Event preview · CPI · Wednesday 14 October 2026, 8:30 AM ET

US CPI October 2026: What September Inflation Means for Gold

By Emanuele Calcina · Published 8 October 2026 · Gold reaction measured on Dukascopy hourly data · Our news-day rules

8:30 AM ET on Wednesday 14 October is when US CPI for September hits the tape, with gold boxed near $4,125 (investingLive). There's no published forecast in our data yet. The last reading ran 0.4% month over month, with core at 0.3% and headline year over year at 3.4%. Gold has leaned toward CPI days all year. Across the last eight releases we measured, the release hour ran a median $42.50, about 2.1 times a normal hour of that day, and gold closed higher two hours later six times out of eight (Dukascopy hourly data). This one lands with the dollar firm and yields high (Reuters). The hour will move.

Forecasts are consensus figures from the TradingView economic calendar and can move before the release.

What September Inflation Means For Gold Right Now

The number that counts is headline CPI month over month for September. Our data shows no forecast posted yet, so anchor on the previous: 0.4% on the month. Core came in at 0.3% last time, core year over year at 2.4%, and headline year over year at 3.4%. All of it drops at 8:30 AM ET together. Gold cares because CPI feeds straight into the rate path. Hotter prints push real yields up and tend to weigh on gold. Softer prints do the opposite.

Gold's been rangebound into this, with the US-Iran stalemate and CPI risk capping the upside (investingLive). Price near $4,125 looks fragile even after a recovery (StoneX). The dollar is firm and yields are high, which has been offsetting waning rate-hike expectations (Reuters). There's even a case going around that AI, not gold, is the new inflation hedge (Seeking Alpha). Against all that noise, a September print that moves the rate story is the one thing that can break the range this week.

How Gold Moved On The Last Prints

We measured the last eight CPI releases on Dukascopy XAUUSD hourly data. The median release-hour range was $42.50, about 2.1 times the $19.90 a normal same-hour candle runs. The biggest was $98.10. On 11 September 2026 the release hour swung $98.10 and gold was up $68.10 two hours later, the largest of the set. On 14 July 2026 the print missed at -0.4% against -0.1% forecast, the hour ran $77.80, and gold added $45.40 over the next two hours. Surprises bring the range.

It doesn't always run big. On 10 April 2026 the hour held to $22.80 and gold drifted just $6.30. On 12 May 2026 the range was $22.00 but gold still faded $27.40 over two hours, so a tight hour does not mean a tame session. Across the eight, gold closed higher two hours out six times and lower twice. Six up, two down. That tilt is worth knowing, though eight prints is a small sample.

Measured on Dukascopy XAUUSD hourly candles, 13 Feb 2026 to 11 Sep 2026. The median release hour covered $42.50 against $19.90 for the same hour on a normal weekday. Two hours later gold was higher 6 times and lower 2 times.

Three Ways The September Print Can Land

Conditional from here, since there's no forecast to beat yet, read it against the 0.4% prior. A hotter print, above that pace, would typically lift yields and the dollar and press gold lower, at least on the first reaction. The July miss shows the reverse: a soft number sent gold up $45.40 in two hours. A cooler print, below the prior and especially a downside surprise, would usually ease yields and soften the dollar, which gives gold room. None of this is a promise. It's the usual map.

In line, right on the prior pace with no surprise, is the trickier one. Our data has a run of zero-surprise prints that still moved. On 12 May 2026 the number came in bang on 0.6% and gold dropped $27.40. On 11 September 2026 it matched at 0.4% and gold ran $68.10. So in line does not mean quiet. Positioning and the component detail can carry the move even when the headline matches. Treat a matching print as live, not a reason to relax.

When It Lands In Your Time Zone

The print drops at 8:30 AM ET, which is the same clock in Toronto. London gets it at 1:30 PM. Madrid, Rome and Amsterdam are at 2:30 PM. Dubai sees it at 4:30 PM. India at 6:00 PM. Wherever you sit, the candle to watch is the one the number lands in, measured in UTC, which is 12:30 PM UTC on Wednesday 14 October. That hour is where our median $42.50 range and the $98.10 outlier both live.

The hour before tends to go thin and nervous. Liquidity pulls back, spreads widen, and price can drift without meaning much. Then the number hits and the first candle does the damage. The hour after is where our 2h move is measured, open of the release hour to close of the next. That's the window that carried $68.10 in September and $45.40 in July. Expect the spread to stay wide through the print and the first few minutes to lie about direction.

How Our Desk Handles The CPI Hour

We keep it simple and boring on purpose. No new entries in the run-up to 8:30 AM ET. The spread is already widening and the range on our measurement runs about 2.1 times a normal hour, so a fresh entry into that is paying up for noise. If a trade is already running, its stop sits outside the expected release-hour range, not inside it, so a $42.50 median swing doesn't knock us out of a position that was working before the number.

After the print, the first move is not treated as the move. Gold on these days likes to spike one way and settle the other, and our data is measured to the close of the next hour for that reason. We let the first candle happen, watch where price actually wants to go once spreads come back in, and work from there. Slow hands beat fast hands on CPI day. The number is an event to manage, not a lottery ticket.

The short version

Gold sits near $4,125 into this, boxed in a range while the dollar and yields do the heavy lifting (investingLive) (Reuters). September CPI is the clearest shot at a rate-story shift this week. Our measurement says the release hour has run a median $42.50, about 2.1 times normal, topping out at $98.10 in September, and gold closed higher two hours later in six of the last eight. None of that tells you Wednesday's direction. It tells you to respect the hour and keep stops outside the range, then let the first move play out before deciding anything. Manage the event.

Questions traders asked

Trading gold on margin carries a high level of risk and you can lose more than your deposit. Nothing on this page is a personal recommendation. Past reactions to a release do not predict the next one.