How strong are soccer teams? : the "host paradox" and other counterintuitive properties of FIFA's former ranking system

I investigate the paradoxes associated with the Fédération Internationale de Football Association (FIFA) point-based ranking of national soccer teams. The ranking has been plagued with paradoxes that incentivize teams to avoid playing friendly matches, i.e., matches that are not part of any official FIFA tournament or preliminaries, and applying other counterintuitive strategies. The most spectacular paradox was the dramatic underrating of the hosts of major tournaments. For a long time, host teams, which were absent from preliminary matches, would play only friendly matches that awarded few points. Here, I present three models that estimate the magnitude of the resulting "host effect" at 14.2-16 positions. Such an estimate counteracts the intuition that a large investment in hosting a tournament should result in an improvement in the host team's standing. However, as discussed here, a given host's low ranking could decrease interest in the tournament, and likely result in a major loss of advertisement revenue.