CPI came in higher than forecast — what should you compare before drawing a conclusion?

01Which number did the headline actually quote?

Your news app says "inflation came in at 3.4%". That number comes from a report on what shoppers pay for food, rent, fuel and the rest — the Consumer Price Index, or CPI. Almost every country publishes its own CPI, but the release that moves crypto, stock and gold charts around the world is the American one, so the very first thing to check is whether your headline is quoting the US report or your own country's. There is also a separate report called PPI, which tracks the prices producers receive rather than the prices you pay in a shop.

Inside one CPI report there are two cuts you need to keep apart. Headline includes everything; core leaves out food and energy, because those swing sharply month to month and the steadier core number is treated as a better picture of the underlying trend. Each of those comes in two forms: the change over the last month, and the change over the last twelve months. That is four numbers from one report.

Look at that table and you can see how easily two people read the same report differently. One says "inflation is 3.4%", another says "core is only 2.4% and falling toward target" — both are quoting the same day's data. Before you form any opinion, say out loud which cell of the table you are talking about.

02Higher than what? Actual, previous and forecast

"Higher" is meaningless on its own. There are three reference points, and they answer different questions. The previous month's number tells you the direction of the trend. The central bank's target tells you how far away policy still is from comfortable — the US central bank, the Federal Reserve (everyone calls it the Fed), tries to keep inflation around 2% over time. The forecast tells you what traders had already positioned for, and that is the one price reacts to most.

Here's a made-up release to show the difference, with illustrative numbers only. Monthly headline CPI was 0.2% last month, the forecast for this month is 0.2%, and the actual comes in at 0.3%. Inflation barely changed in real life, but the surprise is +0.1 point, and the reaction is usually strongest when data surprises the market rather than landing in line with forecasts. Now flip it: actual 0.3% against a forecast of 0.5%. Inflation rose faster than last month, and the market can still rally, because it had braced for worse.

This is why the honest habit is to write the forecast down before the release, with the source you took it from. Economic calendars show a consensus collected from analysts, and different calendars can show slightly different numbers, so note which one you used. If you decide what "high" means after the candle has already moved, you are describing the chart, not reading the data. The page on setting up an economic calendar covers getting those dates and times into your own time zone.

03The step most beginners skip: rate expectations

CPI does not move bitcoin directly. The chain runs through what traders expect the Fed to do: hot inflation can push it to raise rates, cooler inflation gives it room to hold or cut, and every report updates those bets. Prices then adjust to the new expectations — which is why markets can move days or weeks before any decision is announced. Why a US rate decision touches a coin that no government runs is its own topic, covered on the Fed rate and bitcoin page.

The September 2026 example shows how uncertain that step is even for professionals. August CPI came out on 11 September, slightly above expectations. Five days later, on 16 September, the Fed's rate-setting committee raised its main interest rate by a quarter of a percentage point, from 3.50–3.75% to 3.75–4.00%, and every member voted for it. Before that meeting, traders' bets were pointing to a small rate rise; many economists still thought the Fed would leave the rate where it was. Same data, two conclusions.

Some calendars also show the market's own odds for the next rate decision, and that is worth a glance: if the odds already point strongly to a rise, a hot number may barely move the chart, because the move was made days ago. What those probabilities actually show is a useful read once you start watching releases regularly.

04Prepare several reactions, not one prediction

A beginner reads "CPI hot" and concludes "short bitcoin". A more useful exercise is to write down two or three ways the same report could play out, before it lands, and watch which one the market picks.

• Hot and genuinely surprising: traders stop expecting a rate cut soon, and crypto and stocks often fall together.
• Hot headline, soft core: the market may decide the jump was just food and fuel and ignore it, so the first move can reverse within the hour.
• Hot but close to what was already expected: barely a reaction, because positions were set days ago.
• Hot, and crypto rises anyway: crypto has its own drivers — upgrades, regulation, flows and sentiment — and the inflation link is a tendency, not a rule.

That last one matters most for your confidence. If you expect a mechanical rule and the market does the opposite, you will either freeze or chase. If you expected four possible outcomes, the odd one is just another data point.

05What this looks like with 100 USDT

Say you have 100 USDT and you decide you want to be in the market when the number lands. You open a 300 USDT position on bitcoin at 5x, so 60 USDT of your money is set aside as margin. Every 1% move in the price is 3 USDT for you — up or down. That is a planned amount you can live with.

Now picture the first minute after the release. Price can travel two or three percent in one direction, turn around and do the same the other way, all before you have finished reading the headline. On that 300 USDT position each 1% is another 3 USDT, and it moves faster than your thumb. The trade size is the only part of this you control completely, which is exactly why you keep it small on a news release.

06A simple routine for the next release

This is the US inflation report, published by the country's statistics office — the Bureau of Labor Statistics, usually shortened to BLS — and its calendar is set months ahead, so you can plan around it calmly. The next one covers October and lands on 10 November 2026. The BLS says it updates the schedule as needed, usually at least a week before publication, so check the date again a few days before.

1. The day before, write the four numbers you will look for

   Headline monthly, headline yearly, core monthly, core yearly — and the forecast next to each, with the calendar you took it from.

2. Decide which contract you would even trade

   Open the exchange and check the contract is listed, what the minimum order size is and what leverage is allowed on it. The contract specification is in the app next to the pair.

3. Write both scenarios before the data

   "If the actual number beats the forecast by more than X, I expect…" and "if it lands in line, I do nothing." Doing nothing is a complete plan.

4. Watch the first release without a position

   Note what moved, how far and how fast, and whether the move held after thirty minutes. One release observed carefully teaches more than three traded on impulse.

5. Only then try it with a small amount or on demo

   Keep the size at a level where being wrong costs you a few dollars you planned to risk, like the 3 USDT per percent above — not an amount that ruins your week.

If you have never watched a release with a chart open next to it, the easiest way to start is to open a Binance Futures account, find BTCUSDT and leave the chart on the screen at release time with no position at all. Check in the app whether futures and the specific contract are available where you live — that varies by country and by exchange — and whether demo trading is offered to you.