Cohort-based analysis. Part 2: Realised price partitioned by HODL cohorts

6 min readFeb 8, 2021

Continuing our cohort-based analysis series, we will be diving deep into HODLers this time, and find out what is the average price HODLers have originally purchased their BTC for.

To do that, we will begin by showing how large volumes of HODLed bitcoins make BTC price rise and how a particular set of market players are responsible for it. After that, we will be able to find out the average price of these HODLed Bitcoin that was responsible for the bull runs so far.

Who starts the bull run?

Bitcoin has two types of players that are the most interesting to look at: whales and hodlers. Each type of player can be split into cohorts which is what cohort-based analysis about. Whales can be split by the size of their wallet, for example, >10k BTC wallets are a different cohort compared to 1k-10k BTC wallets. HODLers can be split in a similar fashion: HODLers that have been holding for 5–7years, 1day-1week and so on.

The first instance this type of splitting was publicly released for HODLers was thanks to unchained capital which have calculated the number of BTC that belongs to each hodler type. This is shown in the picture below.


What is interesting about this metric is that Bitcoin price has a tendency to rise after a cycle of HODLer accumulation has occurred and these cycles can be noticed on the HODL wave chart:

screenshot from the original unchained capital article

We can also notice that the origin of each cycle is accompanied by a sudden rise in the amount of BTC being transacted, i.e. spike in the <1d band.

As time goes on, part of this transacted BTC is not moved and hence goes into a higher band, i.e. coins from the<1d old band go to 1day-1week old band if the BTC is not touched. If a lot of players decide to not touch their BTC after transacting it, a HODL wave is formed.

The question we want to answer in this article is “at what price did the original HODLers that were responsible for the HODL wave acquire their Bitcoin”

Realised price split by HODLer cohort

For the people who have not already had a look at the previous article, I highly recommend doing so as many concepts talked about in this article are explained there.

Here is a link to it: Realised price (part 1) but here is a short description of what realised price is as well.

Realised price = Realised Cap / Circulating supply of Bitcoin

i.e. Realised price is the average price that the circulating supply of Bitcoin was obtained for by wallets on the blockchain.

Also, I have to mention that all articles that have come out and that will be coming in the future will be as comprehensive and intuitive as they can be. This series of articles will not use complicated language and unnecessary jargon and I believe that is how on-chain analysis should be presented. In reality, we are asking a very simple question: what is the average price HODLers bought their BTC for, and thankfully, we can find the answer.

Going straight into it, in a similar fashion how we found realised cap for address cohorts we can find realised cap for hodler cohorts. Again, I highly recommend having a read of our first article where the concept of realised cap is explained.

After that, dividing realised cap of each HODLer cohort by the number of Bitcoin that belongs to it, we can find the average price that a particular HODLer cohort has obtained its Bitcoins for, i.e.

Realised price split by HODLer cohort = Realised cap of a cohort / Amount of bitcoin that belongs to that cohort.

Data by whalemap

It is important to note that this chart can be used to perform a much deeper analysis of HODLer cohorts behaviour. One can look at what significance the crossovers of bands have on Bitcoin price, analyse the profits or losses a particular HODL cohort is making, or even use the bands as support and resistance.

In short, Realised price split by HODLer cohort contains a plethora of deeper insight which is all contained in one single chart. Hence, anyone interested in analysing this chart further can write to our official email:

In this article, we will touch upon finding the answer to only one question, what is the original price of Bitcoin that is responsible for forming a HODL wave (which are in turn responsible for bull runs so far).

Data by whalemap

In the picture above one can see how Bitcoins in the 1–3month HODL band are migrating to higher bands of 1–3months, 3–6months and so on. We can also see that the peak Realised price of the band is decreasing. This is due to Bitcoins leaving the bands as they get older, i.e. not all HODLers are actually hodling.

Nevertheless, by looking at the yellow 12–18 month band, one can find how much of the HODLers from the original BTC top at $29 are left in Oct 2012 and what is the average price this BTC was acquired for. These are in fact the HODLers who are actually hodling and are forming the HODL wave that Bitcoin price likes so much.

Here are two more examples of such realised price HODL waves with the one on the right showing our latest consolidation and the one on the left showing the origins of the 2017 bull run.

Data by whalemap

One can notice that as long as realised price stays at similar values (HODLers don't cash out), Bitcoin starts its bull run after it finds support at Realised price split by HODLer cohort bands.

This is one of the most important insights of this article.

Another way, HODLer cohort realised price can be used is by looking at the 1–3 month band where it usually provides support during bull runs and marks the beginning of a bear phase when the price gets below it.

Data by whalemap

Furthermore, by popular request, let us compare the 1— 3 month band to the 20-week moving average. Even though there are similarities, the HODLer cohort band is able to provide higher support, resistance and bear signal hit rate than the 20-week moving average. For example, Bitcoin price has fallen below the band earlier compared to the moving average and hence gave a bear signal before the moving average. Furthermore, looking at the metrics from a point of support and resistance, the 1 — 3 month band was able to serve as support and resistance much more often than the moving average. This can be noticed in the figure below.

Data by whalemap


Realised price split by HODLer cohort finds the average price that different types of HODLers have acquired their Bitcoin for. This article showed how the beginning of bull runs is usually accompanied by Bitcoin price finding support at the Realised price bands and as an introduction to the realised price concept, we have also discussed what are HODL waves and how they lie at the origins of historical Bitcoin bull runs. We have also found that the 1–3 month Realised price band was consistently providing supports during bull runs and signalled bear phases when Bitcoin price fell below it.

This finishes the second article from the cohort-based analysis series that the Whalemap team is working on right now.


Exploring blockchains through data science to find actionable insight