Lawrence Chan at Neoticker proposed a charting concept based on this idea. I believe there was an ET chat a few months ago. I would like to put a few queries and comments out at this point: 1. One piece of software that has some very unique properties relating to DOM and Market Depth is ButtonTrader. From a depth perspective it shows volume at each individual level or price or it can accumulate or stack "aggregated" volume. Since both FT71 and Scientist have put forth the following proposition: that the market will tend to trade towards "size", then should not the market trade in the direction of the biggest disparity as shown by the "aggregated" MD volume over time. Is this a valid observation? Kevin Koy in Market and Market Logic also provides some interesting insight here: Another nice feature of ButtonTrader is the volume histogram, which essentially charts the changing quantities of bids and offers in the book and it's lovely tick charts. Really nice stuff. TT also have a volume distribution alongside their ladder, but I don' think it is that useful. 2. I have to agree with FT71 basic premise that the "prints" are far more important than the nominal bids and offers. Thus when a trade is made, this shows conviction on the part of the buyer or seller, especially if the market was hit/taken. J. Peter Steidlmayer the creator of the Market Profile in his books emphases this concept constantly: transactional data is far more important than nominal quotes. This is one of the reasons why I would never trade cash FX; most of the portals do not show this data. 3. Science_Trader posted a raft of links earlier in this thread. Unfortunately having spent most of my school days trying to avoid school, I found most of the material above my head. I was wondering if those with a scientific bent could provide us with a summary in laymen's terms (sorry for being an idiot!). The above is food for thought only! Regards. G.
"One piece of software that has some very unique properties relating to DOM and Market Depth is ButtonTrader. From a depth perspective it shows volume at each individual level or price or it can accumulate or stack "aggregated" volume. Since both FT71 and Scientist have put forth the following proposition: that the market will tend to trade towards "size", then should not the market trade in the direction of the biggest disparity as shown by the "aggregated" MD volume over time. Is this a valid observation?" (UKXGERARD) I use and am impressed by Button Trader. And I do look at cumulative depth a few ticks either side of the market as well as the size at each price. Impressionistically, I don't find that the market heads into the side where the cumulative depth is greater; it may have a slight tendency to do so, but not noticeably. However, it does have a noticeable, and I should have imagined measurable, tendency to move towards deep pockets of supply or demand. This is the behaviour the FT71 has described, and it is familiar, not least because it is exactly the reverse of what a casual theory of market depth would suggest.
This thread has really provided some interesting ideas as far as Bid/Ask Analysis. For example, this afternoon, I created a condition in CQG that would look at the bid/ask ratio of size (best bid/best ask) and would go green/red depending on whether that ratio is above 1.0 or below. If it is above 1.0, then the bids are greater than asks and the indicator is green. Hence, we are expecting it to go up as more bidders bid it up. This is a very crude way to look at it, but it is good enough for this discussion. I put the condition to color the line on a line chart and just watched the ES for the last hour. The ES rallied from 2:45 CST onwards from a low of 1214.25 to a high of 1217.25. The price line was generally RED (meaning the bid/ask ratio is less than 1; hence, the ask is greater than the bid) on the rally. If you follow standard supply/demand wisdom, you would think that the price should go down because the offers were greater than the bids. Actually, the line would be red, then as the offers are lifted, it would switch to green. As soon as the market is bid at that next price up, it would be red again for a while and then green for a second or two and again red. This happened all the way up through the highs. Try this on your charting software. One last comment (sorry about muscling my thoughts on this thread), I want to clarify that I have no intention of saying that the market always goes to size. I don't believe that the market always does anything. Like I said before, there are many factors at play at any given moment; many variables to quickly consider when scalping. The most important of these would be the prints and approaching s/r levels. My intent is to say that oftentimes in a breakout move or trend, I take comfort while scalping long to see offers coming in and prints continuing on those offers (or vice versa). Most people's instinct is to think, supply is exceeding demand so price should fall or whatever. My point is that this is not necessarily true. Generally, seasoned traders will sooner or later come to the realization that in order for one to sustain profitability in the markets, they will have to rethink their perception of mass psychology. One almost has to learn to walk backwards and think through each step to start walking forwards. This is the process that has manifested in me this year. I find that I need to always think of what needs to happen in order for a big player to achieve his max take from the market using the DOM, tape and flow. I will attempt to record the CQG screen with this indicator on it for a bit and show how although the DOM is poised in one direction, price goes the other way. I think this will serve the basis of this discussion effectively. I hope we haven't gone off topic here. Ultimately, the OP's question was about DOM patterns and volume/tape discussions are integral to this. Any thoughts are appreciated.
Following on from FT71 post, here are a couple of interviews with EUREX traders who by definition are master DOM/tape readers. I don't think this is going off topic, but in fact supplementing the discussion. I found the interview with Rotter to be immensely valuable. Notwithstanding his notoriety and the rumor he has "tweaked" software, I think he is a very special and talented trader. Martin_D gave me a heads-up that he will be interviewed in a forthcoming issue of Trader Monthly magazine. My take on trading DOM is this: over time one tends to pick-up (whether consciously or unconsciously) patterns which repeat themselves. It was exactly the same when I traded in the pit, but I believe they were a lot more readable and far easier to pick up on. So the moral of the story is that nothing beats hard work and spending time behind watching those screens. The links: http://www.financial-spread-betting.com/Paul-rotter.html http://www.financial-spread-betting.com/Adam-passaglia.html Anyone have anymore good interviews from Traders' Mag (Germany)? Regards, G.
This is a similar interview that was done a while back where Paul Rotter shared some of his methods by claiming that CCI was a key indicator for him. I believe he is just throwing his audience a small plastic bone. His results have been remarkable. I am sure that he will be seen as a Market Wizard as time goes by. The posts by Science_Trader and Growltiger have really got me digging deeper into CQG's programming language to see if I can come up with something that can reinforce or discount my point about trading towards size and how the DOM configuration impacts all this. Although I am handy with CQG, I tend to not dig to much unless I'm toying with some kind of backtest for a swing system or something. As I work on this, I find that I might be able to come up with a bid/ask indicator that will plot right on prices and will show me if the best bid/ask on the DOM create a pattern that is consistent. I'm guessing by Science_Trader's remarks and the observations from this afternoon that I will find some correlation. I will share the results here for those who are interested and following this thread. BTW, I have been a long time follower of the idea that market trades to size. I find this especially true while watching the ES all day long from 7 am to 3:15 PM every day. However, I will try to confirm this by putting an indicator right on price to see if bid/ask volume is a "ghost" or if we go to/away from it. Good luck everyone on your final days of the year.
The analysis (many years of data on stocks/futures orderbooks) I carried on that type of indicator (whatever on best bid/ask or deeper in the market depth) show that this is almost random. I can confirm that this indicator is mostly inverted with respect to the law of supply&demand, but it is not always the case. It is very difficult to use it as a trading tool. It is interesting to note that this indicator can be good for almost two weeks, and after that won't work for an extended period of time. Personnaly I think supply&demand is a poor theory... and I really wonder why economists still refer to it...
This is what I suspected all along. It is interesting to verify it in the market though (like I don't have enough going on already). I believe that I'm performing this kind of analysis in the back of my mind while trading and it is based on factors other than just the bid/ask (I can't put my finger on it, but it probably includes things such as orders changing higher or lower in the DOM as well as the way the prints are coming). You have provoked some ideas in my mind though about this and I thank you for that.
Abogdan have commercial product. Explain in psike tread. He not active ET current. Product measure flow in flow out at all DOM level. Merchant think is dom base on own experience. But ET like dom thing. You dom. you look up.
Great thread. Though I am not a DOM expert, I thought I'd throw out some of my observations. I have always found the DOM interesting because of the supply/demand base (Disclaimer: I have an economics background......) The biggest problem I find with working with the DOM using the underlying concept of supply and demand is the problem of the ask moving up with greatly diminished size (or bid in a declining market) as FT71 pointed out earlier. I still havn't figured out a good work-around for this problem. My original theory was that the side that was more "weighted" toward the price would be the dominant side, as it would show a more urgent buying/selling interest. I was going to explore this using simple derivitives, but the bid/ask shrinking with movevment screws things up. Also, I have read a few papers on the DOM (known in the economics world as the double auction market). One thing I found in common among most of them is they assume the last trade to be a random event. In other words, the bid/ask will be hit with roughly equal frequency and there is no significance to the location of the last trade or the size of that trade. I think this is wrong, but I havn't found any applicable (academic) research that takes the trade bias into account. Sorry if this sounds incoherent; trying to succinctly convey personal reading/research in a manner that makes sense to others is difficult.