Morning Brief: WTI at $101.23 Meets Friday's Rate and FOMC Test

A timing study needs a normal-return baseline before it can call an observed move unusual. MacKinlay's event-study procedure places the estimation period before the event window and generally excludes event observations from parameter estimation. This avoids allowing the event itself to redefine normal performance. The paper concerns identifiable financial events. Applying its measurement discipline to a Gann-style inspection date is an editorial adaptation; it does not validate the date. Consider three synthetic pre-event daily returns of 1%, 2%, and 3%. Their constant-mean baseline is 2%. If the next completed return is 14%, its deviation is 12 percentage points. If that 14% bar is added to the training sample, the mean rises to 5%, and the reported deviation falls to nine points. No new market information justifies that change. The baseline was contaminated by the observation under test. A live recipe should save the training cut-off and the exact event candle timestamp. If the event mark is shifted after looking at the chart, it becomes a new specification. A short sample mean is useful for showing leakage, not for estimating a reliable trading edge.

Try it

Three synthetic pre-event returns are 1%, 2%, 3%; the event return is 14%. Compute the event deviation from the pre-event mean, then from the contaminated four-day mean.

Limits

The synthetic figures are not observations of any Gann cycle. MacKinlay's estimation-window rule measures event-study designs, not whether a calendar mark causes price movement. Ten daily observations in the recipe below are deliberately small for illustration.

Source

A. Craig MacKinlay, Event Studies in Economics and Finance, Section 2, estimation-window choice, printed p. 15; Section 5, Figure 1 and nonoverlap explanation, printed p. 20. Peer-reviewed Journal of Economic Literature 35(1), 1997, pp. 13-39; article reproduced as a searchable PDF in a public court exhibit. Boston University's hosted image scan confirms article metadata but has no extractable body text. Its empirical example uses 1989-1993 Dow 30 earnings announcements. The Gann-date and crypto adaptation is editorial.

Today

The last completed UTC candle was 2026-10-22 (opened 00:00Z, closed at 2026-10-23T00:00:00Z), with a BTC-USD close of $81,307.16. Its close-to-close return was -3.234181104%; the prior ten returns averaged +0.049735422%. The withheld Oct 22 return is 3.283916527 percentage points below that pre-event baseline. Including that tested return would shift the mean to -0.248802444% and its measured deviation to -2.985378660 points. This is a leakage demonstration at a predeclared inspection mark, not evidence of a Gann turning point or trading edge. Coinbase BTC-USD spot, UTC daily close, Latest session 2026-10-23T00:00:00Z: r_t(%) = 100 * (close_t / close_(t-1) - 1); pre_event_mean(%) = sum(r_D-11 through r_D-2) / 10; result(pp) = r_D-1 - pre_event_mean. Contaminated mean(%) = sum(r_D-11 through r_D-1) / 11. = -3.283916526525 percentage points (inputs: all_return_count=11, contaminated_all_11_mean_pct=-0.248802443715, contaminated_deviation_pp=-2.985378660477, event_return_pct=-3.234181104192, last_completed_close_usd=81307.16, pre_event_mean_pct=0.049735422333, previous_close_usd=84024.67, return_2026_10_12_pct=-2.313371843521, return_2026_10_13_pct=1.663137290883, return_2026_10_14_pct=-1.847970186996, return_2026_10_15_pct=0.692896538509, return_2026_10_16_pct=-0.768791885594, return_2026_10_17_pct=-0.14948458921, return_2026_10_18_pct=0.772491800938, return_2026_10_19_pct=-0.20627545138, return_2026_10_20_pct=1.344820461086, return_2026_10_21_pct=1.309902088614, return_2026_10_22_pct=-3.234181104192, training_return_count=10). Data source 1

Horizons quantity 44 reports the apparent longitude of the Sun in a coordinate system centered on the target body. The target's pole and orbit define its zero direction. For a modeled Mars pole, JPL marks 0 degrees as northern spring, 90 as northern summer solstice, 180 as autumn, and 270 as winter; southern season names reverse. This differs from true anomaly, which starts at periapse, and from an Earth-centered zodiac position of Mars, which uses another origin and observer geometry. Imagine a reading of 89.2 degrees. Mars remains 0.8 degrees short of its northern summer-solstice boundary. A 90-degree label would be premature; neither label is a claim about the next BTC candle. If a timing analyst wants to mark Mars seasons, they must name the target, output quantity, hemisphere and sample timestamp. A paired BTC example can plot its completed daily close at the same UTC instant as the Horizons sample. The chart could show contemporaneous conditions; it does not turn Mars climate geometry into a backtested market relationship.

Try it

JPL quantity 44 hypothetically reads 89.2 degrees for Mars. How far is the 90-degree northern summer-solstice marker, and has Mars reached it?

Limits

This quantity is body-centered and depends on a modeled spin pole. JPL warns that Earth seasons use a different convention and that remote viewing has light-time lag. No market effect follows from the seasonal label.

Source

NASA/JPL Horizons Manual, Observer Table, quantity 44, Apparent target-centered longitude of Sun (App_Lon_Sun). NASA/JPL Solar System Dynamics technical documentation. Mars seasonal labels are astronomical, not market regime evidence.

Today

At 2026-10-22 00:00 UTC and 2026-10-23 00:00 UTC, Mars apparent solar longitude (JPL Horizons Q44) moved from 10.6285044° to 11.1114229° (+0.4829185°). The aligned completed Coinbase BTC-USD daily candles closed at $84024.67 and $81307.16, a -3.234181% move. The Oct 22 UTC candle closed at Oct 23 00:00 UTC; the forming Oct 23 candle is excluded. These paired observations do not establish a causal or predictive relationship. Mars apparent solar longitude (JPL Horizons Q44, geocentric) with Coinbase BTC-USD UTC daily closes, Latest session 2026-10-23T00:00:00Z: Mars ΔL_s=(L_s(T1)-L_s(T0)+360) mod 360; BTC % return=100×(BTC_close(T1)/BTC_close(T0)-1) = 0.4829185 degrees (inputs: btc_change_percent=-3.234181104192376, btc_close_t0_usd=84024.67, btc_close_t1_usd=81307.16, mars_ls_t0_degrees=10.6285044, mars_ls_t1_degrees=11.1114229). Data source 1 Data source 2