Article 22 chart data and assumptions

The October decision is the close of October 22, 2026. The August decision is the close of August 5. Outcome charts end on November 16, the end of the comparison specified for this story. When the figures were retrieved on November 18, both the US daily-price archive and the list archive already contained November 17.

Download the observations

Stock Weekly at the decision Daily at the decision Daily through November 16
TriNet (TNET) Weekly CSV Decision CSV Outcome CSV
Union Pacific (UNP) Weekly CSV Decision CSV Outcome CSV
Zeta Global (ZETA) Weekly CSV Decision CSV Outcome CSV

Each CSV contains the visible chart’s open, high, low, close, volume, benchmark close and dated RS Rating. Weekly rows are labeled by the Monday of the week and show the last included trading session separately. A midweek decision uses that day’s close for the unfinished week. The Weekly price pane displays two simple averages, 10 weeks in teal and 30 weeks in orange. The purple line on the TriNet and Zeta weekly charts marks the base high before the breakout. The 52H tag measures the highest intraday price over the trailing 252 sessions, including the decision session. The percentage beside SPY measures the benchmark over the chart’s visible window. The shorter Union Pacific decision view makes the October 19 low and October 22 recovery easier to compare. The screenshots give the dates of the displayed window and keep the benchmark, price, volume and RS panes aligned.

US stock and SPY prices use the historical as-traded series. A dividend-adjusted app price can differ. RS is the contemporaneous archived rating. Blank values represent missing observations.

How we chose the examples

The US list archive starts July 7, 2026. We searched August during a Confirmed Uptrend and followed the changing market readings from October 21 through November 16. The two October setups were chosen from the October 22 report, when the headline read Uptrend Under Pressure. For breakouts we looked for a recent appearance on a master-derived setup list, a close above a prior high on at least 1.5 times the previous 50 sessions’ average volume, and a rising weekly trend. We read the preceding consolidation and its pullback lows on the chart. For pullbacks we looked for a dated Undercut & Rally or OWL VDU Pullback Pocket Pivot list appearance, an identifiable prior price level, and a recovery inside the base.

TriNet and Union Pacific were chosen after the fact because both appeared on the October 22 close, had published top-50 industry groups and offered different entry points. They had different earnings calendars, businesses and news. Zeta was chosen because its August breakout later reached the series’ partial-sale target. The three examples illustrate the choices available to a trader.

How the orders work

The three October orders are alternative uses of the same hypothetical $40,000 account. The separate August example also starts at $40,000. The usual risk ceiling is $400. A reduced or early entry uses $200. Divide the budget by the distance from the opening fill to the 8% stop, rounded to the nearest cent. Round the shares down and cap the order at available cash and 5% of prior average daily volume. A breakout opening more than 5% above its base high takes half the risk-based shares. The two modeled breakout entries opened within that limit. Union Pacific’s entry requires the next open to hold above its reclaimed low and inside the base.

A new purchase requires a published headline outside Market in Correction on the preceding close. When that preceding report is missing, the exercise skips the new order. The position then follows its own exits. A stop reached during the session fills at the stop, or at the opening price if the stock gaps through it. If a daily bar reaches both the initial stop and the 20% target, the example gives the stop priority. At the target, sell half the shares rounded down, or take the opening price if it gaps above the target. The stop on the rest moves to entry for subsequent sessions.

After the partial sale, the series also models selling the remainder at the close of the largest one-day percentage decline since the weekly advance began. For this calculation its start is the first week after the most recent decline in the 30-week average visible at the decision. A closing fill is a modeling assumption. The tables show price-only results before costs, taxes or dividends. An open gain is valued at the last close.

Reading the Market Compass record

Each CSV row shows what the report dated for that session actually said. An FTD is a follow-through day. The confirmation columns tell you whether the rally was confirmed by an FTD or by a new closing high. The invalidation columns give the date, index and reason recorded for the latest failed rally. An older invalidation can still appear beside a newer, active confirmation. The index codes ^GSPC, ^IXIC and ^NYA mean the S&P 500, Nasdaq Composite and NYSE Composite, respectively.

The Nasdaq’s September 21 confirmation came from a new closing high. On October 22 that confirmation was still active while both the S&P 500 and Nasdaq closed below their 50-day averages. On October 23 the Nasdaq undercut its rally low and the headline moved to correction.

The report dated October 29 still showed a rally in progress. The October 30 report identified an October 29 NYSE Composite FTD. On November 5 the NYSE Composite undercut its rally low. The November 11 report then recorded a Nasdaq Composite FTD. Its reported volume increase was 28.6% that day and 37.6% in the November 12 report, both against the previous session’s volume. The CSV preserves each report’s figure. Its row for November 13 has no headline because that dated report is missing. The November 16 report dates the failure to November 13 and identifies a Nasdaq close below the 200-day average. Download the Market Compass record