Bitcoin: What Our Signals Do
Our Bitcoin signals don't beat buy-and-hold. In a typical quarter, you'd have made more money doing nothing. But in Q2 2022, when BTC dropped 56.8%, our bearish signal fired early and produced +34.77% alpha. That's the value — a systematic exit plan for the moments that ruin portfolios, not a trading edge for ordinary markets.
Most of the time, Bitcoin just runs. You hold, it goes up (or sideways), and any trend-following strategy spends half its time underperforming the index by small margins. That's the honest baseline. Once you accept it, the Q2 2022 number stops looking like a lucky streak and starts looking like exactly what the system was built for.
What the Backtest Showed
We ran MACD(12,26,9) + ADX(14)>20 on five years of daily BTC data across 20 quarters. The headline numbers aren't flattering: mean quarterly alpha of -1.89%, and we beat buy-and-hold in exactly 50% of quarters — a coin flip on average.
What surprised us was how those losses and wins distributed. The losing quarters were mostly small — small enough that a few big wins could overwhelm them entirely. The worst quarter was Q4 2024, where our strategy underperformed by -38.86%. BTC ran hard and we got caught in the timing lag.
But here's the other side. Q2 2022 alone — one quarter out of twenty — delivered +34.77% alpha when BTC fell 56.8%. Q3 2022 added another +9.3%. Q4 2022: +10.1%. Three quarters in a row where the exit signal mattered.
Win rate across individual signals: 36.3%. Roughly one in three signals made money. The system generates around 18 signals per year on the 1D timeframe.
Why ADX Greater Than 20
The ADX threshold isn't arbitrary. We tested every meaningful level from 15 to 30 on five years of daily data. ADX>20 came out clearly best — not because it fires more signals, but because it filters the right ones.
At ADX>30, the strategy beats buy-and-hold in only 30% of quarters. The higher bar sounds safer, but it actually makes things worse. By requiring ADX to hit 30 before a signal counts, you miss the early phase of real trends — exactly when the exit matters most. ADX typically sits around 20–25 when a major bearish move is just getting established. Waiting for 30 means waiting until you've already taken the first 10-15% of the drawdown.
At ADX>15, the signal drowns in noise. You get more alerts, but the quality drops and the strategy's total alpha turns negative.
ADX>20 on the 1D timeframe is where the signal earns its keep. The 4h timeframe, by contrast, is genuinely unprofitable on five years of data — too many signals in choppy conditions, no edge over the noise. We don't use it for trading signals.
The 4h Early Warning Signal
We do track the 4h timeframe as a passive risk indicator, just not as a trading signal.
The 4h bearish MACD signal catches 92.5% of major BTC crashes — 37 out of 40 significant drawdown events across five years — with roughly five days of lead time before the worst of the drop hits. That's genuine early warning value.
The tradeoff: a 57% false positive rate. More than half the 4h bearish signals don't lead to real crashes. That's why we don't send alerts for them — too many false alarms would make the whole system feel unreliable. Instead, the 4h signal surfaces as a passive "risk conditions" indicator on the dashboard. Think of it as the smoke detector going off before you can smell smoke, but also going off occasionally when someone burns toast.
If you see the 4h bearish indicator active and you're holding a meaningful BTC position, that's worth knowing. Not a sell signal. A heads-up.
The Quarters That Mattered — and the One That Hurt
Q2 2022 is the proof point for the entire system. BTC fell 56.8% — one of the worst crypto quarters in the asset's history. Our bearish signal fired early, and the strategy delivered +34.77% alpha versus hold. That's not alpha from clever positioning; it's alpha from not holding during a crash.
Q2 and Q3 2022 combined added +44.1% in alpha over two consecutive quarters of genuine market stress. Those two quarters alone justify the system's existence for BTC.
The hardest quarter was Q4 2024. BTC ran a strong bull rally, and our strategy underperformed by -38.86%. We were too early to call the reversal, then too slow getting back in. That kind of miss is structural — trend-following always gives back some gains at the edges of major rallies. The Q4 2024 result isn't a malfunction. It's what the system costs in good markets.
Who This Serves
BTC's signal profile is classified as Tier 2: bearish signal detection. Not consistent outperformance, not crash elimination — systematic exit signals during the downturns that matter.
If you can hold through 50-60% drawdowns without panic-selling, buy-and-hold has historically outperformed our system on average. That's the honest benchmark, and we mean it.
The value isn't beating the index. It's that a 56.8% drawdown is a life-altering event for most portfolios. Having a systematic signal that fired early in Q2 2022 — based on 5 years of evidence, not gut feel — is worth the average quarterly drag.
We don't know if BTC's crash patterns will repeat. Five years across 20 quarters is meaningful data, but Bitcoin is a 15-year-old asset that hasn't been through every macro cycle. The 2022 performance is evidence, not a promise.
If you already hold Bitcoin, our signals give you a systematic framework for the one question most investors answer emotionally: when to step aside.
This is educational content, not financial advice. Past performance does not guarantee future results. Based on 5-year daily data through Q1 2026, Polygon.io data.