Optimal Bitcoin DCA strategy, courtesy of ML
Market movements are famously unpredictable, rendering past prices useless for accurate future predictions. And yes, all those Medium posts pretending to predict the price of $GOOG with 95% accuracy using machine learning are obviously bullshit.
Timing the market is folly; non-professionals should opt for the sensible strategy of Dollar Cost Averaging (DCA). It involves regularly purchasing a fixed dollar amount of an investment, regardless of its current price, to mitigate volatility. Automating the process further eliminates personal biases.
Still, I’ve been asking myself about ways of optimizing DCA in order to buy lower. So yeah, there are bots (like this one), but their benefit/complexity ratio kinda sucks, and they can do stupid things when price moves are swift.
So, here’s the question: When is the optimal time to schedule DCA buy orders? To find an answer, I examined eight years of BTC price history, fitting a time series model. Analyzing its seasonal components revealed some interesting insights:
- BTC tends to be cheaper on Fridays.
- The beginning of July carries significant price advantages.
- Around 4 am is another opportune time.
- Approximately the 20th day of the month offers favorable prices.
By aligning automatic orders with these points, one can secure the best deals on average. It’s worth noting that halvings have a profound impact on BTC’s price—the further away from a halving, the cheaper BTC tends to be.
