Toby, two final construction details surfaced while translating the Effort versus Result formula into an indicator.
The normalized range/volume quotient is necessarily positive, but the plotted series is signed (+2.21, −0.58, −2.80). What determines the sign—close versus open, close versus the previous close, or another directional measure?
And is each 10-day benchmark calculated from the ten preceding completed periods, excluding the observation being evaluated, or does the rolling average include the current observation?
Toby, Chapter 1’s distinction between concept formation and a “floating abstraction” gives me a clearer way to frame a question I raised under your ORB article.
In the 1989 book, ORB appears operationally in two forms: movement beyond the opening range—the first thirty seconds—in the glossary, and fixed displacement from the opening price in several study tables. The recent ORB article adds a third form: displacement normalized to an n-day average range.
What characteristic do you regard as essential and invariant across these implementations? Are they instances of one higher-order concept—momentum expressed as price displacement from a chosen reference—or distinct concepts that should remain operationally separate? Put another way: which measurements can be omitted without disconnecting the ORB concept from the observations that give it meaning?
Two construction details would also help make the examples exactly reproducible. In the ORB baseline, does “0.80% of the 10-day average range” literally mean 0.008 × the average range, or was 0.80 × the average range—80%—intended? And in the Effort versus Result calculation, what lookback or reference period defines the benchmark for relative range and relative volume, and is the same window used for both?
The three forms aren't equally essential. The thirty-second glossary version and the fixed-displacement version from the '89 book were both arbitrary in the same way: a fixed window and a fixed amount don't adapt to what a market is actually doing. A move that's significant in a quiet market may be noise in a volatile one. Normalizing displacement to the 10-day average range is what makes the measurement mean the same thing across markets and across decades. That's the part I consider essential and invariant: price displacement from a reference point, scaled to what's normal for that market right now. The specific window, thirty seconds, a fixed number of points, can be dropped entirely without losing the concept. The scaling can't be dropped without losing it.
To your two construction questions. Good catch on the 0.80. It's a multiplier, not a percent: 0.80 times the 10-day average range, or 80% of it. The article had it printed as 0.80%, which is off. I've corrected it in the piece. And on Effort vs Result, yes, same window for both sides of the ratio: relative range and relative volume are each measured against their own 10-day average, then range is divided by volume. One lookback, used twice.
Thank you, Toby — that settles it cleanly. “The scaling can’t be dropped without losing it” is the sharpest one-line definition of ORB I’ve seen, and it explains why the fixed-point implementations aged while the concept didn’t. I appreciate the correction on the 0.80; with the same 10-day lookback applied separately to range and volume, the Effort vs Result construction is now fully reproducible.
Toby, two final construction details surfaced while translating the Effort versus Result formula into an indicator.
The normalized range/volume quotient is necessarily positive, but the plotted series is signed (+2.21, −0.58, −2.80). What determines the sign—close versus open, close versus the previous close, or another directional measure?
And is each 10-day benchmark calculated from the ten preceding completed periods, excluding the observation being evaluated, or does the rolling average include the current observation?
Toby, Chapter 1’s distinction between concept formation and a “floating abstraction” gives me a clearer way to frame a question I raised under your ORB article.
In the 1989 book, ORB appears operationally in two forms: movement beyond the opening range—the first thirty seconds—in the glossary, and fixed displacement from the opening price in several study tables. The recent ORB article adds a third form: displacement normalized to an n-day average range.
What characteristic do you regard as essential and invariant across these implementations? Are they instances of one higher-order concept—momentum expressed as price displacement from a chosen reference—or distinct concepts that should remain operationally separate? Put another way: which measurements can be omitted without disconnecting the ORB concept from the observations that give it meaning?
Two construction details would also help make the examples exactly reproducible. In the ORB baseline, does “0.80% of the 10-day average range” literally mean 0.008 × the average range, or was 0.80 × the average range—80%—intended? And in the Effort versus Result calculation, what lookback or reference period defines the benchmark for relative range and relative volume, and is the same window used for both?
Feliks, thanks for the question.
The three forms aren't equally essential. The thirty-second glossary version and the fixed-displacement version from the '89 book were both arbitrary in the same way: a fixed window and a fixed amount don't adapt to what a market is actually doing. A move that's significant in a quiet market may be noise in a volatile one. Normalizing displacement to the 10-day average range is what makes the measurement mean the same thing across markets and across decades. That's the part I consider essential and invariant: price displacement from a reference point, scaled to what's normal for that market right now. The specific window, thirty seconds, a fixed number of points, can be dropped entirely without losing the concept. The scaling can't be dropped without losing it.
To your two construction questions. Good catch on the 0.80. It's a multiplier, not a percent: 0.80 times the 10-day average range, or 80% of it. The article had it printed as 0.80%, which is off. I've corrected it in the piece. And on Effort vs Result, yes, same window for both sides of the ratio: relative range and relative volume are each measured against their own 10-day average, then range is divided by volume. One lookback, used twice.
Thank you, Toby — that settles it cleanly. “The scaling can’t be dropped without losing it” is the sharpest one-line definition of ORB I’ve seen, and it explains why the fixed-point implementations aged while the concept didn’t. I appreciate the correction on the 0.80; with the same 10-day lookback applied separately to range and volume, the Effort vs Result construction is now fully reproducible.