Thanks for starting this thread, it's an excellent topic. I have a gizmo that I set up to show me DOM imbalances visually. I currently have it set up for ratios of 4:1 on the inner 3 levels and 40:1 for the inner pair. When both trigger I'm pretty confident it is a Wall and the other side is weak. T&S (pressure bars) and tick volume complete the picture.
Very good, now when you see a wall do you try and scalp to the wall or wait for it to punch through the wall? Do you avoid trading when there is a balanced bid & ask? Is there any way to measure the volume of pulled orders on the DOM? For exampe, if a wall appears at 13000 at 150 contracts, and you subtract Time and Sales information for trades at 13000. If you would see a lot of contracts dissappear from the DOM it would suggest a punch through that level, correct? Thank you, great thread!
Maybe a bit, but the BID/ASK ratio for traded volume generally looks quite sensible from IB feed. Attached is DAX chart (from IB feed) I saved a little while ago. Constant volume bars 500 contacts wide. BID/ASK ratio is smoothed with 7 EMA. FYI blue bands are developing MP upper and lower value areas and black plot is VWAP. IMHO the bid/ask ratio for traded volume is best viewed on constant volume bars and with a little smoothing. Bar widths and probably amount of smoothing depend on the instrument.
I had a shot at this using the IB feed but ended up with a bit of a mess. Quite possibly it was my code which I will revisit when I get a chance. There are better feeds for this. For example OpenTick book feed has an explicit BOOK CANCEL message for individual order cancels which should do the job nicely.
So far no one has introduced the largest variable in the dynamic. To make use of any information you have to use all of the elements in the data set you are beginning to examine. I am glad to see that many are aware of what they now see as conflict. there isn't any but it sure looks like there is as you become more conscious of what is going on. The people who are excellent at carving the turns did have to go through the process of getting everything on the table. There are also a lot of big money people who are trading. they may not be very skilled but they often make up for this by taking on roles that are designed primarily to deceive. No one really cares about the small fish since they are only around for a period of time before they necissarily dissappear. Some tools that a person can delop for trading are done by analysis Taking apart to find the pieces) other tools are better developed by synthesis (putting the pertinent pieces together). This is one of the last major tools that can be added to one's quiver. To get a leg up, you need to do both analysis and synthesis. Obviously you can see some contributors do one and others do the other. Be sure to understand which way the market is going as you make the effort to fathom what is going on. as a person ventures the stuff that is not in books, it is a good idea to keep taking reference notes. As you discover each false piece that you become conscious of, you have to go back (in a different color) and correct the ribbon from the point where it began. The most important thing along any path is to not deny yourself the trip. It takes many laps to become familiar with the trail. after you know the trail you can make time on it getting somewhere. This is like the requirement the mounties set for people heading out of Skagway to get to Lake Bennet to build their boats and rafts from the headwaters of the Yukon river to ride on down to Dawson and the Klondike. A person had to have 1200 pounds of supplies. You can guess how many times the trip was made up and over the Chillicut Pass across the watershed divide. If any count changes you have to know exactly why and who did it.
1) If I get the signal, I would still trake the trade, but would close it the in moment notice if it does not move my way. 2) The refresh rate I have set for DOM is 200 miliseconds, so I can subtruct all coming trades. Not sure how to represent all of it graphically. I also got a confirmation that T&S will be enabled in Ninja 6.5, sometime this fall.
I'm mainly looking for obvious walls when I am already anticipating change (in trend). When the DOM is translating you would expect price to move to size as a consequence of the trending, driven by market orders. Who'd put on a limit order if price is moving away?
I have witnessed acv 2:1 on the DOM to CREATE breakouts with the trend at major s/r levels, and I have seen this same acv ratio happen to hold price at major s/r levels to then reverse the trend...it is very interesting to watch.