Bitcoin 100-Day Moving Average Strategy (Backtested)
June 23, 2026
A simple trend rule - own bitcoin only when it's trading above its 100-day average, otherwise sit in cash - is one of the most popular ways to ride the upside while sidestepping the worst crashes. Here's what it actually did, on real data, with every number from a live backtest.
The short answer
Over four years of real bitcoin data (Jan 2022 - Jan 2026), holding only while price was above its 100-day moving average and otherwise sitting in cash returned +122.5% versus +86.2% for simply buying and holding - while cutting the worst drawdown from 66.9% to 29.8%.
The headline isn't that it beat buy-and-hold on return; it's that it did so while sitting out the deepest part of the 2022 crash. The catch: it trades, it pays fees, and a sharp whipsaw market can hand back that edge.
The rule
“Hold bitcoin while it's above its 100-day average. Move to cash when it drops below.”
That's the whole input - typed in plain English, exactly as you'd give it to Premiss.
The backtest
BTCUSDT · 2022-01-01 → 2026-01-01
Strategy return
+122.5%
$22,254
Buy & hold
+86.2%
$18,616
Max drawdown
-29.8%
buy & hold -66.9%
Round-trip trades
7
57% win rate
Real backtest on BTCUSDT daily candles, 2022-01-01 to 2026-01-01, with a 0.1% fee per trade. Wins averaged +35.1%, losses -10.2%; best trade +80.8%. These are hypothetical results on past data - see the disclaimer below.
How to read this result
The return edge here came mostly from avoiding catastrophe, not from clever timing. Buy-and-hold fell nearly 67% at its worst; this rule's worst fall was under 30%, because it was in cash for much of the 2022 decline. Less pain is, for most people, the more important number than more return.
Look at the trade log, not just the headline. Across the four years the rule made only a handful of round-trip trades, and a single big winner did a lot of the work. That's normal for trend-following, but it means the result leans on a few decisions - exactly the kind of thing you should test across other periods before trusting.
And notice this is one asset over one four-year window. It's evidence, not proof. The honest way to use it is as a starting point you then vary and re-run.
Where it breaks
The honest part. Every backtest has conditions under which it stops working - here are the ones that matter most for this rule.
- Whipsaw markets are the enemy. In a choppy, sideways market that keeps crossing the 100-day line, the rule buys high and sells low repeatedly, paying fees each time and bleeding the account - the opposite of what it did during the clean 2022 downtrend.
- The moving-average length is a choice. 100 days worked here; 50 or 200 would give different results. If only 100 works and the neighbours fall apart, that's a sign of overfitting rather than a real edge.
- Fees and slippage matter. This test used 0.1% per trade; higher real-world costs, or worse fills, would eat into the edge - more so if you traded a faster version.
- Past performance is not future performance. A trend rule that shone across one bull-and-crash cycle can underperform in the next regime. Treat the backtest as a filter, not a promise.
Why this rule is worth testing
The strategy is deliberately simple, which is part of why it's worth testing: simple rules are harder to overfit. There's no parameter to twist until the past looks perfect - just one line and one threshold.
It also captures a real, intuitive idea: stay invested while the trend is up, step aside when it turns down. The backtest is how we find out whether that intuition survived contact with four years of real prices.
Run this backtest yourself.
Open this exact strategy in Premiss and re-run it on years of real market data - then change it and test your own version.
Frequently asked questions
Does the 100-day moving average strategy beat buy-and-hold for bitcoin?
In this specific backtest over Jan 2022 - Jan 2026 it returned more than buy-and-hold while suffering a far smaller maximum drawdown, mainly by sitting out much of the 2022 crash. That's one asset over one window - it's evidence, not a guarantee, and a whipsaw market could erase the edge.
Why 100 days and not 50 or 200?
100 days is a common medium-term trend filter, slower than 50 (fewer false signals, but later exits) and faster than 200 (quicker to react, but more whipsaws). The robust approach is to test several lengths and prefer a strategy that works across a range, not only at one exact value.
What does 'move to cash' mean here?
When bitcoin closes below its 100-day average, the strategy sells its position and holds cash - earning nothing but also risking nothing - until price climbs back above the average, at which point it buys again.
Keep reading
Premiss is a research and education tool for building and testing trading ideas. It is not financial, investment, or trading advice. Backtested and simulated results are hypothetical and do not guarantee future performance. Trading involves risk, including the possible loss of capital.