Before reading this article, you should have already read the article about investor divergence and have learnt what investor divergence is and what its causes are.
Divergence due to investment volume can be seen when going to a DARWIN's page. The main information, the % of current monthly divergence, can be seen as follows:
When divergence is below 0%, investors will most probably have a lower return than the theoretical return of the DARWIN and when monthly divergence is above 0%, investors will most probably have a higher return than the DARWIN.
In this specific case, monthly divergence is 0.04%. What does this mean? It means that, in case of investing in this DARWIN today, your return in the first month would be probably 0.04% percent higher than the DARWIN's notional return.
Estimated monthly divergence only gets calculated for DARWINs with investors and only if at least 10 orders have been placed for investors.
Now let's see how this % of monthly divergence gets calculated
For this, we'll go to the Divergence tab.
On the one hand, you can see a graph showing a simulation of return divergence over different past timeframes. On the other hand, you can see a graph showing divergence per order.
Let's have a closer look at the graph showing divergence per trade
Here we have one axis showing divergence in pips and the other axis showing latency in seconds.
Please note that from 0.4 seconds onward, the scale doubles with each step: after 0.4 comes 0.8 and then 1.6 seconds and so on.
The dots on the graph are the last 100 orders understanding the opening and the closing of a trade as two separate orders. When less than 100 orders have been placed for investors, the maximum number of orders made is shown. Green dots are orders with positive divergence, that is, investors are getting a better price than the trader, and red dot orders with negative divergence, that is, investors are getting a worse price than the trader. Order size is reflected in the size of the dots.
In the table next to the graph, you can see the summary of the information the graph contains:
- The number of orders
- The average order size
- The average divergence in pips
- The median divergence in pips
- The average latency in seconds
- The median latency in seconds
- The numbers of orders, inside the 100 orders we are looking at, which were replicated with a latency of less than 0.4 seconds or 400 milliseconds
- The average divergence in pips and the median divergence in pips of these shortlisted orders
Now, the question is, why is there a shortlist of orders executed in less than 400 milliseconds?
The reason to group the orders in those executed in more than 400 and those executed in less than 400 milliseconds is the assumption that those executed in more than 400 milliseconds are having divergence due to latency. Latency is a technology problem and we can't have any influence on it beyond optimizing our systems. Looking at a sufficient number of orders, divergence due to latency will be aleatory and equaled out on the long term.
For the other group, the one of the orders executed in less than 400 milliseconds, we make the assumption that divergence is due to investment volume. When trades replicated for investors move a large volume, that is, have a large lot size, the best, top of book spreads are sometimes no longer available which gives rise to little price differences between the trader's trade and investors' replicated trades.
It is this second type of divergence we are interested in when trying to calculate monthly divergence. So, from the average divergence in pips, based on the shortlisted orders, we calculate the DARWIN's monthly divergence.
Simulation of divergence over longer periods
On the other graph, called Divergence in %, you can see a simulation of divergence based on the current monthly divergence %.
You can see:
- The DARWINs return, and
- Investors estimated return considering current monthly divergence
- And total estimated divergence in the selected period
When looking at longer periods, period divergence gets higher. The longer the period, the higher the divergence.
Does it affect divergence if an investor trades in an out all the time in a DARWIN?
Investors trading in and out all the time in a DARWIN do not affect the estimated monthly divergence of the DARWIN. Only trades catalyzed by the DARWIN are included in the calculation of monthly divergence.
Trades catalyzed by the investors by buying and selling the DARWIN while the DARWIN is in an open position, are not represented by the green and red dots of this graph.
As a takeaway from this article, you should have learned that before investing in a DARWIN it is good practice to check out current monthly divergence. A current monthly divergence below -0.5% can result in significantly worse results for investors than the DARWIN's notional return.