Showing posts with label Short-Term Trading. Show all posts
Showing posts with label Short-Term Trading. Show all posts

Friday, September 25, 2026

George Cole and the Discovery of Pivot Points for Day Trading | Toby Crabel

The daily support and resistance lattice that most day traders treat as folk knowledge was first set down in 1936 by George William Cole (1870–1937), though he rarely gets credit for it. Cole was the first author to publish the formula now called Classic or Floor Pivots. Toby Crabel himself uses these pivot levels—not as a standalone system, but as one layer in a larger map of reference points.

"Before the many books and derivations of pit trader's numbers, there was George Cole in 1936. He was the first author to write about the formula so many day traders use now, most of whom have no idea where it originated. I haven't seen any contemporary authors credit Cole for it. Fair enough, the method works well, and the numbers are good reference points for structure in the market. It’s worth knowing where they sit each day in whatever you're trading."

 Cole self-published Graphs and Their Application to Speculation (a sequel to his 1928
book Successful Speculation: A Business) the year before he died, and Donald Mack
reprinted it
in 1998
in the Financial Times / Pitman Traders' Masterclass series.

Cole was not creating a simple day-trading cheat sheet. He wanted speculation to operate like a profession: charts as a visual map of mass psychology, a "law of occurrence or recurrence" in commodity prices, and human judgment required to pull the facts together. That is 1930s technical analysis in the Wyckoff family—slow, pictorial, and commodity-first. On subsequent literature, Crabel is blunt, and he names names:

"Almost all short-term traders have explored these numbers. Larry Williams called them his own. John Hill, Fisher, and Ochoa each added something to them. Carter and Person use them too. I saw traders on the floor in the '80s carrying their "Green Sheets" into the pit, and these numbers were the dominant feature on them. It’s safe to say most traders know about them and have built systems around them, so it pays to know where they sit if you want a read on the market’s mind throughout the day."⁠

These details matter. By the 1980s, the formula was no longer just a book idea; it had become pit infrastructure.

»
 Huh! that fellow is a 'chart trader.'
« 
 
Two Different "Pivots," Often Smashed Together
► Structural and Swing Pivots: A high with lower highs on both sides; a low with higher lows on both sides. Livermore called turning points "pivotal points" and split them into reversal versus continuation. Larry Williams later said he first called those short-term turns "ringed" highs and lows "in deference to the work done in the 1930s by Henry Wheeler Chase." That is swing structure, not a closed-form projection.
► Calculated Floor Levels: Yesterday's high, low, and close, printed into today's map before the open.
 

This is what Crabel means by Cole numbers (15-minute E-mini NASDAQ-100 of January 29, 2024): "R4 to S4 including the pivot (p), are all Cole numbers. I-1 hi and I-1 lo are yesterday's high and low. The two-day high (2 day hi), the all-time high (ath), and swing low (1 day sw lo) are also an important part of market structure."⁠ 

John Person's lineage credits Chase with the formula and Williams with its 1979 revival. The honest history involves two 1930s names, an unread reprint, a popularizer, and a floor practice that was already regarded as "secret numbers" when Person walked onto the CBOT. Crabel is likely right that Cole printed the arithmetic first. Person is likely right that the pits treated it as inherited craft. Neither invented the market’s habit of defending yesterday’s range.

The Formula Without Mysticism 
   Let H, L, C be the prior session's high, low, and close.
 
R4 and S4 are further range multiples, which Crabel plots. PP is the typical price (High + Low + Close / 3). R1 and S1 reflect the opposite extreme through that typical price, while R2 and S2 add or subtract the full prior range. Variants exist simply because people keep reweighting those same three numbers: Woodie doubles the close; Fibonacci stacks 0.382/0.618/1.00 of the range off the Pivot Point; Camarilla builds tight fade and breakout rails off the close; DeMark flips the calculation depending on whether the prior bar closed above or below its open. None of this is new physics. It is a daily map printed from a finished bar. 
 
How Crabel Uses the Numbers
He does not treat Cole levels as a system in themselves. He views them as a "predetermined" framework that sits next to a "dynamic" one:

"These are all predetermined price levels. Once the market opens, there are dynamic reference points to factor in too, and I cover those elsewhere. [...] Price will often poke through a high or low before resuming trend, so the action around the previous day's high and low matters."⁠

That last sentence is the core operational rule. The Cole grid provides the scaffolding; the prior high and low are the live walls. A poke-and-fail through yesterday's extreme, in Crabel's framing, tells you far more than a simple tap of R2.

The dynamic half of the map is the work Crabel is known for. In the 1988 Stocks & Commodities series that became his 1990 book, he defined the open itself as the other reference point of the day:

"Opening range breakout is one of the most important indicators of daily market direction that a trader can utilize. An opening range breakout (ORB) is a trade taken at a predetermined amount above or below the opening range. When the predetermined amount (the "stretch") is computed, a buy stop is placed that amount above the high of the opening range and a sell stop is placed the same amount below the low of the opening range. The first stop that is traded is the position and the other stop is a protective stop."⁠

He was already distinguishing rare trend days from ordinary rotation:

"Early entry is defined as a large price movement in one direction within the first five minutes after the open of the daily session. A study of early entry is essentially a study of price action, and the type of price action that takes place on early entry shows that participants are urgent about entering the market. It is a distinct recognition of either a profitable or dangerous situation. [...] It should be noted that directional moves of this nature are relatively rare and may occur only 10% of the time. Most days (70% to 80%), prices exhibit rotation or choppy action and the first five to 10 minutes of trading are sluggish and directionless without a clear movement away from the opening range.⁠"

The hinge between those two day-types is the principle that still sits under NR4, NR7, inside days, and two-bar and three-bar narrow range:

"The market having a specific nature is constantly changing from a period of movement to a period of rest and back to a period of movement."

 
That is the Principle of Contraction/Expansion. Cole numbers do not tell you which regime you are in; compression plus a move off the open does. Crabel reduced that entire visual tradition to two forces:

"My observations of markets through visual displays of data have led me to a simple conclusion: there are two primary forces at work. One is momentum, which includes the opening range breakout (ORB). The other is mean reversion, which at times can even involve trading in the opposite direction of the ORB. This has always been a useful way to think about markets. But over time, I have come to appreciate that there are many nuances and additional conceptual frameworks that continue to refine this view."⁠

Cole's grid is useful in both regimes. On a rotation day, it marks the likely fade rails. On a momentum day, it marks where the move should pause or accelerate. It does not choose the regime for you.

 
What Changed: The Open Stopped Being the Open
The Cole numbers survived the death of the pit because they do not depend on a clean open. ORB did. Crabel has been explicit about that at length, and the long version is the right one:

"⁠Where's the open? For God's sake. I mean there's so much volume in the 24-hour sessions it's impossible to determine what the open is. So the wonderful thing about open range breakout—back in the day when there were just primary session, domestic session trading—was it was the most vital piece of information, the reference point that you could have. [...] The real problem is where is the reference point? Where the open was a great reference point, but now what other reference points are there in the markets? One is the close of the previous day and the movement off of that."⁠⁠

He has also been careful not to claim invention of the open as a tool—"I didn't invent it"—and to note that Larry Williams was already working the same ground, and was not pleased to see it in print.
 
S&P 500 vs. Daily, Weekly, Monthly, and Quarterly Pivot Levels.
 
What Crabel does claim is the research program: find where order flow concentrates, then trade the imbalance. In the early 1990s, that meant two markers. Now it means a crowd of them:

"⁠In the early nineties I used ORB as a primary, or previous day’s high or low, as a reference point for marking and entering trades. Now we probably have 10 or 15 or maybe even 20 different reference points that exist in any market and different ways of navigating that. [...] The closing of the previous day tends to be much more important now than it ever was.⁠"⁠

Reading the Cole essay against that interview and the layout of Crabel's January 29, 2024 chart, the takeaway becomes obvious. R4 and S4 are the old predetermined lattice. The prior high and low, two-day high, all-time high, and swing low are the reference points that still carry energy now that the open is no longer a single moment.

 
What is Solid, What is Soft
► Solid: The levels are objective and available before the session begins. They establish a sensible bias rule: lean long above PP, lean short below PP, and treat R1 and S1 as the first places the auction should hesitate. Prior-day highs and lows are usually respected more than outer projections because they are actual traded extremes rather than mathematical reflections. Crabel's poke-through-and-fail observation around those extremes remains one of the cleanest intraday tells in trading literature. The green sheets existed because the numbers were shared—and shared levels become structural market points even after their original rationale is forgotten.
►
Soft: The formula does not account for overnight gaps or a Sunday FX open. "Yesterday" is no longer a clean object in 24-hour markets. Outer levels are often wallpaper. Buying S1 and selling R1 as a standalone system is how the method earns its bad name. The edge, when there is one, is confluence: a Cole level + prior high and low + opening range or prior close + the actual swing right in front of you.

That is also why Crabel's credit-where-due note is more than antiquarianism. Cole printed a portable map. The pits made it a common language. Williams, Hill, Fisher, Ochoa, Carter, and Person turned it into product. Camarilla, Woodie, Fibonacci, and DeMark are variants. Crabel's own contribution is the frame around the map: predetermined levels first, then the day’s dynamic reference points, then a decision about whether the session is momentum or mean reversion.

Ochoa's CPR, Cole's Floor/Classic or Traditional, Woodie, DeMark, Fibonacci, Scott's Camarilla.

However, do not give the pivots' arithmetic more metaphysics than it earns. It is a prior-day typical price and a set of reflections. It works when the market is still negotiating yesterday's range. It is noise when the market has already decided today is a different day. The skill is telling those two conditions apart—and that skill, as Toby Crabel keeps repeating, is not in the formula. It is in the structure you put around it.

Reference:
 
Why pivot points work?
Self-fulfilling prophecy.
 Aha!
 
See also:

Friday, September 4, 2026

S&P 500 vs. True & Mean Lunar Nodal Speed | September to December 2026

Financial markets correlate closely with the 4–14-day cycle of the True Lunar Node (North Node/Rahu) as it moves through its retrograde, stationary, and direct phases. This cycle is best illustrated by charting the True Lunar Nodal Speed against, for example, the S&P 500, where "speed" refers to the node's geocentric motion, measured in degrees of longitude per day.
 
Last station: Sep 04 (Fri) 03:57, Retrograde (Rx) → Direct. Next: Sep 07 (Mon) 03:12, local maximum
(+0.0190°/day), then Sep 09 (Wed) 15:18, Direct → Rx. True = Mean on Sep 13 (Sun) 07:35 ( EDT).
 
The True Node is predominantly retrograde, with a negative speed averaging −0.053°/day. It regularly slows, stations (speed = 0°/day), and briefly turns direct, reaching speeds of up to +0.0015°/day, before resuming its retrograde motion. These stationary periods are most pronounced and prolonged near eclipse seasons, which occur roughly every 173 days, when solar perturbations of the lunar orbit are strongest.

Around eclipses, the lunar nodes can shift rapidly between direct, retrograde, and near-
stationary motion, coinciding with sentiment extremes and elevated market volatility.

Expect potential short-term changes in trend when the True Lunar Nodal Speed (blue solid line in the charts above):

► changes direction, shifting from retrograde to direct and back (speed = 0°/day); or
► reaches a maximum or minimum extreme
► equals Mean Nodal Speed 
 
Depending on the season, most True-Node speed swings last 7–8 days (69%).
 
Retrograde (Rx) → Direct repeats every ~13.5 days (6.7–15.8). 
 
Direct is a short poke (~3.6 d); retrograde is the body (~9.9 d); together they make the ~13.5-day station clock. 
 
2026 Q1 — Daily
 OHLC bars of the S&P 500 vs. True and Mean Lunar Nodal Speed;
 annular solar eclipse Feb 17, total lunar eclipse Mar 03.
 
2026 Q1 — 4-hour OHLC S&P 500 vs. True and Mean Lunar Nodal Speed. 
Blue wave is True Node speed on price scale; dashed blue is speed = 0 (direct above, retrograde below); red dashed is mean speed. Blue dates are stations/extrema in ET. Red vertical dates are True speed = Mean speed. Gold = solar, purple = lunar eclipse. 
 
 Q1 2026: True Lunar Nodal Speed at 0°/Day, at Extremes, and Equal to Mean Nodal Speed.
 
However, neither the daily nor the 4-hour chart follows the node. They may answer it, late or early by a bar or two on the 4-hour chart, and sometimes not at all. There hardly is any 1-to-1 correlation. What the speed curve is good for is identifying the hours when a crowd's appetite changes. A station (speed through 0°) is the useful alert: the True Node has stopped adding longitude or stopped subtracting it. That pause often appears as a change of character in the next one to three 4-hour bars—wider range, a failed break, or the first close that refuses the prior drift—rather than as a guaranteed reversal print at the exact minute.

Extrema serve a different purpose. A deep retrograde minimum represents a maximum rate of withdrawal; in Q1, the January 16 and January 30 lows, and in Q3, the July 12 and July 26 lows, fell within or just behind washout clusters already underway. The signal is not "buy the timestamp." It is "the selling has reached a dated climax window; now wait for the 4-hour structure to stop making lower lows after that window." The brief direct maxima are weaker. They last only a few days and often mark nothing more than a pause within an existing drift. Trading them as tops is therefore unreliable; in this sample, they more often mark a pause in the existing drift than a reversal of it.  
  
2026 Q2
— Daily OHLC S&P 500 vs. True and Mean Lunar Nodal Speed.
 
2026 Q2 — 4-hour OHLC S&P 500 vs. True and Mean Lunar Nodal Speed. 


 Q2 2026: True Lunar Nodal Speed at 0°/Day, at Extremes, and Equal to Mean Nodal Speed.
 
True nodal speed equal to mean speed is mostly a mid-wave event, not a turn. Red dates on these charts sit on the slope, not at the crest. Use them as confirmation that the osculating node has fallen back in step with the secular drift, which on the 4-hour chart tends to coincide with a continuation bar rather than a reversal. The April 2026 stretch is the clearest illustration of slope over event: from the April 4 minimum through the April 10–11 station and maximum, the 4-hour market rose with the wave. May 30 is the counterexample: a deep minimum against a market that only chopped at the highs. A minimum that does not meet a 4-hour breakdown is not a trade.

Delay is the practical rule. Weekend stations (e.g. March 21, May 30, June 14) spend their first reaction in Sunday–Monday futures and only then in the cash session; do not treat the Saturday timestamp as a Monday open. Intraday, the usable lag is one to six 4-hour bars. July 31 into August 2 is the best sequence in the file: station, then maximum, while the 4-hour S&P left the July hole and ran. September 4, 03:57 EDT, retrograde to direct, presents the opposite texture—an event at the local high, followed by a fade in the later bars of the same day. The same class of event, two market answers. The distinction is the 4-hour structure already in place when the node arrives.
 
 
2026 Q3 
— Daily OHLC S&P 500 vs. True and Mean Lunar Nodal Speed;
total solar eclipse Aug 12, partial lunar eclipse Aug 28; last S&P 500 bar Sep 04.
 
2026 Q3 — 4-hour OHLC S&P 500 vs. True and Mean Lunar Nodal Speed. 
 
  
Q3 2026: True Lunar Nodal Speed at 0°/Day, at Extremes, and Equal to Mean Nodal Speed.
  
Used as a short-term timing method, then: ignore lobes shorter than six days; treat the seven-to-eight-day swing as the holding window; take stations as tempo alerts and extrema as climax windows; and require the next 4-hour bar to confirm. 
 
2026 Q4 — Nov 09 16:45 is the deepest Rx of the year (−0.241°/day)—furthest from the mean.
 
 
 Q4 2026: True Lunar Nodal Speed at 0°/Day, at Extremes, and Equal to Mean Nodal Speed.
(Date and Time calculated for New York City, ET).
 
And, as always, what is being practiced here is attention and precision, not prediction: any market still has to print a reversal in its own highs and lows. Used this way, the inflection points and crossings in the True and Mean Lunar Nodal Speed curves may serve as a reminder to pay closer attention during the upcoming nodal events listed above. But for a short-term trader simply looking to make money in the markets, better to ignore them altogether and focus on market structure, price action, and risk management instead  (Williams, Unger, Huddleston, Burke).
 
See also: 

Saturday, August 22, 2026

S&P 500 vs. Ap Index: +3-Day Lag and Limits of Multi-Week Forecasting

The chart below illustrates the hypothesis that geomagnetic activity, measured by the planetary Ap index, precedes trend reversals, as geomagnetic disturbances subtly impair collective mood and increase risk aversion. This idea draws on research examining correlations between space weather and financial markets, including evidence of both direct and inverse relationships between Ap—and related Kp and F10.7—readings and subsequent market performance.

S&P 500 vs. Ap Index (Apr-Oct 2026). Projected Ap peaks:
Sep 4 (Fri),  Sep 17–20 (Thu-Sun), Oct 1 (Thu). 
Wait 24–35 or so days and the AP–price correlation will look almost perfect again.
 
Chart Construction and Data Sources
The chart overlays the daily S&P 500 with the Ap index shifted forward by three calendar days—the short lag that currently offers the best balance between the classic weekly effect reported in the literature and practical S&P 500 trading-day alignment. The series is then extended using the NOAA 45-day Ap forecast. Historical daily Ap data are sourced from GFZ Potsdam, while the dashed forward segment represents the latest NOAA SWPC 45-day Ap forecast, issued on August 22, 2026. 
 
Limits of the NOAA 45-Day Forecast for Forward Correlation
However impressive the historical correlation may appear, its reliability as a guide to future relationships is inherently limited. NOAA's 45-day Ap forecast is a relatively low-resolution space-weather projection, it is adjusted on a daily basis, and its predictive skill declines rapidly beyond the first week. Moreover, the forecast activity levels shown in the chart are modest (Ap 8–15) and remain well below classic geomagnetic storm thresholds: Ap 8–15 corresponds roughly to Kp 2–3 (quiet to unsettled conditions), while Ap 48 corresponds to Kp 5, the threshold for a NOAA G1 geomagnetic storm. 
  
Latitude-Dependent Solar Rotation and Active-Region Return Times
Sunspots and active regions do not return to the Earth-facing side of the Sun on a fixed schedule. Because the Sun rotates differentially—faster at the equator (~25 days sidereal, or ~27 days synodic as seen from Earth) and progressively slower at higher latitudes (reaching ~30–35 days near the poles)—the time required for a given region to reappear depends on its heliographic latitude. The standard Carrington frame uses a compromise rotation period of 27.2753 days (synodic), which roughly corresponds to the typical 10–20° latitudes of sunspots. Regions at higher latitudes therefore take longer to rotate back into view, while those near the equator return sooner. 
 
Solar Activity Snapshot: Comparing Sunspot distribution on the Earth-facing and far sides of the Sun (August 22, 2026).
 
From above the Sun's north pole, its rotation is counterclockwise, carrying sunspots from left to right.
 
Reading the Raben Earthside and Farside Maps 
The Raben maps above illustrate this directly: The Earthside view shows currently visible active regions, identified by NOAA numbers and activity-color coding, while the Farside view highlights returning regions with meridian lines estimating the number of days until they may reappear, assuming a uniform rotation rate. In reality, those return times can stretch or compress with latitude. A high-latitude complex visible on the farside today, for example, may take several additional days to rotate back into Earth view compared with a low-latitude region. 
 
How Returning Regions Drive F10.7 and Ap
These returning regions influence both the 10.7 cm radio flux (F10.7) and geomagnetic activity (Ap and Kp). F10.7 serves as a direct proxy for solar EUV/UV output associated with active regions and plages; when a large active complex rotates onto the Earth-facing disk, F10.7 typically rises. Ap, by contrast, responds more indirectly: high-speed solar-wind streams from coronal holes, as well as coronal mass ejections launched from Earth-directed active regions, can disturb the magnetosphere and elevate the planetary Ap index. 
 
Construction of the 27-Day and 45-Day NOAA Forecasts
Consequently, the 27-day forecast for F10.7 and the geomagnetic Ap and Kp indices and the 45-day Ap/F10.7 forecast issued and updated daily by NOAA SWPC, are both built around the expected recurrence of these features through solar rotation. The 27-day forecast is essentially a recurrence forecast, assuming that active regions and coronal holes will reappear roughly one Carrington rotation later. The 45-day forecast extends this approach farther into the future, blending recurrence-based estimates with a longer-term background trend.
The time a Coronal Mass Ejection (CME) takes to reach Earth depends mainly on its density and solar-wind conditions:. fast CMEs (>1,000 km/s) arrive in 1–2 days, average CMEs (500–1,000 km/s) in 2–3 days, and slow CMEs (<500 km/s) in 3–5 days.
The Moon's orbit through Earth's magnetosphere, and the corresponding reduction in solar wind ion flux as it enters the magnetotail cavity near full Moon (0°), provides one example of how the solar wind–magnetosphere configuration can influence geomagnetic conditions. More broadly, the semiannual variation of geomagnetic activity is linked to the interaction between the solar wind and Earth's tilted magnetic field, which typically causes increased geomagnetic disturbances around the equinoxes and lower activity around the solstices.
Why Multi-Week Ap Forecasts Remain a Coarse Guide
That is precisely why attempts to forward correlate 27-day and 45-day Ap forecasts with the S&P 500 are inherently limited. The Sun's differential rotation, the uncertain evolution of active regions—including their growth, decay, or disappearance while on the farside—the variable geoeffectiveness of individual regions, and the chaotic nature of solar-wind–magnetosphere coupling all erode day-to-day predictability.  
 
 
Hence, multi-week Ap and F10.7 forecasts should be interpreted primarily as defining a broad solar-activity envelope rather than as precise day-by-day projections capable of supporting a tight forward correlation with daily S&P 500 returns. By contrast, short-horizon tools—such as the NOAA 3-day forecast, the LSTM-based 72 hour Ap predictor, and real-time L1 solar-wind data—retain greater predictive value for near-term market conditions.
  
See also: