According to the American Gaming Association, more than 40 million Americans fill out an NCAA tournament bracket each year, blowing through more than $2 billion in the process.
While the vast majority of us do it for fun, with no real expectation of winning, that doesn’t mean we can’t take a few steps to improve our odds.
With that in mind, we present our 7 Steps to Win Your Office March Madness Bracket Pool (or at least give yourself a slightly better chance, if you prefer less braggadocio).
1. Embrace Game Theory
The first step to filling out a winning bracket is recognizing that picking the optimal bracket has nothing to do with how much you know about college basketball. It’s nice that you stayed up past midnight ET watching ESPN’s Big Monday every week, and that you can tell us all about trendy NCAA Tournament picks like Murray State and San Francisco. But let’s be honest: no one cares. And it’s not all that helpful.
Instead, filling out a great bracket is an exercise in game theory. The two factors that drive contest results are accuracy and payoffs. You want accurate picks, but perhaps even more important, you want to maximize your reward when those picks hit.
For example, let’s say Gonzaga has a 30% chance of winning the tournament. But out of the 100 entrants in your office pool, 40 have chosen Gonzaga as the winner, meaning the Zags have been chosen at a disproportionately high frequency relative to their actual odds.
In the 30% of scenarios where Gonzaga wins, you only have a 2.5% chance (1 in 40) of overcoming the other entrants who picked the Zags, which means you would have a less than 1% chance of winning the pool. Considering Gonzaga’s status as tournament favorite, those aren’t very strong odds of winning your pool, are they?
Now, let’s say Houston has a 10% chance of winning the tournament. Sure, the Cougars are only a third as likely to win it all as Gonzaga, but now let’s say only five out of 100 in your office pool have picked Houston, a disproportionately low frequency relative to the Cougars’ odds.
In the 10% of scenarios where Houston wins, you have a 20% chance (1 in 5) of winning the pool, which means you have a 2% chance of winning the pool – more than double your odds had you stuck with Gonzaga.
In the scenarios above, you would be accurate the most often by picking Gonzaga, but you would maximize the payoff by selecting Houston. That’s game theory in a nutshell.
2. Leverage Data
And that brings us back to the accuracy component of bracket-building game theory for a moment.
To maximize the accuracy of your picks, the easiest play is to leverage data that already exists. I prefer using Kenpom’s NCAA probabilities but there is a smorgasbord of sites that will do the trick. Most of these sites offer projections that are in the same ballpark as each other. (Probabilities that stray too far from the consensus should be viewed with a healthy dose of skepticism.)
Taking a glance at Kenpom’s probabilities, we see that Gonzaga is indeed the favorite, projected to cut down the nets 27.5% of the time, or a little bit less than what we used in our hypothetical scenario above. Not only are the Zags the favorites, they are three times more likely to win it all than the team with the next-best odds, Arizona at 8.9%.
The other teams with odds that are 5% or better include No. 1 seeds Kansas (6.6%) and Baylor (6.4%), No. 2 seeds Kentucky (6.2%) and Auburn (5.6%), No. 3 seed Tennessee (5.1%) and above-mentioned Houston, a No. 5 seed with a 5.0% chance of prevailing.
3. Understand Public Tendencies
Having a good set of tournament probabilities at your fingertips is great, but the fun really begins once you compare those projections to the public’s appetite for each team.
ESPN’s free bracket contest, Men’s Tournament Challenge, annually attracts more than a million entries. And even better than that, the website offers a Who Picked Whom tab that shows a breakdown of the percentage of participants who selected each team to win in each respective round.
This affords you an opportunity to hunt like a value investor to identify overvalued and undervalued teams and construct your bracket accordingly.
For instance, the public is picking Gonzaga nearly 30% of the time, which is a frequency that slightly outpaces the Zags’ actual chances of winning, which Kenpom pegged at about 27.5%. Remember, in order to maximize your chances of winning an office pool, your goal should be to optimize for payoff.
The next most-favored team by the public is Arizona at 12.2%. Like Gonzaga, the public seems to favor the Wildcats at a rate that outpaces the Wildcats’ true odds, which are closer to 8.9%. The jump from 8.9% to 12.2% doesn’t feel like much, but if you do the math, it means that Arizona is being picked 37% more often than it should be.
If the favorites are overvalued, who does that leave us with? More on that in a minute.
4. Know Your Competition
While ESPN’s data will help you understand trends at a national level, it is imperative to go one level deeper and consider the tendencies of your immediate competition.
If you live in Metro Detroit, you will want to avoid Michigan and Michigan State, as they are near certainties to be selected at a frequency far disproportionate to their actual odds, making them a poor value.
Since UM and MSU aren’t exactly favorites to win it all this year anyway, you may want to consider the likelihood that your officemates will extend that bias to other Big Ten powers, such as Iowa and Purdue.
If you live in the Southeast, you will have to think about the allegiances of your colleagues in a way that helps you understand if there is likely to be an SEC or ACC bias. Same in Texas with the Big 12 or on the West Coast with the Pac-12. You get the idea.
The other important factor with respect to competition is the size of the pool. If you’re in a small pool – let’s say fewer than 100 entries – sticking mostly with favorites might not be a bad idea. However, as the size of the pool grows, your appetite for taking a deeper sleeper should increase.
For instance, if you take Gonzaga and the Zags win it all, emerging from a pack of 30 officemates to post the highest overall score is less daunting than emerging from a pack of 300,000 in a pool that has a million contestants.
5. Pick Your Champ First
Although it’s natural to start with the first round and work toward the middle of the bracket, it is best to identify your champion first. By starting here, you can ensure that you have vetted and selected a winner with a desirable payoff.
In the 2022 NCAA Tournament, here are several teams that have attractive potential payoffs:
Houston Cougars
- Odds of winning: 5.0%
- Frequency among public: 1.3%
- ROI: 285%
UCLA Bruins
- Odds of winning: 3.8%
- Frequency among public: 1.7%
- ROI: 124%
Iowa Hawkeyes
- Odds of winning: 3.5%
- Frequency among public: 1.6%
- ROI: 119%
Texas Tech Red Raiders
- Odds of winning: 4.0%
- Frequency among public: 1.9%
- ROI: 111%
UConn Huskies
- Odds of winning: 0.7%
- Frequency among public: 0.5%
- ROI: 40%
Auburn Tigers
- Odds of winning: 5.6%
- Frequency among public: 4.4%
- ROI: 27%
Tennessee Volunteers
- Odds of winning: 5.1%
- Frequency among public: 4.3%
- ROI: 19%
Baylor Bears
- Odds of winning: 6.4%
- Frequency among public: 5.4%
- ROI: 19%
On the flip side, there are several teams you may want to avoid selecting as your champion, including big names like Gonzaga, Duke, Kentucky, Arizona, Kansas and Wisconsin. It is no surprise that the public loves the blue bloods, and that is a tendency you need to weaponize.
Although we cited Houston earlier in my mock scenario, and then validated that the Cougars are indeed a game theory-friendly selection, we are going to stay in the Lone Star State and go with Texas Tech as our national champion.
We like that Texas Tech is in the bottom half of the West Region with a Duke team that feels like it was overseeded by a couple spots as a parting gift to Coach K in his final tournament. Should the Red Raiders and Blue Devils meet in the Sweet Sixteen, Texas Tech projects to be a small favorite over Duke.
By being in the bottom half of the bracket, Texas Tech would avoid top-seeded Gonzaga until the Elite Eight, which gives us some extra time to see if someone else can knock the Zags out and clear the way for the Red Raiders.
Houston, on the other hand, would likely draw a very strong Illinois team in the second round, then top-seeded Arizona in the Elite Eight. With a couple other game theory-friendly picks in Tennessee and Villanova lurking in the bottom half of that bracket, the South Region feels like it has one too many land mines for our tastes.
6. Work Backward
After selecting your national champion, you want to work backward and select the runner-up and fill in the rest of your Final Four. If you’ve already picked a game theory-friendly national champion, it’s OK to play it more conservatively and stick with favorites to complete your Final Four.
In a deeper pool – think a couple hundred entries or more – it doesn’t hurt to continue using game theory for each of your Final Four selections, as you’ll need to exploit as many edges as possible to emerge from a larger crowd.
With Texas Tech already penciled in as our West champion, we are going to go with Tennessee in the South, Kentucky in the East, and Auburn in the Midwest. That’s one No. 1 seed, one No. 2 seed and a pair of No. 3 seeds, which seems reasonable. (If you’re a typical SEC-hating Midwesterner, you probably want to bleach your eyes after seeing those picks.)
Much like our Texas Tech pick, Tennessee and Auburn appear to be stronger than the public’s perception of them, and while Kentucky doesn’t offer the same value, the Wildcats still own the best odds in the East, and they satisfy our desire to pick at least one No. 1 seed to reach New Orleans.
Since we live in the Midwest, these selections also allow us to steer clear of the Big Ten bias that we assume will cloud some of our peers’ judgement. (Hey, it’s been 22 years since a Big Ten team won the national title anyway, so history is on our side anyway.)
7. Forget About Cinderella
Now that we’ve picked a Final Four and a national champion, we can keep working backward. As we fill in the rest of our bracket, keep in mind that as enticing as it can be to pick upsets, the juice usually isn’t worth the squeeze.
While being the only person in the office pool to have 16-seed UMBC toppling No. 1 Virginia or the only person to have Florida Gulf Coast dunking its way to an improbable Sweet Sixteen is fun for bragging rights, the odds are infinitesimal and the edge you gain on your competition is minimal.
In the unlikely event that you do nail an improbable early round upset or two, those Cinderellas are often easily discarded in the next round, limiting your ability to score major points, especially since the early rounds are only worth a point or two in most formats.
Now that doesn’t mean you have to stick entirely with high seeds, but you want to take sensible risks. For example, there are several double-digit seeds that are either Vegas favorites or small underdogs in the first round, including Michigan, Loyola Chicago, Virginia Tech, San Francisco and Davidson.
Again, if your goal is bragging rights, by all means, have at it with the 13-, 14- and 15-seeds. But if your goal is to maximize your chances at winning the pool — as un-American as it may sounds — it is best to forget about Cinderella.
With that in mind, here is a recap of all my picks by round for the 2022 NCAA Tournament:
First Round: Gonzaga, Memphis, UConn, Arkansas, Alabama, Texas Tech, Michigan State, Duke, Baylor, North Carolina, Saint Mary’s, UCLA, Texas, Purdue, San Francisco, Kentucky, Arizona, Seton Hall, Houston, Illinois, Michigan, Tennessee, Loyola Chicago, Villanova, Kansas, Creighton, Iowa, Providence, Iowa State, Wisconsin, USC, Auburn.
Second Round: Gonzaga, UConn, Texas Tech, Duke, Baylor, UCLA, Purdue, Kentucky, Arizona, Houston, Tennessee, Villanova, Kansas, Iowa, Wisconsin, Auburn.
Sweet Sixteen: Gonzaga, Texas Tech, Baylor, Kentucky, Arizona, Tennessee, Iowa, Auburn.
Elite Eight: Texas Tech, Kentucky, Tennessee, Auburn.
Final Four: Texas Tech, Auburn.
Championship: Texas Tech.
