Selection & how to operateThe obvious part if you've understood all the previous posts.
It's easier to start with how Not to trade .
Wrong - cherry picking "strong" levels. Every level is a level, not better & not worse than another one. Choosing the supposedly strong levels is a subjective thing that reduces expected value & consistency.
Right - operating at each level on a given resolution, you either expect a level to repel prices or to be consumed, you operate accordingly at every level. The more you operate, better for the market, higher your revenues. If there too many levels for you, instead of cherry picking you just move to a lower resolution. Some levels can be effectively skipped because of risk & sizing consideration, but skipping levels an cherry picking levels are 2 completely different mindsets.
Wrong - stopping operation after N loosing trades.
Right - controlling equity as explained in "Sizing & how to manage risk". If you're making loosing trades in a row, you don't stop, you just hit zero size, then you imagine trades or execute on simulator, when your size comes back to a non-zero value you come back to the real account. More you operate - better for the business.
Wrong - waiting for a "confirmation". If you don't have a firm expectation whether a level will repel prices or will be consumed, you don't know what you're doing, read all the posts and understand how it all works.
Right - knowing in advance what you gonna do at each level & keep reevaluating it in real time.
Wrong - making reentries. The activity around levels, especially how levels get cleared, is very well defined. After the scaling in is complete, you either exit at loss/at breakeven when a level gets cleared / positioned in the unexpected side. Or, you scale out while being in the money.
Right - unless there was a mistake caused by a misclick or smth like dat, reentries is an irrelevant concept.
Wrong - working out insurance after the entry.
Right - a hedge should be bought BEFORE scaling in, same goes about placing the stop-losses.
How to operate
Asset selection
Not many people think about it, but it makes sense not only to provide liquidity when & where there's not much of it, but also to consume excessive liquidity when & where there's too much of it, because both cases are unhealthy for the markets. So, we have 2 types of trading instruments then:
1) overquoted ones, such as GE, ZN, or ES many years ago;
2) underquoted ones, such as CL, NQ;
How to distinguish dem?
One way is to take a look at volumes on highest resolution cluster/footprint chart, and compare em with the actual number of bid/asks in the DOM. ZN for example is hugely overquoted, you'll notice that: it has aprox 1000 contract at every bid/ask price, but when these limit orders start to get consumed at one price, the rest orders at the same price just gets cancelled, and you see lesser values on your footprint/cluster chart. The opposite happens on underquoted instruments, they need liquidity.
Why it matters?
You operate the same way on both under and overquoted vehicles, but:
1) on underquoted vehicles you mainly use limit orders, you provide liquidity;
2) on overquoted vehicles you mainly use market orders, you remove liquidity;
Exits at loss vs attempting to get out around breakeven
Both are legit, the latter gives more freedom, but implies not using stop-losses so you have to know 4 sure what's happening and what you're doing.
That's how you trade with stoplosses.
1) In case of trading pops from positioned levels, you simply exit when the support/resistance gets cleared, in case of clearing by price it means you'll have an L, no big deal tho;
2) In case of trading pushes through positioned levels (aka trading clearings aka trading consumptions), same, you're getting an L if you hit the invalidation point. The invalidation point for these trades is the opposite border of the positioning sequence. This border is found the same ways as the front level, just at the opposite side;
3) Trading during a positioning itself. Makes least sense to trade with stop-losses, but in theory: taking an L at the next level past the level you expect to be positioned this or that way. If there is no level past you current level, you try to make a projection, smth like its shown on ZN chart of this post, imagine you were trading positioning of 112'19.
Without stops it's almost the same, it's just instead of taking an immediate loss after an invalidation event, you exit at breakeven when price comes back to the entry zone (in most cases it does). If prices don't go back and hit another level, you simply continue trading there, if that new level you're working with now is supposed to act in the opposite direction from the previous one, you simply reverse your position. If that new level is supposed to work in the same direction as the previous one, you're holding your position further.
This kind of operation assumes very high win rate, low RR ratio and very rare but significant losses. However, if the unexpected happens 2 times in row, chances are the problem is on your side xD
Finally
1) Monitor non-market data in order not to be caught against the momentum surges (eg unless you're a DMM, trading at Jobless Claims release is a BAD IDEA);
2) Pick your main resolution that way you'll be satisfied with the frequency of your operations;
3) Work with all the levels there;
4) Never approach the next level while having a full position, always offload risk on the way, unless you expect the next level to be cleared/positioned in the same direction;
5) Always control risks;
6) Understand that it's all about doing the right thing, and it's totally possible to understand what is right by gaining all the info from all the data.
You should end up trading 100% of positioned levels, trading 50% of positioning processes demselves, and rofl never try to trade smth that looks like "a new level is forming now".
