A few weeks ago, according to TradingView prices, there was a bitcoin trade that could have made a trader over 300% annualised with a single position held for a few days. Perhaps it has to do with Shiba Inu? Not quite.
In our previous, Long Story Short, we covered a basic framework for market making. Today, we break down another powerful strategy that has a major impact on cryptocurrency markets: spread trading.
You may have already heard of the terms “basis trading” and “cash-and-carry” by now, since they’ve become buzzwords in our community. Conceptually, the spread trade is a generic form of those two and is related to the pairs trade and statistical arbitrage. The spread strategy can also be viewed as a variant of the “spot” arbitrage trade, likely involving at least one derivative.
Starting off with example makes it much easier to understand.
In the chart above, we focus on the BTCUSD perpetual swap on Okex and Bitfinex. We see that on October 24, we could have gone long one bitcoin-worth on Okex at a price of around $60,000 and short one bitcoin-worth on Bitfinex at around $61,000. By holding until October 26, we could have sold the Okex long position at around $62,250 and exited the Bitfinex short position at around $60,000.
This generates a profit of about $2,250 on Okex and $1,000 on Bitfinex, or a total of $3,250 in two days, excluding fees. Let’s suppose we used about 2 BTC in margin or about $120,000 at the time, we just earned a little over 2.5% in roughly 2 days.
Let’s generalise this.
While the two products seem different, they are highly related and thus highly correlated. Instead of thinking of them individually, we can think of them jointly by thinking of the “spread” between them. If we consider them in those terms, we can treat their divergence from each other as a “spread price”.
Spread_Price = First_Product_Price — Second_Product_Price
We are now thinking in spread prices (note: negative spread prices can exist). We can see that the spread price is -$1,000 on October 24 and +$2,250 on October 26. By purchasing one bitcoin-worth of the spread on October 24 and selling it on October 26, you would have generated a spread profit of $3,250.
Look familiar? $3,250 was exactly the same profit when we did it individually.
Why does this strategy tend to work? Remember that we are trading the spread between two highly-related products. As a result, the two products have a very strong correlation in both the short-term and long-term. Considering that, we can probably expect the price spread between the two products to be fairly tight on average. However, rare circumstances could cause the two products to diverge. We expect the gap between the two products to snap back “soon enough”, creating trading opportunities for someone paying attention. However, heed our warning, this is not without risk and there are many ways a spread trade could lose money. Note also that everything we covered here assumes certain types of products. The spread price formula above does not apply, for example, to options.
So, now that you understand the fundamentals, what’s a typical spread trade range, how long is “soon enough”, and how can you figure these out? How can you trade the spread on Kollider against Okex or Bitfinex? And what about fees, funding, and futures expirations? Does market making help increase profitability in any way? Alas, that’s a story for another day and is out-of-scope for today’s “Long Story Short”. We’ll leave those as exercises for our ever-curious readers for now.