Pitchfork is running a bracket to determine the best album of the last 30 years, and the whole exercise is causing the predictable mix of engagement and irritation across music discussion spaces. Some people love it. Some people find it reductive. Some people are angry about specific matchups. Everyone has an opinion, which is the point.

What brackets like this reveal, more than any definitive answer they could produce, is how strange the process of canonization actually is. We act as though the canon is something that gets discovered, that a record like OK Computer or Illmatic or Loveless eventually earns its place in the permanent record through accumulated critical consensus and the passage of time. But the reality is that canons are made, not found. They are the product of which critics were influential at which moments, which platforms had the loudest voices, which audiences showed up in sufficient numbers to matter, and which records happened to be reviewed in the right year by the right person at an outlet with enough reach to shape the conversation.

None of that makes the canon wrong. OK Computer is genuinely as good as its reputation suggests. The point is that dozens of other records from the same period are also genuinely that good, and they are not in any bracket because they did not circulate through the right networks at the right time. The brackets do not reveal what was best. They reveal what got canonized, which is a different and more sociologically interesting question.

There is also something worth examining in the 30-year window specifically. Pitchfork was founded in 1996, which means that the frame of this bracket aligns almost exactly with the lifespan of Pitchfork as an institution. The records that land highest in these exercises tend to be the ones that Pitchfork reviewed most favorably at their moment of release, or that achieved retroactive elevation through repeated reference in Pitchfork’s subsequent coverage. This is not a conspiracy. It is just how cultural institutions work. They build taste networks and then those taste networks become self-reinforcing.

The albums that typically get excluded from brackets like this are the ones that were popular but not critically championed, the ones that were critically championed but in genres that the outlet’s core readership does not value, and the ones that emerged from scenes or demographics that the institution spent years treating as peripheral. A 30-year retrospective that does not interrogate those exclusions is not really a history. It is a mirror.

That said: the bracket is fun. The matchups generate genuine arguments about things worth arguing about. The act of ranking forces people to articulate why they value one thing over another, which is a useful exercise even when the rankings themselves are absurd. Music criticism at its best makes the case for paying attention, and brackets make people pay attention to records they might have dismissed or forgotten.

The problem is when the bracket becomes the conclusion rather than the starting point. When the winner gets treated as settled rather than as one data point in an ongoing conversation, the bracket has done the opposite of what good criticism should do: it has closed down rather than opened up.

The 30 best albums of the last 30 years do not fit in a bracket. They barely fit in a list. That is not a failure of the format. It is the nature of the period, which produced more genuinely significant music than any equivalent span in recorded history. Treat the bracket as an invitation to argue, not an authority to defer to.