Archive for Tick16
My monthly update on market internals.
The current snapshot of S&P 500 3-Day Advance Issues, Tick16 Short Term + Long Term as of 2013 May 17 close.
1. Long Term Tick16 (yellow line) moving back up to right below neutral zone.
2. Short Term Tick16 (red line) moving back up to neutral zone.
3. 3-Day Advance Issues (green line) bounced off zero line once and divergence from new high prints in S&P
a. #1 is bearish pointing to potential important top in the making
b. #2 points to breakout mode where next move can be very explosive in either direction
c. #3 Very strong bearish divergence points to a quick selloff or consolidation in the making
Last update predicted a potential short term bottom in the making and that was one powerful rally from that bottom.
Current rally has very strong signature of blow-off top thus it is important not to chase this rally.
This pattern, however, has happened once before back in July-Aug 2012. The 3-Day Advance Issues will correct itself as that is the nature of the statistics. If by the time it has moved back down to below neutral zone while S&P going sideway or pullback just a little only as in late Aug to early Sep 2012 , we will get a powerful short squeeze as a consequence.
This article explains how to translate the interpretation of NYSE Tick Index between the eSignal version and the IQFeed version.
Frame of Reference
I use the eSignal version of NYSE Tick Index extensively for backtesting and signal generation. It is not that it is better than the other ones. It is a habit now as I have monitored the same index for 20 years. I use the eSginal version of NYSE Tick Index to write this series of articles too.
The advantage of the eSignal version is that I have access to all the historical data of the index going all the way back to the 1990s. It enables me to understand the changes happening to the index and how it evolved over time.
Currently eSignal NYSE Tick Index (symbol $TICK) is updated every second, just like the one offered by DTN IQFeed (symbol JTNT.Z). Both of them are updated during US session stock market hours from 9:30 am to 4:00 pm Eastern Time. From these second by second real-time update we can construct minute bar, etc. easily to line up with the other data for comparison.
An Example Chart With Both Versions of NYSE Tick Index
While preparing for the next article on converting trading rules between eSignal and DTN IQFeed versions of NYSE Tick Index, I have been doing data comparison on and off among multiple sources whenever I have time to sit in front of my computer.
So far it is clear that the eSignal NYSE Tick Index is the same as FutureSource, S&P Comstock, and effectively any company that is part of the Interactive Data group. I did the actual checking across these 3 data feeds to make sure they are the same. So I am sure about that. And I am speculating that it would be true for the other brands of data delivering NYSE Tick Index under the same group of companies.
I have access to CQG data and the NYSE Tick Index they offer is extremely similar to that coming from eSignal. Since I do not know if CQG gets its NYSE Tick Index from S&P Comstock or if it generates its own, I cannot claim that they are the same as the data does not match 100%. Let me put it this way. The data looks like coming off the same statistics gathering algorithm but the time where the sampling were done was set at slightly different time schedule.
I no longer have directly access to TradeStation data myself since the beginning of the year. So I do not have updated data from them to cross check if their data is the same as eSignal or not. When I get to the point writing about conversion of analysis between eSignal and Tradestation data, I may need help to obtain the latest tick data on NYSE Tick Index from Tradestation to do so.
I will probably summarize the brokerage versions of NYSE Tick Index in a separate article as the comparison of these data can get very long and boring. I will definitely include MB Trading and Open E Cry. I will also try to include Interactive Brokers if possible. The idea is to turn these comparisons into reference articles. Should you choose to switch from one data source to another, you can quickly tell the difference without the panic attack I experienced.
Back to writing. Later!
My monthly update on market internals.
The current snapshot of S&P 500 3-Day Advance Issues, Tick16 Short Term + Long Term as of 2013 April 19 close.
1. Long Term Tick16 (yellow line) below neutral zone.
2. Short Term Tick16 (red line) moved into oversold zone.
3. 3 Days Advance Issues (green line) diverged from new low printed by S&P
a. #1 is bearish pointing to trend sell mode
b. #2 points to a bounce is due until the line is back to neutral zone
c. #3 strong new momentum low points to start of down trend, or as a minimum end of the current up trend. The divergence points to a potential short term bottom in the making.
Last update predicted a swing long was coming. I did not expect that turning into a major short squeeze / blow-off top off the double bottom in the 3-day advance issues.
The current selloff in progress now is different from the previous one as the strong new momentum low in 3-day advance issues usually points to significant down side risk. Thus counter-trend long should be taken with precaution that they may fail easily.
When I talk about NYSE Tick Index, there is a confusion where my data comes from. The confusion happens because some people find that the Tick Index they loaded onto their charts look completely different from mine. There is a good reason why that is the case and there is no need to panic that your data feed is not providing the same data to you. I will give you a detail account why this discrepancy exists.
The Original NYSE Tick Index
As far as I can recall, the original NYSE Tick Index was introduced by NYSE for the floor traders but it was since discontinued after a short period of time. Thus, there is really no real official NYSE Tick Index since. So what is this NYSE Tick Index I am talking about all along?
The S&P Comstock NYSE Tick Index
Well, it is a long story.
Since NYSE no longer computes its own Tick Index, the data vendor S&P Comstock took the challenge and recreated the index themselves. It made the index available to professional traders who subscribe to their real-time data service. At that time, real-time data was delivered through satellite broadcast, land line direct hook up, and FM transmission.
This effort made by S&P Comstock extended the life of the Tick Index. Since then, many traders developed trading techniques around the Tick Index. Eventually, Tick Index became famous in the professional trading community.
The DBC Tick Index
The major exchanges did not have the facility to allow many direct connections to their data centers in the old days (1980s). S&P Comstock is one of the few data services that secured the right to such direct connection. Hence, the retail data vendors, like Data Broadcasting Corporation (DBC), has to redistribute data from S&P Comstock. This enforced the same NYSE Tick Index distributed to the public at the time.
In short, there was only one NYSE Tick Index in the old days.
The eSignal Tick Index
Fast forward to current time, S&P Comstock, Data Broadcasting Corp., and a few other data vendor companies have merged into one big company. DBC’s data service called Signal in the old days has been rebranded as eSignal since the internet era. The NYSE Tick Index that these companies send out in real-time is still the same NYSE Tick Index.
The Unexpected Effect From More Direct Exchange Connections
A significant change has happened since when the exchanges opened up their data to more direct connections to data vendors. Many data vendors no longer need to redistribute data from S&P Comstock. They can all connect directly to the exchanges and bypass the middle-man. The advantage of that is faster (and timely) data delivery which is very important. But these data vendors and the direct access brokerages who stream their own data to their customers no longer have the NYSE Tick Index available to their clients.
Their solution, instead of licensing the one offered by S&P Comstock, is to create their own versions of the NYSE Tick Index. And that is where the trouble begins. The tick indices created by these firms do not resemble the original NYSE Tick Index. If they do, however, I suspect that it may become a very sensitive issue because it can be borderline infringement of the copyright to the index construction process.
These various favours of the NYSE Tick Index has caused confusions and frustrations among traders.
The TradeStation Tick Index
TradeStation used to redistribute data from one of the major data vendors thus its NYSE Tick Index was the same as the one offered by eSignal. But I have not looked into what TradeStation offers now, so I am not sure if they are still doing the same.
The DTN IQ and DTN IQFeed Tick Index
DTN’s NYSE Tick Index is definitely not the same as the eSignal one. They look similar but there are subtle differences between the two. Hence you cannot directly apply statistical studies done on the eSignal NYSE Tick Index to the DTN one and expect to get the same results. It will take a separate article to explain the difference and how to adapt your work from one to another.
My version of the history may not be completely correct since I am reporting what I know from a client’s point of view. There could be details that I missed. If you know something about this and would like to expand on the story please send me email or comment on the article.
As a subscriber to all these data services over the years, I can say that the worst time during the transition was the time when the direct access brokerages and smaller data vendors started to jump ship from being data redistributors to direct exchange connections. Various cash indices, including the NYSE Tick Index, were stitched together with data they got from the past, yet the real-time one they broadcast are generated by themselves.
Just think about the horror created when moving a trading model from using eSignal data into one using a direct access brokerage’s data, where the trading model no longer fire off signals as expected.
At this point, as the data vendors have all generated their own Tick Index for a long time, it is no longer a big deal switching data services or brokerages. As long as you are aware that the tick indices may not be the same among data sources, you can adjust your trading strategy to deal with the change.
My monthly update on market internals.
The current snapshot of S&P 500 3-Day Advance Issues, Tick16 Short Term + Long Term as of 2013 March 19 close.
1. Long Term Tick16 (yellow line) below neutral zone but not dropping
2. Short Term Tick16 (red line) moved above the neutral zone
3. 3 Days Advance Issues (green line) diverging from new high printed by S&P
a. #1 is bearish pointing to overbough and oversold on 30-min chart and higher timeframe can easily result in reversal.
b. #2 points to buying in underlying components near the all time high pretty much continuously.
c. #3 points to a pullback in advance issues back down to oversold zone may be necessary to correct the prolonged period of divergence we experienced over the past month. As long as price can hold a higher low when the advance issue reading is oversold again like end of February, we will get a tradable swing long coming.
Last update predicted we would experience a significant pullback and we got that in late February.
A selloff is now in progress and how it develops from here will determine if the long term price up trend is over or that another leg up is in store for us. Monitor the way how 3 days advance issues is moving will give us the clues.
I am going to present the first example trading model in this series that exploits the characteristics of the NYSE Tick Index. Most people focus on the intraday behaviour of the Tick Index and forget that it is a great summary statistics tell us what is more likely going to happen the very next trading day. The example trading model Blind Short is a short only day trade model demonstrating this fact.
Over the past 10 years, the model has generated more than $3,000 dollar profit annually on single lot, averaging less than 30 trades each year. Commission of $2.5 per trade per contract is taken into account. Special limited worst case analysis is used with 1 tick disadvantage so that we know the backtest result is a good baseline to measure from.
Following chart shows the emini S&P continuous contract at the top pane, the NYSE Tick Index in the middle pane and the net dollar gain per contract with the Blind Short model in the bottom pane.
The performance has been very consistent even with the raw model presented in the chart. It is interesting how such a simple trading model works so well all these years.
Basic Model Rules
The NYSE Tick Index (aka the $Tick index) and other similar tick indices all demonstrate a similar trending properly. I call it the micro trend channels of the tick indices. These micro trends are easy to recognize visually once you understand how they behave in general.
Be prepared – this article and the upcoming ones on using Tick Indices will be getting longer because the techniques in reading Tick Index is not that simple and requires more explanation comparing to other techniques.
Tick Index Micro Trend Channel
The swing trend definition is exactly the same as the one I wrote about in Art of Chart Reading even though the NYSE Tick Index (and other tick indices) is not based on price.
Swing top and bottom can be defined using 1 minute bar with a 5 minute window.
Once a swing trend has formed you can extend the lines to visualize the channel.
A channel will stay in effect until one side of the channel is breached.
An Example Chart on Trend Day
Following is a trend day in Emini S&P. The chart is in 1-min resolution with Emini S&P in the top pane and the NYSE Tick index in the bottom pane. You can click on the chart to view them in full size.
I have highlighted the micro trend channels on the NYSE Tick Index in the chart.
1. NYSE Tick Index likes to turn and form a swing trend before the price does. In another words, the intraday swing high and low in Emini S&P are often formed with a divergence signal with the NYSE Tick Index.
2. Good turning points usually start near the hour marks.
3. Sustained intraday moves in S&P are always accompanied by a micro trend channel in Tick Index in the same direction.
4. Best intraday swing trend changes happen within a 15 to 30 minute window with clear divergence between price and the NYSE Tick Index.
5. If a strong micro trend channel in Tick Index is formed early in the day during 9:30 to 10:00, it is likely to be a trend day.
An Example Chart on Non-Trend Day
Following chart is emini S&P and NYSE Tick Index in 1-minute resolution. The day showing is a non-trend day.
1. Big swings in NYSE Tick Index translate into only small moves in S&P
2. First micro trend channel failed to last for an hour
3. Folded trend moves in NYSE Tick Index resulting in the micro trend channels themselves diverge from the price swings (e.g. the first 2 down swings in the chart highlighted by the 2 red channels)
4. Likely to break a swing during the 30 to 45 minute mark.
5. A micro trend channel formed around 10 am and failed to sustain the channel for an hour, it is likely that the day is not a trend day.
(Members’ only content below)
My monthly update on market internals.
The current snapshot of S&P 500 3-Day Advance Issues, Tick16 Short Term + Long Term as of 2013 Feb 17 close.
1. Long Term Tick16 (yellow line) below neutral zone but not dropping
2. Short Term Tick16 (red line) pulled back to neutral zone
3. 3 Days Advance Issues (green line) diverging from new high printed by S&P itself
a. #1 in bearish consolidation mode. That means 30-min timeframe intraday has no continuation strength. Both overbought and oversold on 30-min and up would result in reversal easily.
b. #2 points to breakout mode as the index has compressed at neutral zone for days.
c. #3 points to a pullback in breadth (advance issues) back down to neutral zone is necessary. It can correct itself without dragging the price of S&P lower. If so it will be very bullish for S&P.
Last update predicted it was still in favour of the bulls and every push lowered were good long opportunities. That was indeed what happened. This may change if #1 starts to slide lower away from the neutral zone into more bearish level.
It is time to take a look at the basic statistical behaviour of the Tick Index.
Following is a chart of NYSE Tick Index and its 3 distribution graphs. Red one is the distribution of the daily low. Yellow is the distribution of the daily close. Green is the distribution of the daily high.
Visually it is pretty clear the 3 distributions are completely different animals. It is also the primarily reason why it is hard to interpret Tick Index visually if you are not aware of its property. You have to read the positive extremes with a different set of rules from the negative extremes.
Notice the positive extremes (i.e. Tick Index daily high prints) are concentrated within a tight range. In general Tick Index positive extremes behave very similar to price spike highs. Intraday, they spike and retrace quickly from these extremes. making them easily identified.
Local positive extremes are often associated with Emini S&P retracing at least 1 point and majority of the time 1.5 to 2 points.
If you have access to a faster version of the Tick Index, like a custom version you construct yourself locally on your computer, it is possible to get leading signals that is 10 to 15 seconds ahead of the turn in Emini S&P. Hence it is probably the best scalping tool you would ever get.
With the regular Tick Index, the leading effect diminished to a few seconds only, so you have to pay very close attention to the tick by tick update of the index to catch the turn.
Next we have to look at the negative extremes (i.e. Tick Index daily low prints). First, they are NOT concentrated within a tight range like the positive extremes. These negative extremes do not behave like price flush lows. Intraday, they can stay at the negative extremes 2 to 3 times longer than the duration it spends in positive extremes. Negative extremes are best used as down trend triggers or confirmations.
In another words, It is okay to scalp a spike low if the negative extreme printed is not dropping below -1000 (approx.). You would know that if a certain negative extreme is holding by comparing that to the ones made earlier in the day as Tick Index has strong memory effect.
When Tick Index started to fall deep below -1000, however, it may just stay there for a bit longer, and that is enough to break the price level in Emini S&P by 5 points or more in minutes. The delicate change in the role of the Tick Index in negative extremes require the trader to pay a lot of attention to the context of the situation.
I just summarized some of the intraday behaviours above instead of throwing out tables and more tables of the statistics I’ve collected. They are not necessary for delivering the concepts. Those who are interested in gathering the details can easily reproduce the statistics yourself to confirm what I wrote.
For avid chart readers, just open your historical charts with the Tick Index loaded alongside Emini S&P. By flipping through a year or two of your intraday charts, you will be able to internalize the principles I stated above.