The Same Reading, Recounted Every Single Day

Core Idea

Two windows, both counted backward from the most recent daily candle:

The last 20 daily candles. Roughly a trading month of delivery, but a month that ends today rather than on a calendar boundary.

The last 3 daily candles. The shortest useful window there is, and the one an intraday reaction is most often actually respecting.

For each: mark the high, mark the low, mark the equilibrium halfway between. Above equilibrium is premium, below is discount. Nothing new in the construction — what changes is that both windows move forward with every new day, so what was premium on Monday can be discount by Thursday without price having done anything dramatic.

Why It Exists

A fixed anchor has one weakness: late in the month, the range it frames can be a long way behind what price is currently doing. The rolling windows close that gap. They are not a replacement for the anchored 20/40/60 — they are the short-term half of the same reading.

The 3-day window in particular answers a question the longer ranges cannot: of the last three sessions, are we at the top of what price has just done, or the bottom of it? That is frequently the level an intraday move turns at, and it is invisible on a monthly range.

How Professionals Read It

Read the two against each other, and then against the anchored ranges.

Both in discount is the cleanest version of a rolling read: cheap against the last month and cheap against the last three days.

Discount of the 20, premium of the 3 is the common and more interesting one. Price is cheap against the month but has just rallied hard off a low — you are buying into a short-term extreme inside a longer-term discount. That is a timing problem, not a direction problem, and it is exactly what the 3-day window is for.

Premium of both, while the anchored 60-day range says discount is the mirror: expensive against everything recent, still cheap against the quarter.

Practical Application

Mark which PD arrays sit in each window's discount and which sit in its premium, the same as with the anchored ranges. On a 20-day window that list is short and current, which makes it the one you actually work from intraday.

Re-count at the start of every session. That is the whole point of a rolling window, and the only maintenance it needs.

Common Mistakes

• Treating them as a replacement for the anchored ranges. They are the short end of the same reading. Dropping the monthly anchor throws away the context that decides whether the short-term read matters at all.
• Forgetting that they move. A level drawn from the last 20 candles is wrong tomorrow. Levels from the anchored ranges survive the month; these do not survive the day.
• Reading the 3-day window on its own. Three candles is a small sample, and on its own it will have you buying every short-term low regardless of where that sits in the month.
• Counting calendar days. Weekends produce no daily candle. Twenty candles is twenty candles.