I used 10 Levels of the Bid and Ask - not sure if it would be good on just 4 Levels - but if it worked for you - great. I also watched the Inside Bid and Ask - to see if it was larger - on the side I thought the market would move too. Meaning - if the Inside Bid was a lot bigger than the Inside Ask - plus the Ratio was indicating the Bid - the Market would go DOWN. So - if the Inside Ask was a lot bigger than the Inside Bid -plus the Ratio was indicating the Ask - the Market would go UP. And - the 1.5 to 1 Ratio - was the absolute MINIMUM - I prefered around 2 to 1. The larger the Ratio - the better.
Four levels beyond the inside market, so five levels deep. At that time, market depth for Globex futures was only five levels deep. When the market was moving, the cumulative ratio rarely went as high as 2. Did it work for me? No. I did the work. It was a tool I used to help me carve the turns. It was just one tool among several I used for that purpose, which was to capture an additional two ticks off each turn. When it became less effective for me because of hidden orders and synthetic hidden orders from algos, I ditched it without any lingering attachment. I also monitored the inside market as well and T&S. These are what I call “fine” tools. I also have “coarse” tools and “medium” tools. Over the years I’ve simplified the monitoring to using coarse tools only. Carving the turns now is more guesstimating than anything. When I see a turn arriving I have a limit order in advance ahead of the market. It’s like hunting a rabbit. I have to aim ahead of where I see the rabbit to where I think the critter will be. I find this just as effective, if not more effective, as watching the DOM used to be. I don’t scalp for ticks. A couple of ticks on a trade is just break-even for me. I call that a wash trade. Nowadays a wash trade is a couple of points. This is an extraordinary period for stock index futures.
If your system does not work in volatile days your system is weak, obviously. I threw away a dozen system as such. It is easy to get a system that works in only certain types of days. You need a better system that can handle "normal" days and volatile days. Consider isolating a core system and isolating parts that can change for volatile days, e.g. when VIX is over 25. NB that is the hard part about system development. Knowing which parts are positives and which are happen chance. That is why one needs a testing harness to do system development, imo. Most assume the system can be developed and use the market as a test bed, AND then think they have valid results. My system test harness is larger than my actual system. It has to take into account real-time, VARIABLE conditions, all the while remaining simple to execute and manage. Does that make sense? Hope that helps
Couldn't you just record a short video / audio and do the same and post it on YouTube or show your screen live on Youtube live, Facebook live, Twitch or whatever other live stream is popular these days. We've had other members do what you're doing and they only just posted a link in their trade journal to their live stream here at ET or just posted a message post with the link. Yet, the live stream only lasted a few weeks or so although a few live streamed a few months. I would see about 5 - 10 other user names viewing the live stream of trading / conversation but I don't know how many were from Elitetrader.com Strangely, never seen anyone continue doing such beyond a few months even for those trading profitably. wrbtrader
I think a trader that designs one system can just as easily design a few different systems in which each works in different types of trading conditions instead of trying to design one system to work in all types of market conditions as in one size fits all. Thus, one system for what you refer to as "normal days" and another system for "volatile days". Simply, on volatile trading days...the other system for "normal trading days" shouldn't be giving any trade signals or far below what it normally would have signaled. Just the same, I've seen system designers change their systems whenever market conditions change although that would then imply the trader is just as important because the trader will then need to understand the price action to know when it's time to change / switch to another trading system. Note: I'm referring to automated system traders. wrbtrader
I appreciate your response. I guess I could to the live streaming - but that would mean I have to learn something new - which I don't care to do - and it would be like "work" - and I don't do "work". It may be easy - but it is not for me. If someone wants my advice - wants to see what my charts in Real Time - and it doesn't take up too much of my time - I would be glad to bring them into Zoom or Skype (preferably). In my early trading days - trading was hard. When I first started trading - I thought it would be easy. I took a lot of classes - spent many hours trying just about every indicator I could find - and the results was - I kind of sucked at trading. While I did make money - I screwed up a lot - almost had to "hang my head in shame" - as a failed Day Trader - and go back to work. I have several friends in ET and we do talk now and then - in Skype. A few of them wanted me to show some of my Algo charts - and I did - and I kind of got carried away. For many years - I didn't participate in ET - as I just didn't have the time or desire. While I still love daytrading - I don't trade as much now because I don't need too - I make money off my clients who trade large accounts. The main reason I do still trade - is - and I am serious - it keeps my old brain active - and it is fun as hell. Plus - I love to code and that is good and bad - if I have an trading idea - I have to code the damn thing - before I forget. So - to end this long response - I have time to participate on ET - don't know for how long - but I am enjoying it. If what I say and show - helps some trader - great.
I agree. I guess it comes down to the details of how much is the "system" right? Some people consider a set of indicators a system, others include the trade executions, and still others include short term risk management, and others hedging while the trade progresses. Likewise, the criterion for "it works", can be very simple or incomplete. I think the OP was in the "incomplete" camp. I guess the takeaway here is modular parts always helps. Not only to be able to remove or add parts, but also to reuse or adjust. Orthogonality. Also, I would opine, the other take away is, how do you know you have a complete system or at least under what conditions does the system FAIL? If you don't know when it fails, one cannot not say they understand the system. Knowing when it works is NOT ENOUGH. You have to also know when it DOES NOT work. This might sound obvious, but I think many people do not test the limits of their system, but stop after they find where it does work. PS: Finding the system limits took me 2x longer than finding the conditions where it does work.