As this year's IOI is around the corner, I have been considering where I stand and what to aim for this year.

Two metrics to quantify skill

First of all, I would like to mention how I think of skill when it comes to OI. Each competitor's skill is defined by two metrics: untimed skill and timed skill. I think of these similar to how CF problems are rated: someone's untimed skill level is the difficulty level at which they'd be able to solve 50% of problems given a very long time (say 5 to 10 hours, essentially unlimited). Timed skill is defined similarly, but under the time constraint of 100 minutes per problem (the average time that can be allocated per problem at IOI).

That being said,

The dynamics of gold medalists

Inspired by E869120's blog on winning a gold medal at IOI, I have put much thought into what it takes to achieve each kind of medal. When it comes to gold medalists, it seems like there are two main classes: those who were able to be fairly confident (>70% chance):about winning a gold medal going into IOI, and those who were confident about winning a silver, and had some chance (20~70%) at a gold medal. Just for a very rough estimate, I would say that those who rank 16th to 60th in raw skill fall under the second category, which I believe includes myself for this year.

For this group of competitors, playing the safe strategy of spending a roughly equal amount of time on each problem and grabbing as many subtasks as possible would be able to secure a silver medal. However, when it comes to the race for a gold, it is much more dependent on the problemset for that year, which naturally introduces quite a large element of chance.