Bitcoin Gaps

It has been already more than two and a half years, since the Chicago Mercantile Exchange enabled the trading of Bitcoin futures. Famously, the day CME listed Bitcoin futures on December 18, 2017 also marked the first day after Bitcoin’s top at around 20k USD. CME is thereby not the only exchange that is trading Bitcoin futures, but in contrast to other cryptocurrency spot and future exchange trading products, the CME Bitcoin futures are not traded during the weekend and national holidays, which result in trading pauses or so called gaps in a chart. These gaps have an interesting impact on Bitcoin’s price, since they tend to get filled after a certain amount of time. Therefore, we will take a detailed look at the CME gaps in this article and analyze their impact on Bitcoin’s price.

First of all, we have to define what a CME gap is. You could find several definitions for these CME gaps, since every daily candle has not only an opening and closing price, but also a highest and a lowest price. The most common definition though, is the one that shows an actual gap in the chart. It looks at an opening price after a trading pause and compares it to the highest or lowest price on the day before the trading pause. If the opening price is either higher than the highest price or lower than the lowest price before the trading pause, it will result in a so-called CME gap.

After finding a definition for gaps, it is necessary to collect data to analyze the gaps’ impact on Bitcoin’s price. I used the historical trading data of CME Bitcoin futures provided by investing.com, which gave me a data sample of 813 days of trading data. Considering only the days with trading volume, 651 days resided in the data sample, that were used to identify CME gaps as well as data about filling these gaps.

As of today, 58 CME gaps can be found in Bitcoin’s historical trading data. 56 of those have been “filled”, meaning that Bitcoin reached the same prices on the CME futures exchange that it didn’t trade in previously on CME during a trading pause. Naturally, not every gap is filled within one day, there can be many partial fills which lead to an entire fill of a gap over the timespan of many different days.

The average gap size is 224.91 USD. 37 of the 58 gaps have been to the downside, meaning that the gap prices were below the opening price on CME after a trading pause. On the contrary, 21 of the gaps are to the upside, meaning the gap is above the opening price after a trading pause. It shows, that the price has been more likely to go up during trading pauses than it is to go down.

23 of the 56 fills have been “same-day fills”, meaning they were closed on the same day they appeared on the chart. On average, it took 13.9 days to close a CME gap. Excluding same-day fills, it took 23.6 days on average to fill a CME gap. So far, the longest it has taken to close a CME gap completely was 359 days, where a gap from 11,675 USD to 11,795 USD that appeared on August 12, 2019 was filled after 359 days on August 5, 2020.

Although filling a gap is certainly not guaranteed and the time it took to fill gaps has varied in the past, it cannot be ignored that the CME gaps have been one of the better indicators to predict Bitcoin’s price so far. With 56 of 58 gaps filled and all gaps closed from 2017, 2018 and 2019, trading strategies based on CME gap filling have been insightful in the past. Especially when you compare CME gap fills to my recent analyses of other indicators, such as the hammer candle and the daily RSI, that basically had not much predictive impact, it becomes obvious that the CME gaps are an indicator worth considering.