What the 2022 Crash Revealed About Trading Bots

November 2021. BTC hit $69,000. Everywhere you looked, someone was running an automated trading bot and posting screenshots of their returns.

Twelve months later, BTC was at $15,476. A 77% drawdown. And most of those bots had quietly stopped working.

Not because automation was wrong. Because the wrong bot in the wrong market is worse than no bot at all — and 2022 was the most efficient stress test the industry had ever run.

Grid bots: range traders in a trending market

A grid bot places buy and sell orders at regular intervals. Price drops 2%? It buys. Bounces back? It sells. In a sideways, oscillating market this generates steady, mechanical returns. In 2022, BTC didn't oscillate. It trended down for twelve months straight.

Every time BTC dropped another 5%, the grid bot did its job and bought more. By $30,000, most grid bots were sitting on large unrealized losses and running low on capital. Below $20,000 — after LUNA wiped $45 billion from the market in a week and FTX collapsed in November — many either stopped out entirely or left operators holding Bitcoin at an average cost twice the current price.

The structural flaw is simple. Grid bots are built on the assumption of mean reversion. They're designed for markets that oscillate around a stable price. Give them a market that doesn't, and that assumption destroys accounts.

DCA bots: mathematically correct, psychologically brutal

Dollar-cost averaging gets more credit than grid trading, and deservedly so. Investors who committed to buying $10 of BTC per week throughout the entire 2022 bear market outperformed lump-sum buyers by over 33 percentage points. The math works.

What the math doesn't account for is the people running the bots.

When BTC fell from $50,000 to $30,000, most DCA bot operators kept going. From $30,000 to $20,000, some stopped. After FTX — when BTC was at $16,000, every crypto headline was catastrophic, and the Fear & Greed Index was scraping historic lows — most people turned the bot off and walked away. The strategy that would have paid off required them to keep feeding capital into the worst-looking market of the decade.

DCA bots also have no short-side exposure. They can't profit from a downtrend — they can only accumulate through it. That's a design choice that's fine for long-term investors. It's a serious limitation for anyone trying to actually trade.

Trend-following: the only strategy designed for this

While grid bots bled and DCA bots were being abandoned, trend-following funds had the best year in their recorded history.

This wasn't luck. Trend-following is market-neutral by design — it goes short when the trend is down, long when it's up. In 2022, BTC trended down for twelve months. A strategy built to follow trends in any direction turned that drawdown into returns.

The lesson most people draw is wrong

After 2022, a lot of traders concluded that trading bots don't work. What they should have concluded is that bot design determines everything.

Grid bots work in flat, oscillating markets. DCA is a long-term accumulation strategy that requires you to be both solvent and disciplined for years. Trend-following is the only approach that doesn't require a view on direction — it just requires a trend. And crypto markets trend more than they range.

The question isn't whether to automate. It's what you're actually automating.