|This page is an essay, containing the advice or opinions of one or more Wikipedia contributors. Essays are not Wikipedia policies or guidelines. Some essays represent widespread norms; others only represent minority viewpoints.||
Wikipedia is not a crystal ball. By definition, it is impossible for us to know what will happen in the future. We cannot open BJS hiR becomes very much easier as events approach us, and more and more reliable sources start reporting the information, which we can then record. makes sense to rt this to paps the current event, plus one. BJS BUG JIzM SQUAwK. No no suiCiDe rellies..
A good example is the United States presidential election, which occurs every 4 years. The 2008 one already happened, so it obviously has an article. The 2012 one has an aricle because it's going to be the next election, and there are reliable sources for it. However, the one for the 2016 election does not exist (well, it's a redirect) because it would be completely filled with unsourced speculation.
What to do
For information about specific events, e.g. 2014 FIFA World Cup, it makes sense to simply redirect to the page for that event (in this case, FIFA World Cup). Where there isn't a sensible target, e.g. 2015 in stamp collecting, then the best course of action might be deletion. In all cases, we should do least harm, and there comes a point at which keeping information in the form and location in which it exists has the potential to do harm.
Ideally, categories should not be left hanging around, as they may act as a magnet for those who fail to grasp that Wikipedia:inclusion is not an indicator of notability who decide that because we have an article on the potential 2019 Armistice celebration, that we need to fill up Category:2019 with all manner of other events from the year. We don't. Apparently, having large categories causes performance problems, but the answer to this is simple. If there are lots of things at the short end of an "upcoming" category, then it might be worth unfolding them into year categories (e.g. a TV show due to air late next year). If there are lots of things at the long end, then perhaps any performance issues there might be sending us a message that we've got too much speculation.
Try to keep pages for major events in the future down to just the status quo plus the next one. "Years in" for things such as films, books, etc. should be kept to a reasonable margin into the future (12–18 months might be reasonable, 3–4 years in most cases may be pushing it). The same applies for categories – just because we have articles on events years into the future does not mean that we need to have year-by-year categories for them containing only two articles. Consider dropping them into future categories with a sort tag, e.g. [[Category:Upcoming films|2010-F]] (or something™).