The 2018 NCAA tournament was one of the most unpredictable in years. Six double-digit seeds won in the first round. A 16 seed topped a 1 seed for the first time ever. An 11 seed made it to the Final Four.
All that led to some pretty poor performances in our Capital One NCAA March Madness Bracket Challenge, so don't feel too bad if you had an off year last March.
According to data pulled from the millions of brackets in the Bracket Challenge, the average score for 2018 was approximately 57 points. A huge drop from 2017, and the second-lowest average we've seen in the eight years we've run the game.
Here's where those eight years look like:
| Year | Average score |
|---|---|
| 2018 | 57 |
| 2017 | 65.7 |
| 2016 | 68.2 |
| 2015 | 83.3 |
| 2014 | 60.1 |
| 2013 | 70 |
| 2012 | 83 |
| 2011 | 53.1 |
As a refresher, this is how our bracket's traditional scoring works. Players get points for each winner they pick correctly. The points double in each round to reflect the number of teams left, like this:
| Round 1 | Round 2 | Sweet Sixteen | Elite Eight | Final Four | Championship | |
|---|---|---|---|---|---|---|
| Points per correct winner | 1 | 2 | 4 | 8 | 16 | 32 |
So the total possible points for a bracket in a given year — if a player got all 63 games correct — is 192 points.
Nobody ever picks every game correctly. In fact, we've found that on a whole, BCG players only get two-thirds of all picks correct. But a lot of those correct picks come in earlier rounds, when the differences between teams are more stark and correct picks are worth less.
BRACKET STATS: The absurd odds of a perfect NCAA tournament bracket put into perspective
Once you reach the Final Four, one game is worth the equivalent of 16 first-round games, yet less than half of one percent of all brackets in the past eight years have managed four correct Final Four picks.
Fewer than 0.1 percent of players last year — just 54 out of millions — picked all Final Four teams (Loyola-Chicago, Michigan, Villanova, and Kansas). No one picked all four in 2011, when eventual champion Connecticut (a No. 3 seed), Kentucky (4), title-game loser Butler (8) and VCU (11) played in Indianapolis. That's a big reason the average score was so low that year.
In 2013, no one got all the semifinal teams correctly. Louisville, Wichita State, Syracuse and Michigan made their way to Atlanta. While top-seed Louisville was probably the easiest selection of the four. The rest were much longer shots. Wichita State reached the Final Four that year for only the second time in program history. Syracuse hadn’t made it that far since they won the National Championship in 2003. Michigan was returning for the first time since 1993. Michigan and Syracuse were No. 4 seeds. Wichita State was a No. 9.
BRACKETOLOGY: The ultimate guide
The “easiest” bracket of the last seven years was 2015. A whopping 1.6 percent of people correctly predicted all Final Four teams that year. Wisconsin, which had made it to the tournament in each of the past 14 seasons; Kentucky, unbeaten entering the tournament; Duke, a 12-time Final Four participant; and Michigan State, in the semifinals for the seventh time in 18 seasons, advanced to Indianapolis. It was the first time since 2009 that multiple 1 seeds made it to the Final Four. Only Michigan State (a seventh seed) did not have the top seed in its region.
Filling out a bracket is easy. Filling out a good one is not.
