Author: 分析师 卢明宇
Before US CPI is released, support and resistance on a gold chart can still be useful, but they cannot tell you whether the data will be above or below expectations. CPI has a scheduled release time, while real-yield data have their own frequency. The US Treasury's daily real-yield table is not a tick-by-tick quote. Combining the data result, the rate reaction and a candle breakout into one instant signal can encourage chasing the first sharp move. A clearer process is to define observation conditions before the release and then compare real-yield context with closing behaviour.
Confirm the data and time conventions
The Bureau of Labor Statistics CPI schedule lists release times and the relevant time zone. Record the US Eastern Time shown on the source page and convert it to Beijing time using the daylight-saving rule in effect that day. Also record the source of the actual, expected and prior values. Expectations from different calendars may use different survey samples and should not be combined.
Before comparing values, confirm whether they are headline or core CPI, month-on-month or year-on-year, and whether the seasonal-adjustment convention matches. A headline month-on-month number and a core year-on-year number may both be called inflation data, but they do not measure the same expectation gap.
The US Treasury daily real-yield curve provides daily observations for selected maturities. It can help review the direction of real yields, but it is not an intraday quote for the exact second of the CPI release. A higher-frequency study requires a same-time, same-maturity and same-source series.
Use a condition table before the release
Assume, for teaching only, that gold closes at USD 2380 per ounce, with resistance at 2395 and a retest level at 2360. Before CPI, write three conditions. If data are above expectations, real yields and the dollar may rise, so a long breakout needs stricter closing confirmation. If data are below expectations, rate pressure may ease, but price can still spike and retreat. If data are close to expectations, the market may focus on revisions, component details or existing positions. These are scenarios, not a forecast for a particular release.
| Observation step | Record | Do not infer directly |
|---|---|---|
| Before release | Prior high, support, maximum acceptable loss and no-chase condition | CPI must be bullish or bearish for gold |
| Just after release | Actual, expected, prior value and page time zone | The first candle is the new trend |
| Rate reaction | Direction and timestamp of a same-maturity real-yield series | A daily yield equals an instant transaction quote |
| Close confirmation | Candle close, retest, spread and execution status | A breakout must continue or a stop must fill at its line |
The first candle is volatility, not confirmation
Using the same hypothetical values, suppose gold jumps from 2380 to 2403 after CPI and then returns to 2388 during the observation period. Price crossed 2395 but did not close above it. Under a close-confirmation rule, record this as an intraperiod break awaiting confirmation. If the next candle regains 2395 and completes a retest, the technical condition has new evidence. If price falls through 2360, reassess the prewritten risk condition instead of using CPI as a single explanation for the loss.
News and price can also move in opposite directions. Markets trade the change relative to expectations, while the dollar, nominal yields, real yields and risk sentiment can adjust together. Even a below-expectation CPI reading may not lift gold if it was already priced in. Fundamental analysis identifies which variables need observation; technical analysis checks whether price meets the execution condition. Each has a different job.
For a review, record the result as a set of condition tags rather than a single after-the-fact cause. “Below-expectation data, lower daily real-yield reading and a close above resistance” is a record of an observed combination. If price rises without a matching yield move, label it as a divergence requiring further checking. The label describes the combination; it does not prove causality.
Align real-yield observations with price
Build a timeline containing the CPI page's release time, the quote timestamp, the yield-series observation time and the open and close boundaries of the candle. If the Treasury data provide only one daily value for the event date, label it as daily context instead of writing it as the 20:30 intraday reaction. Also record whether the gold quote is spot, futures or a CFD, and which time zone and price source the chart uses.
Gold's three hidden moving averages: rates, the dollar and liquidity discusses cross-market rate and dollar observations. False breakout or trend reversal? explains the division between fundamental context and breakout confirmation. For indicator units, see the gold and silver ATR comparison. These articles do not replace the day's data or product rules.
Put waiting into the trading plan
The main execution risks around CPI are easy to miss: quotes can jump, spreads can widen and a stop trigger does not guarantee the final execution price. Margin trading can also increase required margin or lead to forced liquidation when volatility expands. The 2380, 2395, 2360 and 2403 USD-per-ounce values are teaching assumptions; the release timing and data direction are not a forecast for any particular month. Act only after the source, time, price condition and acceptable loss have been checked.