The KOS effect in practice
It is this KOS property of the network that you can wonder whether it is used enough. Of course, the success of Marktplaats has everything to do with the KOS effect. The chance that you will not succeed on Marktplaats is very small, unless you have a very exotic question (Wanted: politician with decency. We had one, but it was taken off the shelves).
A company is looking for money, a former bank employee has too much money, and under the auspices of the bank and under the responsibility of the parties, an attractive agreement is made for everyone.
In a large umbrella network of brokers, a solution must be found for the current deadlock in the housing market. Everyone waits to buy until their own house is sold. But in a large network, there is a good chance that you will complete a brother cell phone list circle between, for example, four homeowners who simultaneously sell to each other or buy from each other. A buys the house from B, B from C, C from D and he buys the house from A. Should be doable. The bank comes to the rescue for small differences, a notary does the transaction in a slightly longer stroke of the pen and the broker also uses adjusted rates for this series deal. Everyone is satisfied and the housing market is pulled back on track.
The KOS power of a network is much greater than we realize. The condition is that it is organized and that people are not fragmented over multiple networks.
The proof
In a network of 2 people, one connection is possible:
In a network of 3 people, three connections are possible:
In a network of 4 people, six connections are possible:
The formula is as follows: in a network of n people, n(n-1)/2 connections are possible. So: in a network of 23 people, 253 connections are possible.
The probability that two random people have their birthdays on the same day is 1/365 (we do not consider leap years). The probability that two random people do not have their birthdays on the same day is therefore (1-1/365), = 364/365. You have to multiply this probability 253 times – for the network of 23 people with 253 connections. I gave this formula to Wolfram Alpha, and it came up with the following result:
So the probability that none of the 23 people have their birthdays on the same day is 49.95228%.
This means that the probability that two people do have their birthdays on the same day is (1-49.95228), which is more than 50%.
This column was also published in Het Financieele Dagblad .