brief synopsis:
The stochastic oscillator measures the most recent price in relation to the total price range for a selected time period, on a percentage basis of 0 to 100%. A fast stochastic uses a 3-period moving average of the %k-line (represented as a second line referred to as %D). A slow stochastic, or "smoothed" stochastic, can then be smoothed further (double smoothed) by taking yet another 3-period moving average of the previous slow %D-line average.
Oscillator lengths can be tied to underlying market cycles. While precise time cycles can be difficult to pinpoint, a simplified approach would be to use 5, 10, 20, 40 -lengths/cycles to represent the number of trading days in One Week, 1/2-month, 1-month, and 2-month -period respectively. Most default indicators, like the stochastic, use a 14-period length. This length is based on more of a calendar cycle, so a 14-length represents approximately 2-weeks, or a half-cycle (half of a month).
Different markets "cycle" differently; commodities, from softs to metals, have different market cycles, just as security indexes and sectors do, so it's up to you to find a good fit. But to keep it simple and proportionate start with the 10, 20, 40 for the Daily multiples.
While trading divergences of an oscillator can work from time to time, they can't be relied upon solely. An important consideration must be given to the underlying trend while trading Overbought/Oversold (OB/OS) crossover triggers as well. Most buy signals (crossing above 20) can work best in up-trends, while most sell signals (crossing below 80) can be more profitable in down-trends. This is where other indicators and moving averages, etc., come in handy as filters.
So, with that in mind, one should consider trend WHILE ALSO giving less attention to the stochastic oscillator in the early stages of a move, but paying close attention to its signal as the move matures (or nears an end).
Anyway, lots of words. It's time for the charts. The basic strategy here is to use three different lengths on one chart, wait for an alignment, or agreement in direction, of the 3 "cycles" and act off of that alignment. I'll keep the lengths the same (10, 20, 40) and use AAPL and the S & P 500 index.
Can be used daily:
Weekly:Or, you can just use your Slow Stochastics:Or, try Intra-day, using a 5-min chart with lengths of 12, 36, 72 to represent 1-hour, 3-hours, and 6-hours: