It seems that many people try to implement tape reading into their trading, the subject in itself and what information that can be gained from it, to me, is a fascinating subject. Tape reading is an art and frankly tricky business, and I have yet to find any detailed course on it. I will however try and explain my hypothesis on it and how I implement it in trading the S&P. You can take it for what it is, it is simply my two cents, and hopefully something can be gained from it.
As the single most liquid instrument on the planet the ES is prime for the largest players in the world who can throw around millions or even billions into positions. What makes the tape tricky is that it doesn't give you a clear indication as to whether someone is entering or exiting the market. Big players will also build large positions over several different prices in order to push the market in their favor. They will place huge hidden orders on the inside bid or offer and will use algorithms to either constantly refresh the price as contracts print into it, and will also use them to instantly build the preceding offer on a tick up, or a bid on a tick down. They will also use algorithms to place large orders at market in smaller pieces. They will also play both sides of the market, buy, and sell, to keep it stable while they accumulate contracts. The purpose of the tape for me is to only monitor large orders coming in at market. Currently I am using either a 25 or 50 size filter on the Jigsaw Recon Tape with a block filter of 100, and I will treat an instantaneous string of orders of 25 or greater as a block buy or sell.
To me, it's all about where large size steps up to buy or sell the market, and the context of what has happened preceding it. For example lets say that price has put in a new high and is currently four ticks off. Naturally there are those who are going to try and fade it just as there are going to be people that are playing the breakout of that high. This is where the tape can give me clues as to what is more likely to happen. If price reaches the high and you have little buying pressure in the form of large size to push it through and it is met by large size stepping up to sell and/or there are thousands of contracts being absorbed on the offer in the form of large limits or constantly refreshing and instantly building offers you might have an opportunity to fade it. If however price approaches the high at a rapid enough pace, and there is large size stepping up to buy on the tape a couple ticks off I would be more likely to try and play the breakout, and if it doesn't break the high and push through the high, then I will scratch the trade. I think the simplest and most logical rule to follow when playing the break out is "if it doesn't break, get out."
Now, let's say that price has fallen off 6 ticks or more from the high. If there is strong buying in the form of large size down further then that to me indicates that there are large players that are TRYING to push it through the high; they may not succeed, but to me shorting is not an option, because they could easily run it through the high and then frankly you are f****d. But, lets say that price has fallen a few ticks off the high and the tape shows little buying and strong selling comes into the market in the form of large orders. That is how reversals usually happen and I would be looking to short. Usually what will happen is price will drift a few ticks from the high or low and then large size will step up and push it the other way; this combined with longs or shorts covering and those getting squeezed or stops run will cause a reversal and it's also how the "smart money" will get the best, wholesale price. They will be buying into the selling pressure if the market is moving down and is near the bottom, or selling into the buying pressure near the top. The same thing can be applied when the market is in let's say a balanced day, or a trading range. I know that it is going to take some considerable effort buy the buyers or sellers to break the range in either direction due to the fact that so many people will be playing the range, and by looking at the orders coming in on the tape it will tell me if fading the range or buying the bottom of value or selling the top of value is a viable option. I will also look for the rapidity of how price approaches the top or bottom of the range. If it took 4000 ticks or ten minutes to get to the bottom of a trading range and suddenly large orders come into the market and it takes 1500 ticks or 3 minutes to get to the top of the range it could be an indication that price is going to break the other side of the range. If that is the case I will not fade the range or value and look for the market to break and range extension to come into play.
Big players will also defend their position from going against them. Today it was apparent in the morning session of the S&P. After the initial push down, every time price rose to between 1763.50 to 1765, large limits on the offer, constant refreshing of the offer, large selling at market. Those who had forcibly pushed the market down were in no way going to allow it to go against them passed a certain point, even in spite of some size stepping up to bid the market up, and that proved to hold, amazingly even through fed day.
Now let's say the market is rallying up and you see "some" size in the opposite direction on the tape. If it doesn't stop the market chances are it's profit taking. If it's not profit taking it's arbitrage. If the selling stops the market chances are it's going to at the very least become a resistance area, or an area where they squeeze the longs to buy at a better price and then the market goes higher, or an area where the market may pivot and reverse.
In my opinion, in order to successfully read the tape size is the only thing that matters to you. Not what 100 5 lot traders are doing. Or 1000 1 lot traders, but the real size, the size that moves the market. To viably incorporate tape reading into your trading, you need to not only understand where, but understand WHY. If I'm Goldman Sachs and I just sold 10 thousand contracts to push the low and it's bounced off four ticks, am I going to let it reverse on me? No. I'm going to sell 20 thousand more and push it lower, then gradually exit my position into the continued selling pressure and then, THEN the market can reverse.
This again is my two cents, take it for what it's worth.
Last edited by Orion; October 30th, 2013 at 09:22 PM.
The following 23 users say Thank You to Orion for this post:
How long would you say you will hold the position and how do you determine if its the aggressor which is going to move the price or the passive absorber.
I usually take my limit orders are around 1.25 Tick above below the recent range when the US cash opens and market is trying to find the buyers sellers to get an edge and trade lot less but with size and trade first 2 hrs and sometimes last 1 hr.
What i understand is that to go up market has to go down first and vv. and take into account ONH ONL and what Europe did.
Somewhere i read that 8 HFTs in E-Mini S&P does 60% of volume and they profit a $1M a day from their operations and hold the price for a very short time frame where they will put initially a feeler into market to gauge where are buyers sellers are and then take position against and their limit order are filled by the stops.
E-Mini S&P might be FIFO but like you said they flood the market with orders during market open so as to get a guaranteed fill and by reducing their size they don't lose their priority in the Queue which makes it essentially a Pro rata Algorithm.
They use extensively Order Qualifiers to take advantage like IFM, Minimum Display Quantity(Read ICEBERG) etc.
Good Luck trading!
The following 2 users say Thank You to FesxEurex for this post:
Very interesting, it would explain well the instant building of a bid or offer on a tick up or down, the HFT algos probably utilize this to become first in que. I think that the ONL and ONH are very important, I have heard that there is a very high probability that one or the other is breached during the New York session. I think that really the two most likely things that happen is that there is continuation in the direction that the European session Running stops to get a better fill price is also something you see every day. It seems to me that they also run stops to push the market the other way. For example there is a resistance level that has been holding for some time or it is a historical resistance level, you could assume that many, many stops are placed just above it by people playing the pivot of the resistance level. If price comes back to test it they may push it right through and the stops will act as a catalyst to carry the market up and through the resistance level.
I made a video of a trade I took today in which I will try to explain what I am trying to emphasize. Watch it in the highest quality possible or you can't really see the numbers. At the time of this trade price had fallen 7 or so ticks off the high of the session. However, instead of reversing down acceptance of higher prices started to develop, as you can see by the accumulation of contracts in the 58's. Now when I see this acceptance of higher prices on a day like this I am looking for a run of the high and continued vertical development of the profile. I see some relatively strong buying on the tape and Jigsaw depth and sales down below so I figure that the bulls are still in this and they're going to make another run through the highs, and that is the hypothesis for this trade. The trade took 40 minutes to develop so the middle part of the video is played at 5x speed.
The most important confirmation I got in this trade was from 3:15-3:45 in the video. You see a spike in buying, large orders appear on the Tape to the right, and at the current offer price more than 3 thousand contracts trade in a manner of seconds at one price. This made me feel very comfortable that I was on the right side of the market, even as I took a little heat. Another factor in this trade that I liked was the thickness of the market represented by the cumulative depth. What I have found is that usually if you have cumulative depth of over 15 thousand on both the bid and the offer represented by the yellow print on the top and bottom of the levels in the DOM it indicates a strong market. You can see at around 5:50 that there is some size trying to push the market down, but not nearly enough to impact the market greatly, and certainly not enough to justify me exiting the trade.
Eventually I get the desired result of price returning to the high, but what I want to see as price reaches the high is a pop through with large size on the buy side of the market. It didn't happen, this combined with some sizable selling at market and very large limits on the offers made me exit the trade; at this point I was 45 minutes in and didn't want to risk price forming a double top and reversing. Price did eventually push through, but not until 15 or so minutes later. In hindsight I probably should have kept the trade longer, but anyone can say that in hindsight.
What's important is that when you see that kind of size buying down below the high fading is the absolute worst thing you can do, because if it returns to the high chances are it's not on accident, and every time it returns to the high the greater probability that it's going to push through the high, and if it pushes through the high, it could very easily do so by 2 or more points, which is ultimately what happened. In my opinion if you are going to fade you need to do it as close to the high as possible, and if it blows through you get out, unless it is apparent that a full blown sell-off is occurring and you see very heavy selling down below. Hopefully this video can explain more of what I was trying to iterate in my original post. Once again it's just what I have arrived at from my careful observation of how the market behaves, and should be taken for what it is, one traders idea, not an absolute.
Last edited by Orion; November 5th, 2013 at 03:49 PM.
The following 5 users say Thank You to Orion for this post:
Usually i avoid playing the close because its more muddier in the close and big volume is acting on finding the settlement price which they know what they want and we don't know. After they got the settlement price they are free to do whatever they want.
About the cumulative depth its fake volume spammed by bots nothing else is their watch as price reaches that level either it will flip or do a lot of hidden limit orders read icebergs.
I though it will take the 1764 on the close but it fell from it and joke is it has taken the 1764 in overnight session on low volume lol
I have seen Jigsaw maybe you will find this even more interesting Educational Videos - OrderFlowDashPro I dont use them and not promoting them but they have taken Jigsaw one level ahead. Looks like an jigsaw mutant.
Sierra also does some few Jigsaw features in lot less cost
Again i thought its the context which will win because bots will muddy the DOM and the volume price relationship.
The guy at Rancho DInero did a study about the tape size.
Conclusion: It’s still a big win for the little trade. Small lots at high frequency move the ES. 1 – 4 lots dominate the trade. The most frequent big-lot trade could only muster an measly quarter of a percentage of the total trades.
Hidden Limit Orders(Icebergs) break big order into smaller orders and CME stopped supplying this data about BIG orders.
Maybe you see a lot of small orders but they are part of big orders or an E-Mini Trading Room operation.
On market open HFTs flood the market with their orders at each price possible. they put Hidden Limit Order,Queued orders, Activation Pending Orders etc.
And as the price approach their level they reduce their order quantity in the direction they want to go that why you see market keeps flipping and reversing at support and resistance where they are active.
Thanks for the interesting link. He ranks the importance by "lot size", but am I the only one who immediately thought that this in itself does not constitute any evidence towards finding if size controls the market or not ?
If the tape is made up of 98 1-lots and 2 50-lots, the 1-lots win 98%-to-2% in the "lot size" battle, but guess who moved the most size ?
I'd like to have your thoughts on this (and even the author's if he hangs around here).
As a side note, I thought it would be interesting to re-order the study's lot sizes by volume produced.
Please register on futures.io to view futures trading content such as post attachment(s), image(s), and screenshot(s).
Interesting how the 10's and 50's move up, and the 1-2-3-4-5's still demonstrate their overall importance.
The following user says Thank You to krazyanyway for this post:
Really are making important contributions. I want to consult you who have much more experience, if there is a good course to learn how to read the tape .. It would be necessary, right? would be very important for me to hear your opinion.