2022 Bracket Picks Year in Review
We review the performance of our 2022 NCAA bracket picks, including what went right and wrong with our data-driven strategies and analysis.
2022 was a year in which several of the top favorites faltered in the NCAA tournament. No. 1 Kansas was the only team to win its region (the Midwest) as the pre-tournament favorite.
According to pre-tournament betting market odds, No. 1 Gonzaga, No. 1 Arizona, and No. 2 Kentucky were the favorites to win the other regions, yet none of those teams even made the Elite Eight. And although Kansas won the title as a No. 1 seed, the Jayhawks were not a relatively popular champion pick in bracket pools.
Combined with No. 8 North Carolina’s unexpected run to the championship game, 2022 ended up being a year when relatively lower scores were able to win bracket contests.
Picking Kansas as your champion and getting just one other Final Four pick right, combined with decent pick performance in earlier rounds, was likely good enough to win a small or even mid-sized pool.
TeamRankings Performance Overview
Though some of our brackets recommended for multi-entry strategies did feature Kansas making a championship run, none of our Best Brackets for traditional 1-2-4-8-16-32 scoring systems had Kansas as champion. Those Best Brackets also generally avoided strategic bets on eventual Final Four teams Duke and Villanova.
As a result, it was a down year for our bracket picks overall. After digging into the results, though, it’s hard to pinpoint any glaring deficiencies in our research process.
In this post, we’ll first dig a little deeper into how the tournament played out and what it took to win a bracket contests. Then we’ll take a look at what went right and wrong with our pre-tournament analysis and picks.
Overview of the 2022 NCAA Tournament
- Kansas (No. 1 seed, Midwest) completed an improbable second half comeback against North Carolina (No. 8 seed, East) to win the championship.
- The three other No. 1 seeds all failed to reach the Elite Eight, with Baylor losing in the Second Round, and Gonzaga and Arizona both falling in the Sweet Sixteen.
- North Carolina went on a strong run by beating both East No. 1 Baylor and a tough No. 4 seed in UCLA, before also ending Coach Mike Krzyzewski’s final season by beating rival Duke in the Final Four.
- One of the most memorable stories of the 2022 tournament was the deep run by Saint Peter’s, the first No. 15 seed to reach the Elite Eight, with wins over No. 2 Kentucky, No. 7 Murray State, and No. 3 Purdue.
- Most experts considered the ACC to be having a down year entering the tournament, but three of the five ACC teams in the field reached the Elite Eight, and two of them made the Final Four.
- In contrast, the SEC had six teams seeded on the top six seed lines, but No. 4 Arkansas in the West was the only one of the six to survive the first weekend.
What a Winning Bracket Looked Like in 2022
Based on how this year’s tournament played out, and especially because it wasn’t a great year for TeamRankings brackets overall, it’s informative to look at what did work from a picks perspective in 2022.
To do that, we analyzed the top 10 scores in a roughly 300-entry pool with standard 1-2-4-8-16-32 scoring.
(As you review the highlights below, keep in mind that this was not a small pool. So you could have made even more mistakes and still finished at or near the top of a 50-entry or 100-entry pool.)
- All of the top 10 brackets had No. 1 Kansas winning the national title.
- None of them had No. 8 North Carolina in the title game, or the Final Four, or the Elite Eight. Two had UNC in the Sweet 16 over Baylor, but losing after that.
- All of them had exactly one other Final Four team correct besides Kansas. Seven had Villanova out of the South, three picked Duke out of the West. None had three or four Final Four teams right.
- Six of them picked all No. 1 and No. 2 Seeds to the Final Four.
- Four of them had only one team worse than a No. 2 seed in their Final Four. Two had No. 3 Texas Tech, one had No. 3 Purdue, and one picked No. 4 UCLA.
- No. 2 Kentucky was picked to the Final Four in six brackets. Gonzaga was a Final Four team in five of them.
- On average, they picked only 1.7 teams seed worse than No. 5 to reach the Sweet 16.
- They didn’t hit a very high number of early upsets, or dominate the First Round. In fact, they averaged 21.6 wins over the first two days, which was lower than the overall public average.
That Strategy Looks Familiar!
What’s most important to realize here is that the 10 prize-winning brackets in this ~300-entry pool had a lot in common with the strategies that TeamRankings brackets generally employed in 2022. Specifically:
- They did not take a bunch of huge chances with their Final Four picks, even in a large pool. None of them called the UNC run.
- Most had teams that were all among the top 20 teams in the tournament as their Elite Eight picks.
- They didn’t dominate the First Round, but they did outperform their opponents in the final stages of the bracket.
So if you looked at a bracket we recommended in 2022 and thought there was too much “chalk” in the later rounds, remember that in most scoring systems, you win pools with your late-round picks. It’s always going to be tough to get more of those picks right than your opponents do, but in the years when it happens, it really pays off.
Several of our specific late-round “bets” this year (and particularly our Best Brackets featuring deep runs by Gonzaga, Kentucky, and Iowa) didn’t pan out. But in principle, the strategy behind our picks for traditional scoring systems was very much in alignment with what ended up winning in 2022.
TeamRankings Performance Detail
Here is how our bracket picks compared to the public for the 2022 tournament in each round:
|Bracket Type||Correct R1 Picks||Correct R2 Picks||Correct Sweet 16 Picks||Correct Elite 8 Picks||Correct Finalist Picks||Correct Champ Picks|
|TR "Best Brackets"|
For 1-2-4-8-16-32 Scoring
|TR "Best Brackets"|
For All Scoring Rules
|All TR Brackets|
For 1-2-4-8-16-32 Scoring
|All TR Brackets||22.0||8.8||1.62||0.62||0.17||0.11|
|The General Public||22.0||8.0||1.87||0.84||0.22||0.09|
Many of our brackets took some early losses that had long-term impacts on scoring, in the middle rounds of the tournament especially. The most notable of those losses were No. 2 Kentucky and No. 5 Iowa in the First Round. On the brighter side, a larger percentage of our alternate brackets (for multi-entry strategies) had Kansas as champion than the public did.
In general, our Best Brackets for traditional 1-2-4-8-16-32 scoring systems focused on Kentucky, Gonzaga, and Houston (as a long shot in the largest pool sizes) as NCAA champion picks. None of those teams reached the Final Four, while eventual Final Four teams Kansas, Duke, and Villanova were each picked by at least 18% of the public to make it that far.
As a result, most of our Best Brackets for traditional scoring systems finished relatively poorly in most pools. Depending on pool size, several of them finished in the bottom 33% on ESPN, while our Best Bracket for the smallest pool sizes finished in the 82nd percentile. As mentioned in the introduction, some alternate brackets for multi-entry strategies fared much better, such as this one in the 99th percentile.
What remains to be seen is the overall performance of our 2022 brackets for non-traditional scoring systems, which we’ll learn more about once our subscriber survey results come back. Overall, it still wasn’t a great year for those brackets, but anecdotally we’ve gotten word of at least some prize-winning outcomes.
Putting Expectations In Context
When reviewing bracket performance, it’s critical to remember that the goal is to be one of very few people in your pool who win a prize. “Coming close” to winning a prize, but not winning, is worth the same as coming in last place: nothing. And more often than not, “coming close” is just an illusion of success.
We design our Best Brackets for traditional scoring to make calculated bets on certain late-round picks. If several of them are correct, our brackets will do great. If we don’t get them right, and some No. 1 and/or No. 2 seeds that we didn’t pick make the Final Four instead, our brackets will do poorly. That’s simply the boom-or-bust nature of bracket pools in which late round picks are worth so many points.
Our lean toward No. 5 Houston as a dark horse champion pick in very large bracket pools this year is a great example of this phenomenon. Those brackets finished poorly overall compared to the public, but subscribers using them were in a fantastic position in big pools going into the Elite Eight. Long shot Houston had just upset No. 1 Arizona and was looking like the best team in the tournament after the Sweet 16.
Some horrible luck (including shooting 1-for-20 from three) against Villanova ended Houston’s run, but after three rounds, those brackets gave some of our subscribers a realistic chance of taking down some huge pots.
What Worked and Didn’t Work in our 2022 Brackets
As we mentioned above, a successful bracket picking strategy, especially for larger bracket pools, requires making some educated bets against the general public.
Our decisions regarding which teams to bet on in our brackets often trace back to each team’s tournament rating, which often includes an adjustment we make to its full-season rating based on factors like lineup changes, injuries, and COVID-related issues. Over the years, these adjustments have benefited our bracket picks overall.
(As a quick reminder, we provided subscribers with all our adjusted tournament predictive ratings in our NCAA Bracket Picks product. We also provided a detailed writeup containing our thoughts on the 2022 bracket and associated pick strategy.)
Because our system customizes subscriber bracket picks based on a pool’s size and scoring system (e.g. upset bonuses will change recommended picks to include more upset picks), the optimized brackets our customers receive can differ quite a bit from pool to pool. Nevertheless, there are some over-arching themes that impacted 2022 results for most of our subscribers.
1) Kentucky and Iowa losing early hurt most Best Brackets
No. 2 Kentucky and No. 5 Iowa both lost on the opening Thursday of the tournament, as big favorites. Both teams showed up frequently in the Final Four (or beyond) in our Best Brackets for various types of pools.
Those two losses left many TR subscribers hoping for lots of other upsets and general chaos in certain regions. Some of those upsets came through, especially in the East, where the Elite Eight matchup between a No. 8 seed and a No. 15 seed (both unpopular picks) helped limit the negative impact of Kentucky’s early loss. But the overall impact of these two losses was still significant.
Sometimes we take a stand on teams, trusting our analysis even more than the implications of betting market odds. That was not the case with Kentucky or Iowa. Favoring these two teams was primarily the result of a disconnect between public picking rates in bracket contests and betting market odds.
Iowa was featured in the Final Four in a good subset of TR brackets, including some Best Brackets, and was a finalist or NCAA champion pick for some pools.
The Hawkeyes had been hot coming into the tournament, but for our final bracket release the day before the First Round started, we didn’t rate Iowa any better than the market expected them to be. Our round-by-round advancement odds for Iowa were in line with what the betting market implied.
We knew going in that Iowa was far from the most likely team to make the Final Four, or to win the tournament. (Based on futures odds, Iowa was the 13th most likely team to win it all this year.) However, as an undervalued relative long shot with a still-realistic chance to win, the Hawkeyes fit the profile of a strong pick in pools where you need to take some risks to maximize your chance to win.
According to objective odds, Iowa had a bit less than a 1-in-5 chance to make the Final Four, but only about 1-in-10 brackets were picking the Hawkeyes to make it that far. The odds weren’t great, but if it happened, the payoff would be huge.
Kentucky was almost universally our East Region Final Four pick in our Best Brackets for traditional scoring. The Wildcats also showed up as the NCAA champion pick in our Best Brackets for traditional scoring at certain pool sizes, and in plenty of alternate brackets.
As we all know, Kentucky suffered a massive upset at the hands of undervalued and relatively unknown Saint Peter’s, in a game that went into overtime. Saint Peter’s backed up that win by also beating both Murray State and Purdue in two more big upsets, implying that they were probably a tougher draw than most people expected.
But as of the tipoff of the First Round game, Kentucky was still the third mostly likely team to win the tournament according to futures odds. Going in, you would have gotten a bigger payday betting on Duke, Villanova, or even Kansas to win the tournament than betting on Kentucky. So picking Kentucky to make the final game, or even to win it all, wasn’t going out on a limb at all.
It shouldn’t be forgotten that the entire magical Cinderella run of Saint Peter’s this year was made possible by one missed Kentucky free throw in regulation. That’s how thin the margin of error can be in bracket pools.
In the long term, picking teams with profiles like Kentucky and Iowa will pay off. But it didn’t work out this year.
2) Our Team Ratings Adjustments did a Good Job Identifying Underrated Teams, But Their Positive Impact on Brackets was Limited
Saint Peter’s was one of the biggest surprise stories of the 2022 NCAA Tournament. The Peacocks were also the recipient our third-biggest pre-tournament positive ratings adjustment of all 68 teams. In short, we recognized going into the tournament that Saint Peter’s was a significantly better team than their season stats implied.
In total, based on our manual review of team data, we made judgment calls to bump up the power rating of 17 different tournament teams by more than one point compared to their season average rating.
Over two-thirds of the teams we adjusted upward were double-digit seeds in the tournament. A couple of those teams faced off with each other in the First Four play-in games, and some others (Arkansas vs. Vermont, for example) met in the First Round.
However, in First Round games featuring a team we adjusted upwards playing a team we did not adjust upwards, the team we adjusted upwards went 7-6 (despite typically being an underdog) and 12-1 against the spread.
The teams we adjusted upward the most before the tournament started included:
- No. 8 North Carolina: Made Championship Game
- No. 15 Saint Peter’s: Made Elite Eight
- No. 4 Arkansas: Beat Gonzaga and went to Elite Eight
- No. 12 New Mexico State: Beat Connecticut in First Round
- No. 12 Richmond: Beat Iowa in First Round
- No. 11 Notre Dame: Beat Alabama in First Round, came within a couple points of beating Texas Tech and advancing to Sweet 16.
An Underrated Team Isn’t Always the Better Bracket Pick
If anything, we weren’t aggressive enough with our team ratings adjustments in 2022, and some of the teams we thought were playing better than their full season numbers implied ended up costing us big (e.g. Richmond and Saint Peter’s).
Still, just because we thought teams like North Carolina, Saint Peter’s and Richmond were significantly underrated doesn’t mean that it makes sense to pick them to win one or more games in bracket pools.
For example, by our numbers, Saint Peter’s should have been closer to a 15-point underdog against Kentucky than an 18.5-point underdog (the actual betting line). But actually predicting Saint Peter’s to win would have been utterly reckless in most bracket pools.
(Where our adjusted numbers helped most was in betting advice, and riding a team like North Carolina against the point spread was a big contributor to our success this year with our Staff Betting Picks.)
3) Compared to Most Years, Weaker Top Seeds Won More Games
Ultimately, 2022 ended up being a year in which teams that looked like they were solid values underachieved, while teams it most often makes sense to fade came through with key wins.
Over the course of the last eight years, that’s been an anomaly. Generally speaking, teams that are power-rated higher win more games in the NCAA tournament, once you control for seed number.
Comparing Historical Data
Below are a couple of tables that illustrate just how unusual 2022 was. First, we define whether a team was “properly seeded,” “over-seeded,” or “under-seeded” by its pre-tournament power rating. For example:
- A No. 2 seed that ranked in our top four teams in predictive ratings would be “under-seeded.”
- A No. 2 seed that ranked between No. 5 and No. 8 in our predictive ratings would be “properly seeded.”
- A No. 2 seed that ranked outside our top eight would be “over-seeded.”
We then compare each team’s actual wins to the expected average wins for that seed, based on 20 years of NCAA Tournament data. For example, No. 2 seeds have won 2.4 games in the NCAA Tournament on average. So a No. 2 seed that advanced to the Final Four would count as having four actual wins, and 2.4 expected wins.
Here are the results from 2015 to 2022, based on the team’s power rating versus seed.
Top Five Seeds that Were Under-seeded or Properly Seeded
|Year||Expected Wins||Actual Wins||Difference|
The only other year that under-seeded or properly seeded teams underachieved seed-based expectations was in 2018, a wild year in which No. 11 Loyola-Chicago went on its Final Four run.
But in that year, the team we had rated as the top team, Villanova, still won the national title. So while several teams underperformed, the highest leverage team still did enough to win a lot of pools.
That wasn’t the case this year, in which notable under-performers in this category included Kentucky, Iowa, Gonzaga, Arizona, Tennessee, and Connecticut.
Top Five Seeds that Were Over-seeded
|Year||Expected Wins||Actual Wins||Difference|
In the other six most recent tournaments, teams that were over-seeded based on our power ratings largely lived up to that expectation as a group. 2018 was the only other year in which they did not underperform seed-based win expectations.
But 2022 was at a completely different level. Several teams that did not rate particularly highly for their seed number won the key games, and in other cases, benefited by other upsets opening up an easier path forward in the tournament.
For example, both Villanova and Duke were outside the top 10 in our predictive ratings based on their full-season results, and both teams reached the Final Four.
In addition, Providence’s power rating was low for its seed, as its No. 4 seeding benefited from an extremely good (and likely at least somewhat lucky) record in close games during the regular season. But Providence’s tournament prospects got a boost when likely Second Round opponent Iowa was upset in the First Round.
Wrapping It Up
Bracket contests are risky business, especially when one or two game results not going your way can spell doom. So in any given year, even smart strategies can fail miserably.
Over time, betting on undervalued teams is clearly the better approach, as the data above shows. But it didn’t work out in 2022, and it didn’t deliver pool wins for most of our subscribers.
While we clearly prefer the opposite outcome, the “not win” outcome will always be the more common one. If bracket pools happened weekly or even monthly, it would be a lot quicker to translate the value of our approach into prizes and profits. That would be especially nice for newer subscribers who haven’t yet tasted the thrill of victory using our picks.
But when the frequency of competition is once per year, and the expected frequency of winning is significantly longer than that, patience is required.
If you were an NCAA Bracket Picks subscriber in 2022, we appreciate your business. Our commitment to you is that we will continue to improve our research and strategies, and work as hard as we can to deliver winning long term results. Have a great summer and we hope to see you back in 2023.
(If you want to read some of our past yearly bracket picks recaps, you can check out our 2021 Brackets in Review, 2019 Brackets in Review, 2018 Brackets in Review and 2017 Brackets in Review articles.)