Advanced concepts

Section 01

More than one time frame

A practical hierarchy is: the higher time frame describes the condition, the lower time frame times the entry. Write which one wins an argument. If the weekly trend is up and the hourly trend is down, a pullback buyer and a breakdown seller are looking at true facts. Only one of them is allowed to trade under your rules. Switching to whichever chart agrees with the trade you already want is not multi-time-frame analysis. It is shopping.

Do not stack five clocks. Two are a conversation. Five are an excuse to wait until you chase.

Section 02

Gaps and the stop you do not have

A gap is a discontinuity. The market did not trade the prices in between, at least not on the venue you are looking at. Earnings, economic releases, halts, and policy shocks cause them. Your stop-market order cannot fill inside a gap that never traded. The Swiss franc study is the extreme version. Ordinary earnings gaps are the weekly version. If a single gap through your stop would violate your loss cap, the position is too big, or it should be flat before the event.

Options price an implied move into known events. You do not need the formula to use the idea: the market's option prices embed a range for the event, and that range is often wider than a normal day. Trading the event without knowing that range is volunteering to be surprised by a number professionals already wrote down.

Section 03

Correlation and heat

Portfolio heat is the sum of what you lose if every open idea hits its stop together. If the ideas are the same bet in costume, assume they will. Three long semiconductor stocks, a long semiconductor index, and a short volatility product are not five independent R. In 2008, and again in March 2020, the costume came off quickly.

Correlation measured on a calm year understates the crisis correlation. Stress is when you needed the hedge. A hedge that only works on quiet Tuesdays is a decoration. If you hedge, write what would make the hedge itself the problem, including the cost of holding it.

Section 04

Short selling, without the myth

Short interest is the open short position, reported on a lag. It is not the same number as the shares sold short during a single day. A high short interest means a crowd will have to buy if they cover. It does not mean they must cover today. Borrow can tighten, fees can rise, and a broker can refuse the locate. The loss on a short stock, with no exit, is not capped by zero the way a long stock's loss is. Define the cover before the entry. Reread the GameStop study as a lesson in crowding and plumbing, not as a pattern to hunt.

Section 05

Execution is part of the result

• Use a limit when you care more about price than about being filled. Use a market when you care more about being out than about the penny. Being out of a broken idea is usually the market-order case.
• Partial fills are normal in size. Your journal should record the average fill, not the price you aimed at.
• Moving a stop further away after entry is a new trade, larger than the one you sized. Log it as a rule break even if it then wins.
• Cancelling a target to "let it run" without a trail is also a new trade. Write the trail first if that is the style.

Section 06

The mind will edit the tape

Hersh Shefrin and Meir Statman named the disposition effect in 1985: investors tend to sell winners too soon and hold losers too long. Loss aversion, from the work of Daniel Kahneman and Amos Tversky, is the heavier pain of a loss compared with the pleasure of an equal gain. You do not need a diagnosis. You need a procedure that does not ask you to feel brave at the moment of the exit.

Revenge trading is size and frequency going up after a loss, in order to get the money back from the market, which is not a counterparty that owes you anything. Recency is trusting the last five trades more than the journal. Both are detectable on the desk if you fill it in. A followed-or-not flag that is honest will embarrass you faster than a mood will. That is the flag's job.

Section 07

Research habits that keep you employable by your own money

• Freeze the rule, including the stop and the cost assumption, before you look at the score.
• Keep a record of the ideas you threw away. Testing twenty variants and publishing the winner is not a test. David Aronson's Evidence-Based Technical Analysis is the stern version of this paragraph.
• Prefer a result that still exists after you worsen the fills and shift the dates.
• Do not optimize a length to the exact bar. If 20 works and 19 and 21 are disasters, you found a coincidence.

Section 08

Methods this manual will not teach as systems

Elliott Wave sorts swings into nested impulses and corrections, often with Fibonacci ratios attached. Live, two competent counters often disagree about which wave is underway, and the count is easy to repair after the move. That is too much freedom for a rule you claim to test. Harmonic patterns with animal names stack several precise ratios. The arithmetic is exact and the placement is flexible, which is how any chart grows a bat if you need one. Marketing pages attach win rates to these names. Treat those rates as unproven unless the definition was frozen, the sample includes the failures, and the costs are in.

Wyckoff's language of accumulation and distribution is a way to talk about supply and demand. The phases are easy to label in hindsight and hard to specify in advance. If you use the language, translate it into marks a stranger could repeat. If you cannot, keep it as commentary and do not size it.

Section 09

Standing aside

No-trade conditions belong on the same page as the setup. A short list that earns its keep: you are angry, you are tired, the event is inside your hold time and you have no gap rule, the spread is wider than the target, you cannot state where the idea is wrong, or you already have the heat budget spent. Flat is not a failure of imagination. It is a position with a known result of zero, which beats an unknown result you sized from boredom.

When you do trade, the standard is dull. Follow the rule. Record the R. Review on a schedule, not after each tick. Change the system only when the kill switch or the research rule says so. WorldBase can give you the language and the arithmetic. It cannot feel the loss for you, and it should not tell you the next trade.