Three-line break as a one-parameter mechanical filter

Fewer lines than periods

A three-line break chart adds a line only when that period’s move meets a mode-dependent test. Prices that stay inside the recent-line channel are omitted, so the chart has fewer lines than periods. In typical examples, about two-thirds of high-low-close bars collapse to one line.

After a first-period basis line, each later line runs from the prior line’s price to the current close. Three same-direction lines define trending mode, and a reverse line is drawn only when the close exceeds the high or low of those three lines.

The construction uses a single price per period, so the same rules can be applied to series that lack high-low-close or intradaily data.

One reversal count sets the mode

The only first-pass parameter is the reversal line count, commonly three. That count is simpler to set than a point-and-figure box size. Three-line and four-line versions of the same yen series stay identical over large segments, so the trend reading often does not change with the count.

Entry, exit, and abstention from mode

A mechanical trading system is a complete mapping from observed market state into entry, exit, and abstention without a discretionary override. Trend following participates only while a defined trend state is in force and liquidates, reverses, or stands aside when that state fails.

The elementary mechanical reading is to enter when upward trending mode first appears, liquidate or reverse when mode reverses, and stand aside while direction is mixed.

Sequence quality on the historical examples

On the cash-index example, 34 signals appeared and 28 were runs of at least three same-direction lines. Isolated one- or two-line interruptions were treated as false steps that the rules would unwind on the next valid line.

Five continuously linked futures series from 1992 through 1996 produced 160 apparent reversals, of which 40 were not three-or-more-line sequences. On the three-line yen chart, eight of 28 signals reversed after only a short contrary run.

Confirmation lines and market screens

Two evaluation adjustments were compared: wait for a second same-direction confirmation line, or raise the reversal hurdle to four lines. The four-line yen chart counted 20 of 26 signals as accurate over five years, at the cost of later participation.

A market was treated as poorly matched when successive signals failed to capture the move or reversed too often. Among a few dozen futures series, lumber, coffee, and Nikkei were judged to give the most usable signals, while the charts were used as trend confirmation on nearly every series examined.

Time out of the market

The historical comparison omitted transaction charges and treated reduced time in the market as part of the evaluation. It reported long-side exposure of 50.7 percent over 430 days for lumber from July 1993 to December 1996 and 45.0 percent over 337 days for Nikkei from January 1992 to January 1994.