NCAA Tournament Picks: Creating a Contrarian Bracket That Can Win Your Pool
Fred Kfoury III/Icon Sportswire via Getty Images. Pictured: The Villanova Wildcats.
There are many different ways to fill out a bracket this time of year.
You can dive deep into matchups, you can randomly select teams out of a hat, or you can pick your favorite mascots. This article will give you a better way to go about selecting teams with the ultimate goal of winning your pool.
How do you actually win your pool, though? Seriously, give this question some thought. The way I see it, there are two ways to end up victorious at the end of the tournament:
- Pick a lot of the games correctly (this is very hard).
- Pick teams that give you the most to gain over the field if they win.
Instead of trying to get all of the games correct, picking teams that most people aren’t picking gives us a better chance to win.
As an example, roughly 35% of brackets are picking Gonzaga to win it all. If Gonzaga does end up winning it all, you are competing with 35% of other people to win your pool, and it now depends on who got most of the other games correct.
Be Contrarian, Not Crazy
You want to avoid the most popular teams where you can, but this doesn’t mean taking all of the 16-seeds to win the first round either.
For this article, my process is looking at our Action Network probabilities from Sean Koerner and comparing them to how many people are taking each team to win each round.
When people are picking teams more than the probabilities suggest, that’s usually the time to pick an upset.
Build Inside Out
For most brackets, the scoring is doubled each subsequent round. This means that picking the champion is the most important part of your bracket because it’s worth the most points.
This may seem backwards, but I don’t care about the matchups along the way. I’m trying to maximize my leverage against everyone else in the pool with the picks that are worth the most points.
Let’s find a champion.
Gonzaga is the most likely champion per our probabilities at 24.7%, but over 35% of brackets are taking the Bulldogs, so we’re avoiding them.
Looking for championship value leads us to four options: Baylor, Michigan, Houston, and Iowa.
The Iowa Hawkeyes are being selected to win it all in just 3.5% of brackets and have an 8.7% chance of winning, which is the highest edge of any team.
For this bracket, I’m taking Iowa although any of the four options are acceptable. If you are doing multiple brackets, you should be selecting different champions each time and worrying less about changing first-round games you are unsure about.
The Final Four
Using the same method above, I can fill out the rest of the Final Four teams in my bracket. Houston is a 2-seed but is not being picked like a normal 2-seed. It makes a great contrarian play to make a run and meet Iowa in the championship game.
For the South Region, the choice came down to Arkansas, Purdue, and Wisconsin. Having a 9-seed Wisconsin make a deep run would certainly be a differentiating path, but Purdue offered the most pure value.
The East Region is the one time I considered taking the favorite. The 1-seed Michigan Wolverines are being picked just slightly above what our projections are predicting. The field is “over-picking” Alabama and Texas, which puts Michigan in play, but I ended up going with another 4-seed in Florida State.
Colorado is a good dark-horse play in this region.
The real question is how far to advance Gonzaga. In this scenario, I have it losing to Creighton in the Sweet 16. Creighton has the best value to reach the Elite Eight in the region.
For first-round games, I typically want to get at least a 5% edge to take a major upset, although that is an arbitrary number I’ve used with some success over the years. For 7/10 and 8/9 matchups, I’ll always take whichever teams are offering value, which in this case are VCU and Missouri.
This region really starts to fly off the rails.
As I said above, Alabama and Texas are popular picks, so it’s a good play to have them losing early. Maryland making a big run is stomach-churning, but the data liked either the Terrapins or UConn to make that same run.
If you do multiple entries, you may want to alternate those given how close the first game is projected to be.
The top two teams in this region are both knocked out in the second round by Wisconsin and Florida, respectively. People are picking Winthrop to upset Villanova too often, likely because of the history of the 5/12 matchup.
Utah State is one of the best value upsets in the entire bracket.
There is a theme of 1-seeds going down early if you had not caught on yet.
This time, Loyola Chicago plays the role of David to Illinois’ Goliath. Tennessee is the beneficiary as it gets an easier path to the Elite Eight. Note that I did not have a single 12-seed beating a 5-seed, and I’m perfectly fine with that.
We do have an upset here, as Liberty is only being picked 14% of the time but has roughly a 25% chance to take down Cade Cunningham and Oklahoma State.
Rutgers is being picked 40% of the time but should be favored, making it the best value play in the first round.
If this bracket looks ugly to you, that is a feature and not a bug.
Every time I get finished with one of these brackets, I usually hate the look of it, but the results have shown me this is the best way to increase your chances of winning.
This is March. Nobody knows what’s going to happen. So, the best way to win is to try to do something different than everyone else.