FX Claims, Tested · #1

Trade with the trend?

“The trend is your friend. If the 50-day average is above the 200-day, look for longs.”

Verdict: not supported

Seven major pairs · 2004–2026 · last run 1 Oct 2026

Why we tested this one first

Every currency pair page on this site shows a trend label: Uptrend, Downtrend or Mixed. Before testing anyone else's ideas, it seemed only fair to check whether our own label actually means anything.

The rule

Uptrend	Price above its 50-day average, and the 50-day above the 200-day
Downtrend	Price below its 50-day average, and the 50-day below the 200-day
Mixed	Anything else

It's the textbook definition, and close to what a lot of trading courses teach as the first filter: only take trades in the direction of the trend.

How we tested it

All seven majors, daily closes from 2004 (2006 for AUD/USD) to today. No cherry-picking a pair that happened to work.
No hindsight: each day's label uses only prices known at that day's close, and any trade happens from the next day.
Trading costs included: 0.02% every time the position changes, roughly two pips on EUR/USD for spread and slippage.
Two halves: 2004–2014 and 2015–today, because anything real should show up in both.
Not included: overnight swap. On a real account that adds or subtracts a little depending on the rate gap.

Question 1: does the label predict the next month?

Average move over the next 20 trading days, depending on the label that day. If the label worked, the Uptrend column would be clearly positive and the Downtrend column clearly negative.

Pair	Time in uptrend	Next 20 days after uptrend	Time in downtrend	Next 20 days after downtrend	After mixed	Any day
EUR/USD	31%	−0.02%	30%	−0.10%	+0.12%	+0.01%
GBP/USD	30%	−0.34%	27%	−0.21%	+0.20%	−0.07%
USD/JPY	32%	+0.08%	25%	+0.25%	+0.18%	+0.16%
AUD/USD	31%	−0.13%	28%	−0.11%	+0.21%	+0.02%
USD/CAD	29%	+0.11%	26%	+0.17%	−0.07%	+0.04%
USD/CHF	24%	−0.44%	33%	+0.04%	−0.03%	−0.11%
NZD/USD	29%	−0.10%	26%	+0.10%	+0.00%	+0.00%

They aren't. The numbers are tiny, point in both directions, and none of them is distinguishable from random chance once you allow for how noisy currency moves are. After an "Uptrend" label, pairs went on to rise about as often as on any other day.

Question 2: would following it have made money?

Long when the label says Uptrend, short on Downtrend, flat when Mixed, spread equally across all seven pairs. This is how each unit of money would have grown, after costs.

Yearly return

−1.21%

Worst drawdown

-36%

Winning years

6 of 21

Best year

2008: +11.7%

Year by year

The rule did its job in 2008, when the financial crisis created huge, one-way moves. Most other years it gave a little back, and the slow bleed wins over time.

Pair by pair

"Sharpe" measures return for the risk taken: above zero means it made money, around 0.5 or more is where a strategy starts to look genuinely useful. Every pair is close to zero or negative, and most got worse in the second half.

Pair	Yearly return	Sharpe	Worst drawdown	Sharpe 2004–14	Sharpe 2015–now	Position changes
EUR/USD	+0.13%	+0.02	-29%	+0.14	-0.13	444
GBP/USD	−1.03%	-0.14	-37%	+0.05	-0.34	447
USD/JPY	−1.88%	-0.25	-45%	-0.27	-0.23	523
AUD/USD	−0.32%	-0.03	-45%	+0.24	-0.34	426
USD/CAD	−1.95%	-0.29	-47%	-0.37	-0.21	484
USD/CHF	−2.47%	-0.30	-59%	+0.04	-0.69	523
NZD/USD	−0.76%	-0.08	-54%	+0.17	-0.40	468

What this means for you.

The trend label is a description, not a forecast. It tells you where a pair has been. That's still useful context, but on its own it gave no edge on where the pair went next.
"Only trade with the trend" isn't the safety net it's sold as, at least not in this simple form on daily FX charts. Over twenty years it would have cost you money after spreads.
It shone in a crisis. In 2008 it made 12%. That's the pattern trend-followers often describe: long dull stretches, then a few big years. The catch is surviving the dull stretches, and here they lasted over a decade.

What this test doesn't prove.

It tests one popular rule on daily charts. Other timeframes, other averages, or trend-following across dozens of markets (the way the big funds do it) are different claims.
It uses end-of-day data from Yahoo Finance, which has occasional bad ticks. We removed obvious ones, but it isn't institutional-grade data.
Past results can't tell you what happens next. A strategy can fail for twenty years and then work, or the other way round.

Risk warning: CFDs and spread bets are complex instruments and come with a high risk of losing money rapidly due to leverage. Most retail accounts lose money when trading them. You should consider whether you understand how they work and whether you can afford to take the high risk of losing your money. Satdish provides education, not financial advice.

Data: Yahoo Finance daily closes for EUR/USD, GBP/USD, USD/JPY, AUD/USD, USD/CAD, USD/CHF and NZD/USD. Forward-return averages use every day; significance checks use non-overlapping 20-day periods so the same move isn't counted twenty times. Re-run monthly.