How HFT Has Changed Tape Reading

[Do Or Die 10]

There have been significant structural changes in US Equity Markets over past few years. Daytrading NYSE was a totally difference ball game (way easier) until 2008. There have been some research that HFT provides liquidity and reduces spreads. However, the factual reality is totally different- HFT is silently bleeding the system. The irony is that regulators tend to restrict short selling, talk about new taxes and what not to restrict retail traders; but they are unwilling to put a simple cancellation fee of 1 cent (say) per order to discipline HFT.

 

Prices are nothing but the result of traders’ order submission discretion, which in turn is influenced by overall market structure. If the way that orders are submitted is changed, it changes the sub-minute price patterns as well. Currently it is all It is noise, subterfuge, and manipulation.

 

Knight Capital Group has examined the significant structural changes in US Equity Markets and how those changes affect traders and investors. To download the excellent reports, click here and here. Mentioned below is some synopsis from the study pertaining to retail intraday traders.

The average trade size has come down dramatically. If the average Joe trader sees thick offer and punches BUY for 1000 shares, he may end up chasing those offers which cancel lightening fast. (in illiquid stocks)

The quote rates in 2007 were about 100 times lower than the peaks of today. This simply means 'more noise', there is way too much data now to decipher trade signals.

 

Now the most troubling news:

 

 

This orders cancellation/execution or quotes-per-trade ratio shows the relative number of' 'fake' orders in market. The quotes per trade ration has gone from manageable 6 in 2007 to a formidable 50 in these times. The size on the bid/ask is very likely to be fake.

 

To see more clearly how HFT manipulates, see the Nannex reports.

The Disruptor, will sell (or buy) enough contracts to cause a market disruption. At the same exact time, this algo softens up the market in ETFs such as SPY, IWM, QQQ, DIA and other market index symbols and options on these symbols. When the disruptor strikes, many professional arbitrageurs who had placed their bids and offers in the emini suddenly find themselves long or short, and when they go to hedge with ETFs or options, find that market soft and sloppy and get poor fills. Naturally, many of these arbitrageurs realize the strategy no longer works, so they no longer post their bids and offers in the emini. Other HFT algos teach the same lesson -- bids or offers resting in the book will only become liabilities to those who can't compete on speed.

http://www.nanex.net/Research/EMini2/EMini2.html

The news about the budget impasse, the S&P downgrade, and the Fed Announcement after a tumultuous week had less impact on traffic rates than talk of a potential tax on financial transactions in Europe. This temper tantrum exposes the willingness of certain types of traders to wreak havoc on our quotation system to make their point

http://www.nanex.net/research/MsgRates/EquityMessageRates.html

 

 

 

Posting a comment will only take you 2 minutes, but it will be the strongest motivation for me to share something better.

[steve46 12]

I for one, would like to read something better....at least something more accurate than what you have posted.

 

I read the tape every day in front of my students...in fact reading the tape is the way we confirm that our trade selection is valid....We have no problem finding favorable entries as long as we stay within our rules..

 

Reading your comment, the first thing that strikes me is that you don't know what "read the tape" is.....it is certainly not reading a DOM display...and second, you do not seem to have information to share but are simply offering a tease in the effort to get someone here to supply you with information....how about providing something of (current) value instead of outdated commentary..

 

Thanks

Steve

[Do Or Die 10]

Re-posting with the images:

The average trade size has come down dramatically. If the average Joe trader sees thick offer and punches BUY for 1000 shares, he may end up chasing those offers which cancel lightening fast. (in illiquid stocks)

The quote rates in 2007 were about 100 times lower than the peaks of today. This simply means 'more noise', there is way too much data now to decipher trade signals.

 

Now the most troubling news:

 

 

This orders cancellation/execution or quotes-per-trade ratio shows the relative number of' 'fake' orders in market. The quotes per trade ration has gone from manageable 6 in 2007 to a formidable 50 in these times. The size on the bid/ask is very likely to be fake.

 

To see more clearly how HFT manipulates, see the Nannex reports.

 

Nanex - HFT Killed the EMini

 

Nanex - HFT is Out-of-Control

 

[steve46 12]

Once again for those capable of abstract thought....if you are trying to read the tape, and you are using the DOM, you are using the wrong tool for this purpose...the first "quote" talks about "thick offers".....referring to the DOM....apparently the poster doesn't understand the difference....bids and offers on the DOM can be faked, can be pulled, and they can be disguised in a number of ways.....in contrast, data appearing on the time & sales strip represents ACTUAL transactions...thus those of us who actually trade use it instead of the DOM.....

 

I think we are done here....

 

As an aside, I take some pleasure in thinking that this person is continuing (in the presence of evidence to the contrary) to trade using a DOM to read the market...by all means continue as you were......

Edited August 22, 2011 by MadMarketScientist
language

[Do Or Die 10]

Yes, I'm an arrogant prick full of overweening pride who is nowhere as smart as he thinks he is. And yes, I am using a slow old laptop running decade-old software on a thirty-year-old paradigm. And it's the most shameful mistake on entire TL to have said 'tape reading' instead of 'scalping'.

 

Now that I have got your brain working could you please get back to: http://www.traderslaboratory.com/forums/psychology/10512-sin-predicting-anticipatory-trading.html#post125013

 

Until then, please stay in the confines of http://www.traderslaboratory.com/forums/technical-analysis/10597-early-warning-again-2.html#post126099

[MightyMouse 12]

Steve46,

 

The poster posted his opinion of a topic and gave information, good or bad, on the topic.

 

You can simply respond to what you agree with or respond with what you disagree with.

 

You seem to have the habit of attacking the individual who posted which makes the forum an unpleasant experience for all.

 

Even if you where an individual who has empirically shown that he is a superior trader, it does not earn you the right to bash the individual on a personal level, especially when he did not warrant it. However, you have repeatedly refused to substantiate your ability and instead choose to inflate your frail ego at the expense of others. Somehow you hold onto the ideal that you are superior.

 

To answer a question you posed of me. No, I have never dealt with difficult children. However, I suspect that you have had the unfortunate experience of raising difficult children.

 

I have my suspicions on how they became difficult.

[Tradewinds 12]

I'm constantly canceling orders, but it's not because I'm trying to manipulate the market. The only way I'd never cancel an order, is if I were trading perfectly, or I let my order get filled even though I knew it was a bad decision. I hope I don't intentionally let my orders get filled even though I know it's a bad decision. I can guess at what price level the market is going to hit, but I can't be right 100% of the time, that means my order will go unfilled. I have to cancel it. There is no other choice. If I move the order to another price level, it's basically the same as cancelling and replacing it. Cancelled orders are unavoidable. I'm not convinced that the increase in cancelled orders is directly related to people trying to manipulate the market.

[Do Or Die 10]

I'm constantly canceling orders...

 

 

You do not place orders intentionally to make the ask/bid side appear thick... HFT trading is totally different than retail trading.

 

The quotes per trade ratio has gone up 8 times... do you think that is because of individual traders like you.

Edited August 21, 2011 by Do Or Die
mistake

[Tradewinds 12]

You do not place orders intentionally to make the ask/bid side appear thick... HFT trading is totally different than retail trading.

 

The quotes per trade ratio has gone up 8 times... do you think that is because of individual traders like you.

 

I don't know. I doubt it.

 

I'm all for creating a playing field in the market that it fair to everyone.

[Tradewinds 12]

You do not place orders intentionally to make the ask/bid side appear thick... .

 

I'd like some clarification on the term "Thick". I just did a web search, but didn't find anything that made me feel like I understand what "thick" means. I'm not familiar with the term.

[Tradewinds 12]

Here is something else I just thought of. Sometimes I click an order to cancel it, and I get a msg stating that the order can not be cancelled. So my question is this, if the HFT traders are entering an order, don't they risk being filled before they can cancel? Or maybe they have just enough of an "edge" so they can enter an order, and cancel it without worrying to much about being unintentionally filled.

[Do Or Die 10]

It's commonly used on trading floors...

Momentum Day Trading techniques

 

Option Trading | Lenny Dykstra - Deep In-The-Money Options Strategies - Who Took My Money? | 1option.com

 

Thick refers to when the bid/ask looks stable and 'liquid'. Similarly thin refers when you cannot hit/take more than 500 shares at best bid/ask.

[RichardTodd 10]

This is really not a very useful article. Losing traders need a scapegoat, and today's scapegoat of choice is HFT. There was a time when retail traders had to phone in orders, while pit traders could trade many times faster with better information. Guess who the scapegoat was back then?

 

The article talks of "significant changes" as if they are a novelty. There are significant changes in progress all the time. Financial markets are the most exciting and lucrative competition on the planet, so don't expect that to ever stop. Strong competitors are invigorated by strong competition. Weak competitors write articles about how scary it is.

 

The article says there is "way too much" data now, but doesn't give the standard by which that conclusion was reached. Don't decide for your readers how much data is "too much" for them. The article says that HFT reduces sub-minute orders and price patterns to noise, but this cannot logically be true as long as there is more than one HFT competitor. Think about it: if two competitors have made noise for each other, then neither could make money in the aftermath. Since the claim here is that HFT "bleeds" the markets, then there is a signal in that noise somewhere, and there is no sense in complaining that it's hard to find. Who ever expected trading to get easier over time just isn't thinking straight. The state of the art progresses, and competition gets harder. It's true in all endeavors, be it sports, business, art, or trading.

 

Laughably, the article pronounces:

 

The size on the bid/ask is very likely to be fake.

 

I got news for you... this was always the case. My competition isn't going to just show me their hand? What a shock! At this point it sounds like the author hasn't even played a hand of poker, much less traded a stock. I'm sure that's not true, though, so I'm a little dumbfounded.

 

Does anyone seriously imagine that there was a time when everyone was telling the truth about their intentions in the market, and no one was exploiting technological advantages over their competition? There are stories of rice traders centuries ago forming rooftop networks of flag-wavers to exchange market data faster, for crying out loud. No doubt someone back then wrote an article about how you couldn't discern trading signals against such high speed activity. They were wrong then, and this article is wrong now.

 

All any rational trader wants is free competition. I want to be free to cancel my orders as often as I want, so I don't have a grudge against others canceling orders at high rates. I want to be free to mislead and misdirect if I want, so I don't have a grudge against others doing so. The only kinds of speed restrictions that are sensible are ones that protect the integrity of the market's IT infrastructure.

[Do Or Die 10]

Okay Sir... I've got news for you too.

 

This is really not a very useful article. Losing traders need a scapegoat, and today's scapegoat of choice is HFT. There was a time when retail traders had to phone in orders, while pit traders could trade many times faster with better information. Guess who the scapegoat was back then?

 

Feel free to assume I'm a losing trader, like you are free to assume tomorrow market will go up/dpwn.

The article talks of "significant changes" as if they are a novelty. There are significant changes in progress all the time. Financial markets are the most exciting and lucrative competition on the planet, so don't expect that to ever stop. Strong competitors are invigorated by strong competition. Weak competitors write articles about how scary it is.

Ever heard of 'regime shifts'? please google for it. Whenever there is a regime shift a certain type of traders get thrown out of the market.

The article says there is "way too much" data now, but doesn't give the standard by which that conclusion was reached. Don't decide for your readers how much data is "too much" for them.

I expect readers understand how to read graphs and how to read English.

The article says that HFT reduces sub-minute orders and price patterns to noise, but this cannot logically be true as long as there is more than one HFT competitor. Think about it: if two competitors have made noise for each other, then neither could make money in the aftermath. Since the claim here is that HFT "bleeds" the markets, then there is a signal in that noise somewhere, and there is no sense in complaining that it's hard to find. Who ever expected trading to get easier over time just isn't thinking straight. The state of the art progresses, and competition gets harder. It's true in all endeavors, be it sports, business, art, or trading.

Again it gets harder for certain trading styles, and it does not changes at all for other trading styles.

 

I got news for you... this was always the case.

:rofl:

 

Can you relate a similar regime shift in past? Let me tell you one. A LOT of nyse scalpers went out of business when nyse became hybrid.

 

This regime shift again puts certain scalpers out of business. OK I hear your assumption I'm one of them.

 

BTW, I did appreciate you made such a detailed reply.

[MightyMouse 12]

I'm constantly canceling orders, but it's not because I'm trying to manipulate the market. The only way I'd never cancel an order, is if I were trading perfectly, or I let my order get filled even though I knew it was a bad decision. I hope I don't intentionally let my orders get filled even though I know it's a bad decision. I can guess at what price level the market is going to hit, but I can't be right 100% of the time, that means my order will go unfilled. I have to cancel it. There is no other choice. If I move the order to another price level, it's basically the same as cancelling and replacing it. Cancelled orders are unavoidable. I'm not convinced that the increase in cancelled orders is directly related to people trying to manipulate the market.

 

When the bid is stacked on the DOM one would guess that there are many traders trying to buy at the same time. since there appears to be more buyers than sellers. One could look at it and think the market is poised to go higher. This guess could cause an unsuspecting buyer (either a weak long or a scared short) to buy via market order so as to not be left behind. Those with the large buy orders are actually trying to get buyers to jump ahead of the bid and buy right into their standing sell orders. The buy orders are almost all fakes and are really sellers who have standing sell orders and want to get filled as high as possible. Once sold and presumably ( for this example) short, then they will distribute more contracts or shares into the standing buy orders and attempt to force the weak longs who enter long to sell lower into the sellers standing buy orders.

 

There are multiple variations of how this can unfold. you can get clues by watching the DOM and T&S. So, for example, if the bid is stacked and there are a lot of trades going off at the ask, you can expect a different reaction than if the ask is stacked and there are a lot of trades going off at the ask.

 

You can gain good information by watching both.

[Tradewinds 12]

The article talks of "significant changes" as if they are a novelty. There are significant changes in progress all the time.

 

I agree. And on a similar note, I'd like to add:

 

The nanex.net article starts off by stating the the huge increase in data is "insane". An increase in data isn't insane, or bad. It's just an increase in data. Period.

 

The biggest concern seems to be liquidity. I guess that the implication here, is that traders are abandoning markets like the SP 500 e-mini futures, and therefore the liquidity is dropping. I guess that's the issue I'd like to know more about. Are traders and investors really abandoning the SP 500 emini? And if they are, why?

[Do Or Die 10]

The nanex.net article starts off by stating the the huge increase in data is "insane". An increase in data isn't insane, or bad. It's just an increase in data. Period.

 

Consider two cases:

1. There is one signal per 24 Bytes

2.There is one signal per 192 Bytes

 

This is not just increase in data. This is increase in noise.

 

Edit: You are saying that Nanex isn't just wrong, they are stupid.

[RichardTodd 10]

Whenever there is a regime shift a certain type of traders get thrown out of the market.

 

That statement really sums up my problem with all the HFT hype. A trader does not get thrown out of the market. No one forces them to place so many losing trades. They failed to adapt and they either lost all their money or closed up shop all by themselves. Presenting losing traders as victims who were "thrown out" is not productive as long as the marketplace is free.

 

That doesn't mean I'm assuming you are a loser, but you are taking the position that a loser takes to feel better about himself. The position that something happened to him, rather than taking responsibility for his own failure to remain competitive. The position that there's "too much" data to deal with, and that cancellations should be taxed to "discipline" the new opponents. The fact is the marketplace is always shifting, and when there are big pockets of failure we come in after the fact as label it a regime change. But that's just a matter of scale. When a little volatility bankrupts a few funds, it's a smaller example of the same failure to adapt, and no one gives the event a fancy label. No pity-party for them!

 

I expect readers understand how to read graphs and how to read English.

 

And I expected the same of you, until you took my comment about giving no standard by which we can judge how much data is "too much," and responded in this vacuous way. You have graphs that show there is more data. Then you tell us that the increase is noise and that there is too much data, but you don't support those statements.

 

 

Can you relate a similar regime shift in past? Let me tell you one. A LOT of nyse scalpers went out of business when nyse became hybrid.

 

Here's two more examples I've personally witnessed: The rise of SOES, and decimalization of stock prices. Boo hoo for all those that tried doing the same thing they did yesterday and falling on their face. Let's play some violins for when they kept falling on their face until they had no money left, instead of adapting to the new conditions. But let's not, under any conditions, say that they were thrown out, as if they had no part in their own undoing.

[Do Or Die 10]

Here's two more examples I've personally witnessed: The rise of SOES, and decimalization of stock prices. Boo hoo for all those that tried doing the same thing they did yesterday and falling on their face. Let's play some violins for when they kept falling on their face until they had no money left, instead of adapting to the new conditions. But let's not, under any conditions, say that they were thrown out, as if they had no part in their own undoing.

 

Looks like I(we) can sum up here.

 

There is a significant difference between these regime shifts and the current one which we are talking.

 

NYSE going Hybrid and Decimalization were good for the over all market microstructure. The NYSE specialists lost their monopoly and provided better level playing field to more market participants.

 

However, this regime of HFT going full throttle is a drag on entire system. I do not want to elaborate on this but you may agree after a little research- HFT is giving less and taking more. By this article I do not want to show sympathy who might have lost their trading edge; just suggesting for a better level playing field.

 

You have graphs that show there is more data. Then you tell us that the increase is noise and that there is too much data, but you don't support those statements.

 

Yes, maybe I should have given these two explanations earlier.

When the bid is stacked on the DOM one would guess that there are many traders trying to buy at the same time. since there appears to be more buyers than sellers. One could look at it and think the market is poised to go higher. This guess could cause an unsuspecting buyer (either a weak long or a scared short) to buy via market order so as to not be left behind. Those with the large buy orders are actually trying to get buyers to jump ahead of the bid and buy right into their standing sell orders. The buy orders are almost all fakes and are really sellers who have standing sell orders and want to get filled as high as possible. Once sold and presumably ( for this example) short, then they will distribute more contracts or shares into the standing buy orders and attempt to force the weak longs who enter long to sell lower into the sellers standing buy orders.

 

There are multiple variations of how this can unfold. you can get clues by watching the DOM and T&S. So, for example, if the bid is stacked and there are a lot of trades going off at the ask, you can expect a different reaction than if the ask is stacked and there are a lot of trades going off at the ask.

 

You can gain good information by watching both.

Compare this with the below version. The market data for such judgements has just become more noisy, as the graphs show.

Consider two cases:

1. There is one signal per 24 Bytes

2.There is one signal per 192 Bytes

 

This is not just increase in data. This is increase in noise.

 

Edit: You are saying that Nanex isn't just wrong, they are stupid.

[Tradewinds 12]

Consider two cases:

1. There is one signal per 24 Bytes

2.There is one signal per 192 Bytes

 

This is not just increase in data. This is increase in noise.

 

It depends upon what data we are talking about, and how the trader is processing that data. If all the new data is a lot of orders that just get cancelled, then yes, I see your point. If the HFT traders are entering orders, then a split second later cancelling them, then I'd like to see whoever controls the market look for a pattern of this behavior on an individual basis, and do something about it.

[Tradewinds 12]

When the bid is stacked on the DOM one would guess that there are many traders trying to buy at the same time. since there appears to be more buyers than sellers.

 

I've never been able to make any sense of the DOM on a trading ladder. And I really don't know how to analyze Depth of Market. I've watched the numbers for the DOM on a trading ladder, and it makes no sense to me at all. I have never been able to see any correlation that I feel I can trust. And the DOM numbers change so fast and so dramatically, that I find it just a lot of "noise". I ignore those numbers, and I would just have them shut off if I could.

[RichardTodd 10]

However, this regime of HFT going full throttle is a drag on entire system. I do not want to elaborate on this but you may agree after a little research- HFT is giving less and taking more. By this article I do not want to show sympathy who might have lost their trading edge; just suggesting for a better level playing field.

 

Anyone who makes money trading takes more than they give, and is a drag on everyone else's profits. There's nothing wrong with that as long as they are playing by the same rules as everyone else. It's just like an athlete complaining if a stronger, faster, more accurate athlete walks on to the field. "now I can't score as many points!"

 

As long as I am not illegally excluded from starting my own HFT business, the playing field is level already. If you have evidence of collusion or whatever, that should definitely be stomped out. But your article makes no such claims. Your article just says there's more activity generating more data. I don't see why anyone should have a problem with that.

[Do Or Die 10]

Looks to me like we are discussing for the sake of discussion alone.

 

Anyone who makes money trading takes more than they give, and is a drag on everyone else's profits.

I said HFT is a drag in relation to microstructure changes it has brought in the markets. The article also says it's "bleeding" the system and I though it will be simple to relate that to flash crashes and front running. People who trade with size (more than 2000 shares in thin stocks, on any time frame) just know about the difference in fills they are getting. You certainly are not aware of that... but there's lot it to say in a single post. I will have to post a full article trying to convey from basic what HFT has brought and what it has taken (and again in relation to microstructure only, not why certain people are losing).

There's nothing wrong with that as long as they are playing by the same rules as everyone else...

They are not playing with the same rules as anyone else, I will show why shortly.

 

Your article just says there's more activity generating more data. I don't see why anyone should have a problem with that.

My article is about a regime shift. "How HFT Has Changed Tape Reading". Maybe I should have used "How HFT Has Changed Scalping" because people on this forum relate tape reading to Wyckoff. And my article empirically shows the regime change and the people that are effected by it.

 

Your criticism is similar to the newbies in my other thread who start complaining about the maths without trying to grasp the logic. The math ofcourse was right though explained later.

[jackb 10]

 

They are not playing with the same rules as anyone else, I will show why shortly.

 

 

RTodd is way out ahead on this so far. So you will need something substantial with your next post. And exactly what definition of "shortly" are you using?

[Do Or Die 10]

RTodd is way out ahead on this so far...

 

 

Never took it as contest; if its a contest I give up.

 

BTW shortly has a discretionary connotation which depends on my interest level