Yeah this uses average and a different type of calculation called the United Nations System of National Accounts (SNA 2008). They’re also the data the study in the article uses for its basis. From the graph info:

Average National income within the p0p100 percentile group. National income aims to measure the total income available to the residents of a given country. It is equal to the gross domestic product (the total value of goods and services produced on the territory of a given country during a given year), minus fixed capital used in production processes (e.g. replacement of obsolete machines or maintenance of roads) plus the net foreign income earned by residents in the rest of the world.
National income has many limitations. However it is the only income concept that has an internationally agreed definition (established by the United Nations System of National Accounts, see SNA 2008). So we use it as our reference concept (with tax havens correction). To be improved.The national economy - in the national accounts sense - includes all domestic sectors, i.e. all entities that are resident of a given country (in the sense of their economic activity), whether they belong to the private sector, the corporate sector, the governement sector. The population is comprised of individuals over age 20. The base unit is the individual (rather than the household). This is equivalent to assuming no sharing of resources within couples.

[National income]=[Net domestic product]+[Net foreign income]

WID.world estimations as a proportion of GDP based on the following; 1950–2021: estimated from other components. These estimates are then anchored to GDP (see GDP variable for details). The estimates of national accounts subcomponents in the WID are based on official country data and use the methodology presented in the DINA guidelines. We stress that these subcomponents estimates are more fragile than those of main aggregates such as national income. Countries may use classifications used are not always fully consistent with other countries or over time. Series breaks with no real economic significance can appear as a result. The WID include these estimates to provide a centralized source for this official data, so that it can be exploited more directly. We encourage users of this data to be careful and to pay attention to the source of the data, which we systematically indicate.

No idea how much that affects things, but it’s the official government numbers basically standardised across countries to try to factor in access to infrastructure, etc. Definitely not perfect because it still has tp be an estimate, but still at least based on the same official source.

Unless they genuinely fucked up, which is still possible.

But basically, most everyone you know is likely in the global top 15%-20% even if none of you look individually rich, because you all have access to a high standard of living.