Research12 July 2026· 5 min read

Why your backtested win rate doesn't survive contact with real news events

You backtested a strategy. It shows a 62% win rate over five years. You go live. Six weeks in, you’re at 47%.

Most retail traders blame themselves. I must be executing it wrong, or I broke a rule somewhere. Sometimes that’s true. Often it isn’t.

There’s a specific reason a lot of backtests overstate performance, and curve fitting isn’t it. The culprit is news event contamination.

What contamination actually is

Every backtest is a set of trades taken on some entry signal. What most retail backtesting tools don’t check: were any of those trades entered within a few minutes of a scheduled news release?

If a trade was open when NFP hit, or when CPI dropped, or when the FOMC statement came out, the price action during that trade wasn’t driven by your setup. It was driven by news flow that had nothing to do with the technical pattern you were testing. That trade tells you nothing about whether your strategy has an edge. It tells you the market moved because of news, and your position was along for the ride.

If it moved in your direction, a lucky win that inflates your backtested win rate. If against, a loss that got blamed on your strategy but wasn’t caused by it. Either way, that trade is noise pretending to be signal.

How much this matters

We looked at 653,879 short-window reactions to macro events (1-minute through end-of-day) across 41 instruments over sixteen years. About 4.8% of them had a second high-impact event land inside their measurement window: contaminated, unable to attribute the price move to the release that started it.

That’s not the number for your backtest. Yours probably has more, because a trading strategy overlaps news events at whatever rate the news occurs, which for major FX pairs is several times a week.

Take a strategy that averages one trade per day on EURUSD. Now count the high-impact events involving USD or EUR in a typical month. About a dozenimpact-3 releases: NFP, CPI, PPI, Fed speakers, ECB, retail sales, PMIs, jobless claims. Add in the mid-tier stuff and you’re looking at several events every week where any open EURUSD trade is being pushed around by something that isn’t your entry logic.

If you’re not filtering those out, a meaningful chunk of your “wins” and “losses” aren’t measuring your strategy. They’re measuring the news.

Why live results differ

The reason live performance drops is that in real time, you feel every one of those news-driven moves. You cut the wins early because the news made price spike and you didn’t trust it. You hold the losses longer because you’re waiting for the “news distortion” to fade. Your execution deviates from your backtest not because you’re undisciplined, but because your backtest included trades where price action was noise, and your live self correctly recognized them as noise, just too late to avoid the position entirely.

What to do about it

Two options.

Filter events out of your backtest.Before you count a trade in your stats, check whether it overlapped a high-impact release. If it did, either exclude it or bucket it separately as a “news trade” so you can see whether your strategy actually has an edge outside those windows.
Stop trading through them.If your strategy isn’t specifically designed for news reactions, close positions before major releases and reopen after. You’ll take fewer trades. Your win rate on the remaining ones, the trades your setup was actually built for, will be a lot closer to what you backtested.

News Impact Explorer shows you exactly which events, on which instruments, produce measurable reactions. It’s the same filter list you’d want before running any backtest or leaving a trade open into a release. Free to use: tryvantage.co/news-impact

Keep reading

EURUSD Thu 12:30 is quiet 15% of the time. Here's why.127 of 832 Thursdays at 12:30 UTC were calm on EURUSD. Why the number isn't zero, and why the tool marks this cell DST-ambiguous.Four XAUUSD reaction records fell in four weeksFed, GDP, NFP, CPI: four US macro events between June 17 and July 14 each broke the largest 15-minute XAUUSD reaction on record, by 1.74x to 2.33x. What the data shows, and what it doesn't.How to check what a release does to your pairEvery scheduled release has a track record. Here is how to read it in News Impact Explorer, panel by panel, using NFP on gold as the worked example.What actually happens to EURUSD in the 15 minutes after NFP (and the next five days)195 releases, sorted by surprise, measured from 15 minutes out to 5 days. The story isn't drift. It's asymmetric: beats mean-revert, misses hold, in-line does nothing.The Bloomberg terminal problem: why institutional edge on macro events was never available to retailThe pros really do have data you don't. Here's exactly what it is, why the gap stayed open for so long, and why it's finally closing.