Seed Distributions for March Madness 2018: A Tool for Bracketologists

Bracketodds: Our Story

Welcome to Bracketodds. What started (in 2007) as a simple inquiry on how seeds advance in the NCAA Men's basketball tournament has grown into a STEM (Science, Technology, Engineering, and Mathematics) Learning Laboratory in Computer Science at the University of Illinois at Urbana-Champaign. Both undergraduate and graduate students have been involved in conducting research, analyzing data, and shaping this web site (first launched in 2011) into an informative tool for March Madness neophytes all the way up to skilled bracketologists. The site retains its academic focus, by providing data and information to help everyone better enjoy the tournament, and answer questions as the outcomes of the games unfold. Thank you for helping make Bracketodds a mainstay of March Madness. Let the games begin!

Let us Fill in a Bracket for You

Not sure which 5-12 upset to pick? Which No. 13 or No. 14 seed will pull an upset in the first round? How many No. 1 seeds should be in your Final Four? Click here. and we will fill in a bracket for you based on the bracketodds analytics model. This model samples from the 9,223,372,036,854,775,808 possible brackets, in the appropriate proportions, based on historical performance of the seeds. You will notice that many of the brackets generated have a No. 1 seed winning the National Championship. That is because No. 1 seeds have won the National Championship almost 60% of the tournaments since 1985 (19 of 32). We do not attempt to predict who will win, like other websites or sports pundits; we just sample across the 9+ quintillions possible brackets. We will fill in the 64 teams (the First Four winners are added as they become available), so you can print out your own analytically designed bracket.

Like purchasing lottery tickets, most tickets do not win, but a few do. Our model samples across the 9,223,372,036,854,775,808 possible brackets in a manner that is representative of what has occurred in the tournaments since 1985. Using the lottery analogy, this is like knowing which numbers come up most often, and picking such numbers in the right combination to enhance the likelihood of matching the winning combination (which for the tournament, is the winners of the 63 games). Therefore, to get the best results requires you to consider a basket of brackets, which from among that set, you may find a few that will end up scoring well. If you have a particular team that you think will win the national championship, keep sampling until you find a bracket with that team as the winner. We do not guarantee a winning bracket; just an appropriate mix of upsets and favorites to make you look like a true bracketologist. Each screen refresh will result in a unique bracket. If you end up with a superb bracket, do let us know at .

Tips for Building Your Bracket

Click here.

Perfect Rounds and Perfect Brackets: An Inside Look (2018)

Click here.

Round of 64: Graveyard for Busted Brackets

Click here.

Upsets in the Round of 64

Click here.

How Far Does Each Seed Advance in the Tournament?

Click here.

Seed Projections for 2018

Click here.

Power Conferences Versus Mid-Majors

Click here.

Why Pick Favorites?

Click here.

Select a Round of the Tournament:

Tournament Background

Under the current format of the NCAA Men's Basketball Tournament, 68 teams are selected to compete in a single elimination format1. The Selection Committee seeds each of the teams according to factors such as their regular season records and their performance in their respective conference tournament. The tournament bracket is structured into four regions, each containing 16 teams. Narrowing the field to these 64 teams requires eight teams to participate in four play-in games, termed the First Four.

In the round of 64, each seed No. 1 plays seed No. 16, seed No. 2 plays seed No. 15, and so on, through seed No. 8 who plays seed No. 9. The winner of each game advances in the tournament, with the possible sets of seed match-ups in the round of 32 given by {1,8,9,16}, {2,7,10,15}, {3,6,11,14}, and {4,5,12,13}, corresponding to the four games in each region. The possible sets of seed match-ups in the Sweet Sixteen are given by {1,4,5,8,9,12,13,16} and {2,3,6,7,10,11,14,15}. In the remaining rounds of the tournament (the Elite Eight, the National Semifinals, and the National Championship), the possible sets of seed match-ups include the entire set of 16 seeds.

The winners of each region are designated as the Final Four, where the National Semifinals and National Championship determine the NCAA Men's Basketball Tournament champion over the final weekend of play. While any seed combination reaching the Final Four is possible for these final three games, the probability of each seed combination is not uniformly distributed.

Probabilistic Analysis

Using data from the past 33 tournaments (1985 through 2017), prior seed match-ups and winners can be used to identify a distribution that models the probability of certain seed combinations playing in each round of the tournament. This is accomplished by determining the frequency that each seed reaches a given round, then fitting this data to a truncated geometric distribution2, a nonnegative discrete random variable formed by the number of independent and identically distributed Bernoulli random variables, with success probability p (defined as the probability that the higher seed wins a particular game) that occurs until reaching the first success. Given that the NCAA tournament is single elimination, then a team cannot reach a round unless it has won in all the previous rounds.

Peer-Reviewed Research Papers

1 Jacobson, S. H., King, D. M., 2009, “Seeding in the NCAA Men’s Basketball Tournament: When is a Higher Seed Better?” Journal of Gambling Business and Economics, 3(2), 63-87. Click here for pdf.

2 Jacobson, S. H., Nikolaev, A. G., King, D.M.,  Lee, A. J., 2011, “Seed distributions for the NCAA Men’s Basketball Tournament”, OMEGA, 39(6), 719-724. Click here for pdf.

3 Khatibi, A., King, D. M., Jacobson, S. H., 2015, “Modeling the Winning Seed Distributions of the NCAA Division I Men’s Basketball Tournament”, OMEGA, 50(1), 141-148. Click here for pdf.

4 Dutta, S., Jacobson, S. H., Sauppe, J.J., 2017, “Identifying NCAA Tournament Upsets using Balance Optimization Subset Selection”, Journal of Quantitative Analysis in Sports, xx(x), xxx-xxx. Click here for pdf.

5 Dutta, S., Jacobson, S.H., 2017, “Modeling the NCAA Basketball Tournament Selection Process Using a Decision Tree”, Journal of Sports Analytics, xx(x), xxx-xxx. Click here for pdf.

About BracketOdds

Related Links


The authors of this web site would like to thank Douglas King (University of Illinois at Urbana-Champaign), Adrian Lee (CITERI), Alexander Nikolaev (University of Buffalo), and Arash Khatibi (University of Illinois at Urbana-Champaign) for their comments and feedback on this web site.

Web Developers in 2011-2012: Ammar Rizwan, Emon Dai (Students, Department of Computer Science, University of Illinois at Urbana-Champaign)

Web Developer in 2015: Andrew Yang (Student, Department of Computer Science, University of Illinois at Urbana-Champaign)

Contributors in 2016-2017: Arash Khatibi, Kevin Li, Daniel Zurawski (Students, Department of Computer Science, University of Illinois at Urbana-Champaign)

This is a student-driven project supervised by Professor Sheldon H. Jacobson (Department of Computer Science, University of Illinois at Urbana-Champaign).


For comments and feedback, send an email to .

Follow us on twitter: @shjanalytics.

Media requests? Send an email to  or call 217-244-7275.

How You Can Help

If you would like to support the ongoing student development activities for this project, tax-deductible contributions can be made to the Department of Computer Science at the University of Illinois. Under "Your Gift. Your Choice", type "Bracketodds Project". THANK YOU for your support. ©  2018 - The Board of Trustees of the University of Illinois.
All Rights Reserved.

free web tracker