Gold macro

    GOLD SEASONALITY

    Seasonality in gold is real, small and frequently misused. Here is the measurement, the sample size, and an honest statement of what it can and cannot support.

    On Gold Hunts data since January 2015, January is the strongest month for XAUUSD with an average return of +3.72% and 72.7% of months positive. Sample sizes are around eleven observations per month, so monthly seasonality is a weak prior for position sizing, not a trading signal.

    THE MONTHLY MEASUREMENT

    Every seasonality claim depends entirely on the sample it was built from, so start with ours: month-end XAUUSD closes from January 2015 to the present, giving roughly eleven observations per calendar month. That is a small sample by any statistical standard. We publish it because the alternative — repeating decades-old seasonal folklore built on a different volatility regime and a different monetary system — is worse.

    The measurement produces two numbers per month that matter: the average return, which tells you the historical magnitude, and the percentage of months that were positive, which tells you how consistent that magnitude was. A month with a large average driven by one enormous outlier is a different animal from a month with a modest average and a high hit rate, and only reading both columns together protects you from that confusion.

    What the numbers support

    January stands out on both metrics: the strongest average at +3.72% and the highest consistency at 72.7% positive. That combination is the only genuinely notable feature of the sample, and even it rests on eleven observations. The plausible explanation is a real-world mechanism — new-year allocation into commodity and reserve baskets, Asian physical demand around Lunar New Year — but a plausible mechanism plus eleven data points is a hypothesis, not an edge.

    The rest of the distribution is unremarkable. No month has an average large enough relative to gold's monthly volatility to justify a trade on the calendar alone. Anyone presenting a monthly seasonality chart as a trading plan is exploiting the reader's pattern-matching instinct.

    What the numbers do not support

    They do not support entering long gold on the first trading day of January and exiting at month end. They do not support avoiding a month because its average is negative. And they do not support any claim about a specific future month, because a distribution of eleven values gives you almost no confidence about the twelfth.

    The honest use is asymmetry at the margin: in a month whose historical profile is favourable, and where your structural and macro read already points the same way, you can justify sizing at the top of your normal range rather than the middle. That is the entire legitimate application.

    MonthYearsAvg %Median %Positive %Best %Worst %
    January113.724.8072.712.85-2.66
    February120.92-0.0150.010.74-6.10
    March120.88-0.2050.09.64-11.26
    April121.781.3166.77.24-2.11
    May120.080.0150.07.79-6.07
    June12-0.81-1.5733.38.80-11.77
    July121.271.5566.710.90-6.56
    August122.221.1450.013.80-3.11
    September11-0.59-2.9827.311.70-4.71
    October111.621.8863.67.27-2.93
    November11-0.93-0.5145.58.29-8.16
    December112.202.2572.76.72-1.81

    Average XAUUSD month-over-month return and share of positive months by calendar month, month-end closes since January 2015. Roughly eleven observations per month.

    Average monthly XAUUSD return by calendar month, same sample. Read the height alongside the hit rate in the table above, never on its own.

    INTRADAY SEASONALITY IS THE USEFUL KIND

    While monthly seasonality is statistically thin, intraday seasonality in gold is strong, mechanically explained and built on thousands of observations. Across 2,599 hourly bars, the average XAUUSD range at 13:00 UTC is $24.09 while the 21:00 UTC rollover hour averages $2.80 — a ratio of nearly nine to one, driven by the clock rather than the calendar.

    At the session level, New York captures 52.6% of the average daily range, London 33.1%, Asia 10.8% and the rollover window 3.6%. These proportions are stable, explained by where liquidity and scheduled data sit, and they have direct consequences for every stop, target and expected-range assumption in a trade plan. This is where a trader looking for exploitable regularity should spend their attention.

    Average XAUUSD hourly high-low range by UTC hour across 2,599 hourly bars — the strongest and most reliable seasonal pattern in the instrument.
    Session (UTC)DaysShare of daily range %Avg range ($)
    Asia10845.040.72
    London10843.440.45
    New York10852.649.74
    Rollover and late New York10731.020.67

    Share of the average daily range captured by each session. Asia 00:00–06:00, London 06:00–13:00, New York 13:00–21:00, rollover 21:00–00:00 UTC.

    Sample size discipline

    Eleven observations per calendar month versus 2,599 hourly bars. When two patterns disagree about where your attention belongs, the one with two hundred times the sample wins.

    Turn intraday seasonality into rules

    HOW A GOLD TRADER USES THIS

    Use the calendar for expectation setting and the clock for execution. Knowing that January has historically been strong is useful only as a tiebreaker when your macro read and your structure already agree. Knowing that the 13:00 UTC hour moves nine times as much as the 21:00 UTC hour changes where you place stops, which sessions you trade, and how you size — every day.

    And journal both. Tag each trade with the session it was taken in and the month, then review at the year end. Most traders discover their results are concentrated in one or two sessions and roughly random across the calendar, which is exactly what this data predicts.

    Seasonality discipline

    1. 01

      Never take a trade on calendar seasonality alone.

      Pass test: A structural reason exists independent of the month.

    2. 02

      Always quote the sample size when using a seasonal claim.

      Pass test: Roughly eleven observations per calendar month here.

    3. 03

      Scale expected range to the session, not to the daily average.

      Pass test: Range assumption differs between Asia and New York.

    4. 04

      Tag every journal entry with session and month.

      Pass test: Year-end review can separate the two effects.

    Related pages

    HOW TO READ THIS WITHOUT FOOLING YOURSELF

    A macro relationship is a prior, not a trade. It tells you which side of gold has the wind behind it over weeks and months; it does not tell you where your stop goes or whether this hour is a good hour to click. Every number on this page is a historical measurement over a stated sample, and gold has repeatedly spent months trading against its own textbook driver.

    Leveraged gold carries a high risk of rapid loss. Nothing here is financial advice, a signal, or a forecast of returns. Use the relationship to set context, then let your own tested rules and your own journalled results decide the position.

    Turn context into a tested process

    THE REST OF THE PILLAR

    The rest of the gold macro pillar

    GOLD TRADING FAQ

    What is the best month for gold historically?
    On Gold Hunts data since January 2015, January is strongest with an average XAUUSD return of +3.72% and 72.7% of months positive. The sample is only about eleven observations, so treat it as a weak prior rather than a signal.
    Is gold seasonality tradeable?
    Not on its own. No calendar month in our sample has an average return large enough relative to gold's monthly volatility to justify a position without an independent structural or macro reason.
    Why is January strong for gold?
    The plausible mechanism is new-year allocation into commodity and reserve baskets combined with Asian physical demand around Lunar New Year. With eleven observations, that remains a hypothesis rather than a demonstrated cause.
    What is the strongest seasonal pattern in gold?
    Intraday, not monthly. Across 2,599 hourly bars the 13:00 UTC hour averages $24.09 of range while the 21:00 UTC rollover hour averages $2.80. New York captures 52.6% of the average daily range.
    What sample does this page use?
    Month-end XAUUSD closes from January 2015 to date for the monthly table, and 2,599 hourly bars with timestamps normalised to true UTC for the intraday profile. Both are downloadable as CSV.

    TRACK YOUR OWN GOLD RESULTS FREE

    Live XAUUSD signals on a public ledger, a free gold trade journal that syncs with MetaTrader 5, and the macro pulse before the London open.