My Tape Reading Software & Techniques

[Predictor 11]

I wanted to share what I've been finding that works versus what hasn't been useful.

 

First, the summaries are AMAZING. To be honest, I had not used our OrderFlow Bars in my own trading but instead focused exclusively on the OF monitor. However, setting the OF Bar to for example a 1K volume bar and tracking the summary information is a very useful way to see if the order flow is correlated. You can view each 1K block as an "order". I like that our summaries are printed near the bars too. It makes it easy to see what information goes with what bar.

 

Likewise, the summaries on the Volume Inventory Tracker has really made it far more powerful because now I can see much clearer the activity in a range bound market. I've always focused on the Order Flow monitor which is still one of the best way to see the directional moves.. but it gets more difficult to read during range bound or ultra fast markets. I was able to track the accumulations through our inventory tracker even though it was too fast to read.

 

I track the high volume areas within bars and find the distribution useful. I'm still undecided on the shading. I don't think looking at the difference volume numbers inside the bar is useful. I like the color coding without numbers. Of course, this is just how I read the market.. others will sure use different approaches.

 

Some people claim that order flow always drives/leads the market. The reality is that the order flow can be highly mean reverting or trending-- just like the market. Trends are created by highly synchronized/one sided order flow. Both our OrderFlow Monitor and the summary stats help to see whether the order flow is correlated (trending) or mean reverting.

 

We have an experimental measure that I'd like to apply to the summaries as well. First, and I could be wrong but I don't think watching for trade size is likely to be useful. Some algorithms may look for round number executions but I've never used trade size. Our SynchronousVolume measures how one or two sided the order flow is. Its similar to the difference volume but focused on whether orders come in runs or in mixed fashion. I've found that when SynchronousVolume is higher then a certain level then the market almost always sees follow through. The problem is so far I'm only tracking it on a per price level which is pretty much the domain of HFT. One of the studies we'll be looking at is when we collect this information on a bar by bar basis whether or not it proves useful.

 

One of the reasons the order flow is mean reverting is if you take a range market.. all buys are at the highs and sells at lows. The LQ providers just flip them out. They do provide a valuable service for the market though in the ES. As long as they are "real" I don't have a problem with them.. we actually need them. I feel many of these LQ providers are using semi martingale strategies as well which is often why we see these support/resistances in the market as a function not only as a function of technical traders but of liquidity providers trying to create break even "games", just a thought.. The question then is what will cause the range to break? As stated above, the LQ providers seem sensitive to the correlation of the order flow.. As they pick up on that and respond to it. Where will highly correlated OF come from? You guessed it, large institutions/traders are spitting in large orders that are broken up in time or space to prevent showing "naive tracking", i.e big prints. We have some measures we're looking at to help us gain an even better edge...

 

... But first priority is release the software. I feel functionality wise that its ready.

Edited December 13, 2012 by Predictor

[nelsonhimes 10]

Now in the trading world all are looking for something new which are really helpful for them in particular way.. And this are one of it according to me

[Predictor 11]

Nelson, thanks for your interest. Remember, I said that I didn't see any manipulating in the DEPTH in the ES. Well, last night in after hours around 10:57-12AM I seen activity that looked like it could be manipulation. I warn anyone who thinks of trying this that the CFTC recently fined a firm for manipulation.

 

I can't say for certain if the activity was manipulating but someone flashed large size on the DEPTH in the inside level a few times and then immediately pulled it. Remember, how I've wrote that traders respond to order book imbalances. There was a large OB imbalance above the market in the where many traders were trying to clear. A trader flashed large size as we attempted to move into this zone. This triggered what appeared to be HFT market orders. The manipulation was easy for me to read and I hypothesize the intent was a large futures trader was attempting to slow down the rally to acquire more contracts.

 

My hypothesis for this activity:

 

Large futures trader was accumulating inventory but market was rallying too fast. So, he flashes large size near highs to trigger HFT response (sell market orders). He then captures these contracts and resells them. The response was for about 3 ticks.

 

It looks like the HFT robots responded immediately. But, after the first spoof they quit responding. The alternate explanation is that a large trader really was trying to exit but posting the large size triggered the HFT response and he pulled his orders. I don't believe this though because this happened a few times and the size never stayed for more then a split second. It easy to see in my histogram view.

---

 

This is just my interpretation. It happened very fast, and I cant know if this was actually what happened. I do not see this activity as a big deal for my trading and it seems fairly rare. Given it was a lower liquidity market/time then this may indicate that these games are more likely to be played when liquidity/activity is lower/risk is lower and an action is more likely to generate a response..

 

If anyone wants to check it was the 12-12 ES contract around 10:57PM.

Edited December 13, 2012 by Predictor

[Predictor 11]

Here you can see our how using 1k volume bars provides with our net/summary information gives us a different view into the order flow...

 

[Predictor 11]

Regarding my last video, I also want to share more about how I read the tape and how our clearable Invetory Tracking/Volume Accumulator can be used. What I do is I closely track the MFE and MAE in my trades and attempt to exit them with a small profit or loss BEFORE my stop is hit. It doesn't always work that way and sometimes I will exit too soon or decide to be more head strong.

 

Tracking the MFE/MAE takes tremendous amount of effort. One of the nice things about our Volume Tracking Accumulator is that if you clear it when you enter a trade then you can track all of the following from your your trade entry very easily:

 

1. The max adverse and max fav excursions since entry

2. If there has been more NET buying or selling since entry.

3. The prices where most of the buying and selling is occuring.

4. The total amount of activity that has taken place.

 

Hope it helps.. In general, I find my worst losers never FE (favorable movement). In backtesting, I've found this true in most of systems. Likewise, I've found my best trades typically have very little AE (adverse movement). Tracking this helps me to see how well the trade is actually working.

[Predictor 11]

Using 1 tick range OrderFlow Bars for reading the tape.. comparison with 1000 volume OrderFlow Bars...

 

[Predictor 11]

In this video, I reveal our tape inspector, a time/sales analysis tool, to help explain how our OrderFlow Monitoring columns work...

 

[Predictor 11]

... New video showing our powerful time & sales inspector!

 

[johnw 10]

....................................................

 

One of the experimental measures we have in the program is what we call Synchronous Volume. With this measure, we measure how one or two sided the order flow is.. what we've found is that when the "one sidedness" of the order flow passes a threshold that the price almost always follows through whereas when the order flow is two-sided then we have reversion. This one-sidedness appears to be the best indicator of what will drive price -- not just imbalance. In one respect, this measure is a "modern day" equivalent of being able to measure large size trades -- which are split up now.

 

For order execution, if I feel a reversal is imminent then one technique that I use often is to shoot in a market order when the imbalance in the book has reached an extreme (light pink/blue) and the order flow has reversed to become dominant on the other side. It is quite often possible with this technique for the trade to become profitable without ever taking any heat. This works well in volatile times/markets.

 

gm Pred,

 

Firstly I admire your tenacity, it is a very good quality to possess.

My initial reaction to your new program is that you are producing enough information to drown a moose let alone a would-be Trader .... but really this is not a matter for my concern.

 

However I would be interested in an expansion of your thoughts contained in the two paras above .....

SynchronousVolume

imbalance in the book ...how are you measuring it.

 

thank you

[Pedro01 10]

I am curious about one thing.

 

When you copied the Jigsaw Trading products, why did you not even change the color scheme?

 

You have it all - even down to the yellow/white highlighting....

 

I think it's a great compliment to Jigsaw that you were afraid to even change the colors, lest your might miss something!

[Predictor 11]

johnw,

 

Contrary to increasing the information demands on the trader, nearly every aspect of our software produces higher quality information while reducing the data required for the trader to process. This has the effect of slowing down the market. For example, the consolidated tape scrolls about 10x slower then traditional tape, and if we process it to the same level as we do in the OrderFlow Monitor, which I'm thinking we will at some point in the future, then it would slow down by another factor of 2. Obviously, not every trader will have the tape up or need it and those that do will probably just use it to gauge overall market activity.

 

There are some cases though when our consolidated tape would be very useful. If the market has a large bid/ask spread and a narrow tick then our software won't allow tracking such an instrument well because there will be a large number of blank rows. On such an instrument which we really aren't designed for, the consolidated tape could still prove valuable. It can also prove valuable because we don't clear cells in the OrderFlow Monitor, as such if a trader wonders if a certain volume came through then they can do so in the tape.

 

I don't say this as marketing spiel. But, I believe that we've produced the most powerful and at the same time easiest-to-use tape reading software ever, and I'm sure that many experienced tape readers will agree. I think one of the reasons we've been able to do this is that we had a clear knowledge of what we were trying to accomplish.

 

SynchronousVolume measures how 1 or 2 sided the order flow is. At market lows and highs, we will generally see a lower synchronicity because the order flow becomes 2 sided as market buy orders are shot in whereas we see price follow-through when we have 1 sided order flow. The reason for this is that the liquidity providers can't flip out the traders and pull out or reprice. I've shared most every algorithm fully here but am not sharing the specifics for this one, at least for now.

 

The exact method to calculate order book imbalances is customizable. However, I've found that measuring differences in the inside levels to work well. Its a hunch on my part that the market will likely respond to inside levels while being attracted to outside levels. We just sum the levels the traders want to compute (take the resting buys-resting sells). We find that at extremes that high frequency traders respond to these imbalances and will shoot in market orders in front of them to capture a few ticks. I would never trade just on the imbalance but use them for order entry enhancement. It takes a strongly trending market to blast through order book imbalances. OB imbalances often are coincident with the market highs and lows, as well. We find high volume in these areas due to the limit order exhaustion, as well.

 

Pedro, as I said I wanted the very best tape reading software and no existing software met my requirements. There are good programs out there. But, you won't find our best features in any other program and you won't find any other program offering such a complete package -- unless they copied us. We know the temptation will be high as larger companies realize we've "one upped them" -- and that's the only reason we've made an effort to TM our best displays like PressureVolume. Pedro, I certainly encourage anyone to try the other programs out there. We're not at all concerned because I personally know that our software is the best. If it weren't we wouldn't have developed it because it was very expensive to develop it and only developed out of necessity.

Edited December 19, 2012 by Predictor

[johnw 10]

johnw,

 

...........................................

The exact method to calculate order book imbalances is customizable. However, I've found that measuring differences in the inside levels to work well. Its a hunch on my part that the market will likely respond to inside levels while being attracted to outside levels. We just sum the levels the traders want to compute (take the resting buys-resting sells). We find that at extremes that high frequency traders respond to these imbalances and will shoot in market orders in front of them to capture a few ticks. I would never trade just on the imbalance but use them for order entry enhancement. It takes a strongly trending market to blast through order book imbalances. OB imbalances often are coincident with the market highs and lows, as well. We find high volume in these areas due to the limit order exhaustion, as well.

 

 

Thanks Pred,

I am learning slowly, however I am unfamiliar with 'inside and outside' levels

I guess these are trade terms and I would appreciate an explanation of them and how they apply.

thanks

[Predictor 11]

john...we can selectively compute the resting orders based on the inner most (best bid/best ask) or outer most (worst bid/worst ask) or just take a difference for all levels. I only pay attention to these numbers when they hit extremes and then I see the color highlights. So, it doesn't take any real effort. We call this LimitResistance or BidAskAtPrice or just orderbook imbalances.. I also look at high volume areas as possible LimitResistance but that form is not caused by imbalances visible in the book by a refilling limit/iceberg order.

 

..... Often a large imbalance in the book will be backed up by these hidden orders. I've started tracking inventory also... its possible with our software to now see where most traders are short and long from... it is possible to combine this to get new insights into the market... This inventory tracking provides a whole new level of ability to read OF

 

Thanks Pred,

I am learning slowly, however I am unfamiliar with 'inside and outside' levels

I guess these are trade terms and I would appreciate an explanation of them and how they apply.

thanks

Edited December 19, 2012 by Predictor

[Predictor 11]

Pedro, you may want to ask who copied who because I can see that others, for the first time, are copying some of my many innovations -- like dominant side highlighting -- which I never seen any program have until a vendor posted in the day trading thread today. They even copied the color.

 

Even so, I still still encourage any trader to try the competition.. even though they're attempting to copy some of our innovations. They don't appear to be anywhere close. And, there are other more basic reasons I built my own software and one is that it is very difficult to get this stuff right. And, I'm sure we'll have some bugs when we release -- all software does but I encourage you to actually debug the output of my software and compare it to other programs which you think are similar. I suspect you'll find that the algorithms we built for doing this stuff are superior: hint: actually work.

[Tams 20]

pedro, you may want to ask who copied who because i can see that others, for the first time, are copying some of my many innovations -- like dominant side highlighting -- which i never seen any program have until a vendor posted in the day trading thread today. They even copied the color.

 

Even so, i still still encourage any trader to try the competition.. Even though they're attempting to copy some of our innovations. They don't appear to be anywhere close. And, there are other more basic reasons i built my own software and one is that it is very difficult to get this stuff right. And, i'm sure we'll have some bugs when we release -- all software does but i encourage you to actually debug the output of my software and compare it to other programs which you think are similar. I suspect you'll find that the algorithms we built for doing this stuff are superior: Hint: Actually work.

 

lol !

[Predictor 11]

I was running our time/sales today in my live trading and liked having it up. I realized we could do better... check out our Matrix TIme & Sales Display...

 

[traderbarry 10]

3 questions:

 

1. In your videos, sometimes you say "I" and other times "we." Is it just you, or a team of developers/traders, who developed this software? I am guessing it is just you but the we sounds like you are a big firm.

 

2. You mention that your software likely has bugs as does all software. What protocols are you using to ensure the software bugs are minimized? This is critical with my money of the line.

 

3. It seems like many of the topics you discuss are specific to your software ---- meaning it would be hard for me to try it without your software. If that be true, then this whole thread is just one big advert for your product. Is not that against the forum rules?

[Predictor 11]

1. In your videos, sometimes you say "I" and other times "we." Is it just you, or a team of developers/traders, who developed this software? I am guessing it is just you but the we sounds like you are a big firm.

 

We sounds like a big firm? Why? Many people refer to their company as "we" instead of I. We are not a big company. Do you really think I'm a great enough programmer, yes I'm good, but to do all this myself? Much of it is my handiwork, and I'm proud of it. But, I've a partner. Many "big name" products were developed by only 1 or 2 solid programmers.

 

2. You mention that your software likely has bugs as does all software. What protocols are you using to ensure the software bugs are minimized? This is critical with my money of the line.

 

Look, I mentioned this in the CONTEXT of encouraging our users to check the algorithms and output of our software and other software out there. It has been my experience and one of the reasons I developed my own software was because I found that not only was there no product that did what I needed but that even rudimentary programs had basic flaws and errors in their output.

 

Do you ask other companies this? Do you go to Microsoft and ask them what "protocols"? I have personally never went and asked a company this. I trade with the software live and with real money. I'm serious about the quality and have extensive software development experience. As with any software, there is a risk of problems. This is true for any trading software out there and no we won't be responsible if a trader loses money with our software -- regardless of the cause. That'd be stupid and if that were the litmus test there wouldn't be any trading software. Our software doesn't place orders. I recommend that all traders have a second machine not running anything but the order routing software in case of any bugs.

 

3. It seems like many of the topics you discuss are specific to your software ---- meaning it would be hard for me to try it without your software. If that be true, then this whole thread is just one big advert for your product. Is not that against the forum rules?

 

Have you actually read most of my posts here? I think if the forum administrators didn't see the value then it wouldn't be here. No, they've actually encouraged me to share more because they see the value.

Edited December 21, 2012 by Predictor

[traderbarry 10]

You seem very defensive I'm trying to figure out why.

 

A 1 or 2 person firm may be better than a large firm and I am glad you clarified. If the forum admins allow this thread then ok.

 

Those questions you answered.

 

You side-stepped number 2 and that concerns.me. Getting offended at a legit question is not a good answer. But if that is your answer I will pass on the product and this thread.

 

You must not remember me from the Meetup talk do you?

[steve46 12]

Traderbarry

 

If you are interested you ought to check out the thread "day trading the Emini Futures"

 

It may answer your questions (or perhaps raise a few more)

 

Good luck

Steve

[Predictor 11]

>>You side-stepped number 2 and that concerns.me. Getting offended at a legit question is >>not a good answer. But if that is your answer I will pass on the product and this thread.

 

 

Good.. don't need you. You sound like you might be a high maintenance customer. Honestly, you come on here with 1 post and make/imply some claims about my company and motivations. I have a lot of folks trolling me here and so that's not the right way to approach someone you want to do business with.

 

Look, we make every effort to make our software as highest quality as possible. I have a test plan for checking a huge number of things to ensure they work. I trade using my own software -- so I've a huge incentive to make sure we get it right.

 

I've got about 20 emails from people who are wanting to buy it and aren't asking questions about what testing methodology we use. Sorry, if I come off as abrupt.. next time be more friendly and you'll get a friendlier response.

Edited December 21, 2012 by Predictor

[traderbarry 10]

I did not realize that I needed multiple posts before I could ask the master a question.

 

Please forgive me.

[TL Staff 10]

>>You side-stepped number 2 and that concerns.me. Getting offended at a legit question is >>not a good answer. But if that is your answer I will pass on the product and this thread.

 

 

Good.. don't need you. You sound like complainer. Honestly, you come on here with 1 post. I have a lot of folks trolling me here. I had a guy wanting to buy a system for 3k and he sounded like he'd be a problem customer, I was glad I didn't make the sell.

 

Look, we make every effort to make our software as highest quality as possible. I have a test plan for checking a huge number of things to ensure they work. I trade using my own software -- so I've a huge incentive to make sure we get it right. I cited that some programs out there had basic errors in their algorithms. No I won't name names.

 

I'm going to be offering the program for a very affordable price. My goal is to make it affordable even for someone with a small account. The cost for that is that I'm not going to be able to "hold hands". I'll do my best but we'll have to make some rules regarding the support up front. It might be we offer 2 versions.. low support/high support.

 

I built this for myself and don't need to sell it but I've got about 20 emails from people who are wanting to buy it and aren't asking questions about what testing methodology we use. Sorry if I come off as abrupt but again if you seen the amount of trolls that post nonsense you'd understand.

 

Hi guys,

 

Seems like these threads get a lot of heat.

 

We definitely want useful posts and threads for traders at this point. Selling software is not really the focus of our forums, but evaluating them is, so traders can make their own decisions.

 

Carry on the conversations without shutting anyone down, and maybe in our upcoming forum re-org we will try to put discussions related to vendor sales in a specific area. The questions about the software are definitely valid, since trust is important.

[Predictor 11]

TL.. see my revised post. Thanks.

 

Barry, again I apologize if I came out as rude but it sounded like you were trying to MAKE statements about my software and my company. Also, I traded with the software in early early alpha when it did crash. I had a backup plan. All good traders should have backup plans. I personally recommend that the trader have a separate "clean" machine for placing orders. You also suggested that the admins take my thread down. Again, as I said if you want do business with someone then be friendly and you'll get a more friendly response.

 

Now in all seriousness, if you're really that concerned that a program might glitch up (one that doesn't place orders) then maybe you should rethink things. Our software isn't a buy/sell indicator but a powerful tool made for professionals. As for the algorithms/output data, I recommend everyone to validate that for themselves... I do.

 

Thanks for your interest. I hope your question is answered. I don't have time to go into any more details about our test plans. We will have a ticket system for customers to submit issues and resolve them.

Edited December 21, 2012 by Predictor

[Predictor 11]

One of the misconceptions about tape reading is that its all about the market orders. In fact, the skilled tape reader can determine where the larger limit order traders are sitting off market. Why is this important? Because often entering with the limit order trader is the lowest risk trade.

 

Obviously, if it were as easy as watching for buying or selling then anyone could tape read but the real skill lies in detecting what the buying or selling means. With training, practice, and experience it is possible to determine where traders are finding value.

 

Often shorter term market order driven traders will drive price away from the slower moving limit order traders. It is important to see how the limit order traders respond. Often, they respond very slowly and these moves tend to reverse quickly leading to stop outs.

 

There are two forms of Limit Resistance . The first form are visible orders that are shown in the book. We find that when many short term traders try to do the same thing at the same time that it leads to reversals or at least makes it difficult for the market to continue to move because it creates a situation of over supply. The second form isn't found in the book but is the result of a replenished offer/bid. This results in high volume printing on a level which will often be contrary.

 

On a longer term basis, it is possible to track where most market order traders are long and short from. Watching how the price moves in relation to these imbalances often provides critical insight.

Edited December 22, 2012 by Predictor