"So, what's the bot doing?"
I get that question a lot. And the answer is usually disappointing:
"Nothing."
Since October 2024, my main bot has been sitting in cash. 184 days. No trade. No buy. No sell. Just... waiting.
And that's exactly the point. Because Bitcoin dropped 36% in those 184 days. And my portfolio? 0% change.
My bot did nothing. And got everything right.
In this article I'll tell you the story of my first and most important bot. HOW it thinks, WHY I trust it, and WHAT it taught me. The logic behind its filters I explain in the newsletter.
Why "The Watchdog"?
Because it watches. Doesn't hunt. Doesn't speculate. It stands at the gate and decides one single question:
"Should I be in Bitcoin right now, or not?"
That's all. No complicated trades. No leverage. No shorting. Just: IN or OUT. BTC or cash.
Sounds simple? It is. And that's exactly why it works.
How It Thinks
The Watchdog uses three independent filters. Each filter measures something different:
Filter 1 looks at the long-term trend. Where is Bitcoin moving over weeks and months? Up or down?
Filter 2 measures the behavior of the crowd. Is everyone euphoric and buying blindly? Or is there calm and rationality?
Filter 3 checks market conditions. Is the environment favorable for risky assets, or is there headwind?
The rule: Only when all three filters say "yes" does the bot buy. If even one is red, it stays in cash.
What exactly the filters measure? What data they use? What parameters? I'm not telling. Not because I'm mean, but because it's my edge. If everyone uses the same strategy, it stops working.
Remember: You see the RESULTS of the bot in real time, via the traffic light on our website. You just don't see the WHY.
The Numbers
Here's where it gets concrete. 6 years of backtesting on real Bitcoin data:
| Metric | The Watchdog | Simple HODL |
|---|---|---|
| Return (backtest) | +2,942% | +644% |
| Max drawdown | -35% | -77% |
| Time in cash | 57% | 0% |
| Trades per year | ~4 | 0 |
| Factor better than HODL | 3x | -- |
Important: The +2,942% is a historical backtest result. The bulk of it comes from Bitcoin growing massively in those 6 years (from ~$3,500 to over $80,000). This growth won't repeat itself like that because the market cap is much larger today. The number you should remember isn't 2,942%. It's the 3x. The bot beat HODL by a factor of 3. That ratio is the real value.
Read the drawdown line again. HODL has -77% maximum loss. The Watchdog only -35%. That means: HODL investors had to watch their portfolio collapse by three quarters. The Watchdog investors had at most a third in losses.
And still 3x more return. How does that work?
The Secret: Avoidance
The Watchdog doesn't make money by catching the BEST days. It makes money by AVOIDING the WORST days.
2022: Bitcoin fell from $69,000 to $15,500. The Watchdog? Had been in cash for months. Lost nothing.
2024: Bitcoin dropped 25% in the fall. The Watchdog? In cash since October.
2025: While others are panicking, my portfolio is at plus/minus zero.
That sounds unspectacular. It is. But the math behind it is brutal: If you lose -50%, you need +100% just to get back to zero. If you lose -77%, you need +335%. The Watchdog avoids that.
Remember: It's more important to avoid losses than to maximize gains. Sounds boring. But it's true.
57% in Cash. And Still Better
This confuses most people. "Your bot sits in cash HALF the time? Then it misses half the gains!"
True. It misses gains. But it also misses losses. And because losses mathematically WEIGH more than gains (crash recovery: -50% requires +100%), it wins on balance.
An analogy: Imagine two drivers. One always goes 200 km/h. The other goes 120 km/h but brakes BEFORE the curves.
After 1,000 kilometers? The 120 driver arrived faster. Because the 200 driver flew off the road three times and had to get repairs.
The Watchdog drives 120. Boring. But it arrives.
My Hardest Moment with The Watchdog
November 2024. The bot sells at $91,500. Goes to cash.
Bitcoin then rises to $95,000. Then $100,000. Then $108,000.
Everyone on Twitter: "TO THE MOON!" My bot: In cash. My brain: "YOU IDIOT BOT!"
It was unbearable. I seriously considered overriding the bot. Buying manually. "Just this once."
I didn't.
Three months later: Bitcoin at $84,000. The bot was right. Not perfect. It didn't sell at the absolute top. But it recognized the downtrend BEFORE it came.
Remember: The hardest moment in bot trading isn't when the bot loses. It's when it looks "wrong" -- and you need the discipline to trust it anyway.
What The Watchdog Taught Me
1. Patience is a superpower. 184 days of doing nothing. While everyone else is trading, panicking, FOMO-ing. The bot waits. Patiently. Emotionlessly. No human can do that.
2. The best returns come from what you DON'T do. Not buying the dip. Not panic selling. Not listening to Twitter. Just trusting the filters.
3. Few trades = low risk. 4 trades per year. 4 decisions. 4 times right or wrong. That's manageable. 200 trades per year? 200 opportunities to make a mistake.
4. I need the bot more than it needs me. The bot would be just as good without me. I'd be significantly worse without the bot. That's the most honest realization.
How to See The Watchdog Live
The traffic light on our website shows you in real time what The Watchdog is saying:
- Green = "I'm in Bitcoin"
- Red = "I'm in cash"
- 3 Filters = Uptrend, buying pressure, sentiment
You see the signal. You see how many days it's been active. You see the current BTC price. What you DON'T see: the exact filters and parameters.
Newsletter subscribers also get a notification when the signal CHANGES.
-> The Scout: Short-Term Opportunities -- Opportunities during cash phases
-> The Genius: Scanning the DeFi Market -- My most experimental bot
-> The Most Boring Strategy That Works -- 3x better than HODL in backtesting
-> DCA Savings Plan Calculator -- What a savings plan would have returned
-> What Does the Bot Say Right Now? -- Current signal
Your Dominic, the guy who trusts his bot more than himself. And sleeps better because of it.
Disclaimer: This is not investment advice. All backtest results are based on historical data and do not guarantee future results. Only invest what you can afford to lose.
This bot in the post-mortem ledger: See /post-mortems →
Every retired strategy and failed walk-forward — documented publicly.
