Tiered Tournament Brackets

I’ve never been a fan of conventional tournament brackets. They are essentially set up as coronation runs for the top seed. The top seed gets a less arduous road because they play the lowest-ranked available opponent. As a side effect of this philosophy, the regional brackets with the #1 and #4 overall seeds are generally considered easier than the brackets with the #2 and #3 overall seeds.

Some might say to populate the entire bracket randomly, so that there are no “favored nations”. I’m not a big fan of this method because it diminishes the season-long accomplishment of the best teams. I think the best course is an meeting of the two strategies: a little bit of favored nation combined with a bit of randomness to make things interesting.

I suggest a tiered system. Tier A has the top four teams; Tier B, the next four; Tier C, the next eight; Tier D, the next sixteen; and Tier E, the final thirty-two. Within each tier, we place each team into their brackets based on random selection.

How does it work on an actual bracket? The actual positions (though not necessarily the seedings) are the same as now. The Tier As are placed in each region’s #1 position, which region being randomly chosen. The Tier Bs, likewise, are randomly placed at the #2 position. The Tier Cs get the #3 and #4 positions. The Tier Ds the 5-thru-8 positions, and the Tier Es randomly fill in the rest of the field.

In the first round, every team from the top half of the draw will face a lower half team. This means that the #1 overall seed could face an opponent ranked at either #64 (as now) or as high as #33. While it should still be an easier game regardless, at least there is the hope of competition. Looking at it from the other side, the #64-ranked team in the tournament could face an upper-tier team ranked as low as #32. While this doesn’t exactly change the 64’s underdog status, at least it gives them a fighting chance to advance past the first round.

After the first round, things tend to shake out as usual. Top schools will likely be playing each other in the final three rounds. A “Cinderella” team will make a heck of a run every now and again. In the end, though, the cream should still tend to rise to the top…at least in this system there is more of a chance that everyone will have been tested.

As for game locations… since the evolution is to pre-selected sites, there is no reason why, if the highest seed in the initial site group is a host site that they can’t stay at home. Think of it as another perk of getting a high seed. On the other hand, if you want things to be as neutral as possible, then the sites should be pre-selected before the bracket is drawn and let the schools land where they may.

Lastly, there is always the concern that schools out of the same conference will end up in the same region in an unbalanced way. Economics might encourage some committee massaging of the places, but I figure that if its fated that teams from the four most represented conference all find themselves in their own conference-heavy regions, who are we to judge that it’s wrong?

Realistically, if this unlikely event were to happen, there would be a huge outcry. There could possibly be some leeway for calculating a new bracket, but I find that once again tilts toward a perception of malfeasance. Trust that the tiers are separating the top teams from meeting too early. After that, let the chips fall where they may.

So, what sort of pairing would this method come up with? I dashed off a little program that lets you see for yourself. Just hop on over to my Tiered Bracket page and give it a test.