March Madness predictions: Here's how your bracket will do if you only pick the better-seeded team
Filling out a March Madness bracket can be challenging. There are dozens of statistics to research, 68 teams to examine, and only a few days to do it.
If you’re committed, we’ve got you covered with more than 50 stories focused on bracket advice in our BracketIQ hub.
But what if there were an easier way to fill out your bracket? One that required no research, no planning, no guesswork.
What if you picked a bracket where the better seed won every game?
|MARCH MADNESS SHOP|
We looked at data from our Bracket Challenge Game from 2011, and compared a bracket that picked using the overall seed rankings against the average brackets that year. Here’s how they fared:
|Year||Seed-based bracket score||Average BCG score||Difference|
Apart from 2015, seed-based brackets perform much better than average. On average, a bracket that picks based solely on the overall seed ranking will net 16.2 points more than the average bracket. Since each round is worth 32 points, that’s the equivalent of two correct Final Four picks.
Of course, two of these years stand out and somewhat skew the numbers.
Those would be 2012 and 2013, when the No. 1 overall seed won the championship. Those two (Kentucky in 2012, and Louisville in 2013*) are the only two No. 1 overall seeds to play in or win the title in the past 10 years.
There is a lesson to be found here for more involved bracket pickers: First round upsets are fun on paper, but knocking off top-seeded teams early and picking Cinderellas to make deep runs is a volatile tactic.
Here's how the seed-based brackets correct picks break down by round:
|Year||First round||Second round||Sweet 16||Elite Eight||Final Four||Championship game|
Take this, from our Lessons from Bracket Challenge Game winners:
“It's not worth stressing over trying to get the first couple rounds perfect. That's a nearly impossible task. But the early missed picks can turn out to be pretty inconsequential if you get the later rounds right. In general, it’s better to miss an upset than to predict one, be wrong, and have that winning team go on another two or three rounds – or even worse, to the Final Four.”
Back to the seed-based bracket.
We’ve established that you’ll do much better than average. But that doesn’t mean this is necessarily a good strategy to win your pool. For the four years we have full data for — 2014-2017 — the seed-based brackets placed in the 71st percentile on average.
The 2016 seed-based bracket does the best comparatively, placing in the 86th percentile, despite not picking the champion correctly.
Moral of the story: If you’re tired of being embarrassed in your pool by your 3-year-old nephew who literally can’t read a bracket, play it safe and try picking the better-seeded team this year. If nothing less than first place will do, add a little more analysis to your process.
*NOTE: Louisville’s participation in the 2013 Division I men’s basketball championship was later vacated by the Committee on Infractions.