That NCAA Tournament buzz is in the air. And with the announcement of FanDuel's Bracket Pick'em game, there is even more reason to be excited about March.
To make sure your Bracket Pick'em picks set you up for success, we need to dig into some strategy surrounding the new game. And you've come to the right place.
When it's all said and done and the final net is cut down, which five-team combinations would have stood at the top of the heap each year?
Let's dive in.
Past Optimal Combinations
Let's info dump. Here are the optimal combinations possible each year over the last 10 years, based on FanDuel's Bracket Pick'em scoring (champions are in bold).
| Year | Optimal Combination | Seed Total | Total Points | Final Four Teams | 1-Seeds |
|---|---|---|---|---|---|
| 2017 | 7 South Carolina (1295) 1 North Carolina (935) 11 Xavier (935) 3 Oregon (555) 1 Gonzaga (435) | 23 | 4,155 | 4 | 2 |
| 2016 | 2 Villanova (1870) 10 Syracuse (1850) 6 Notre Dame ( 510) 1 North Carolina (435) 11 Gonzaga (385) | 30 | 5,050 | 3 | 1 |
| 2015 | 7 Michigan State (1295) 1 Duke (935) 1 Wisconsin (435) 11 UCLA (385) 4 Louisville (340) | 24 | 3,390 | 3 | 2 |
| 2014 | 7 Connecticut (6545) 8 Kentucky (3480) 11 Dayton (935) 11 Tennessee (385) 2 Wisconsin (370) | 39 | 11,715 | 3 | 0 |
| 2013 | 4 Michigan (1740) 9 Wichita State (1665) 1 Louisville (935) 4 Syracuse (740) 15 Florida Gulf Coast (525) | 33 | 5,605 | 4 | 1 |
| 2012 | 1 Kentucky (935) 2 Kansas (870) 4 Louisville (740) 7 Florida (595) 13 Ohio (455) | 27 | 3,595 | 3 | 1 |
| 2011 | 8 Butler (3480) 3 Connecticut (2805) 11 VCU (2035) 4 Kentucky (740) 5 Arizona (425) | 31 | 9,485 | 4 | 0 |
| 2010 | 5 Butler (2175) 1 Duke (935) 5 Michigan State (925) 6 Tennessee (510) 12 Cornell (420) | 29 | 4,965 | 3 | 1 |
| 2009 | 1 North Carolina (935) 2 Michigan State (870) 3 Villanova (555) 12 Arizona (420) 3 Missouri (255) | 21 | 3,035 | 3 | 1 |
| 2008 | 1 Kansas (935) 10 Davidson (850) 1 Memphis (435) 12 Western Kentucky (420) 12 Villanova (420) | 36 | 3,060 | 2 | 2 |
As it works out, these past 10 perfect combinations happened to be just the five highest-scoring teams. The combined seeds naturally topped 20, thanks to the multipliers. In fact, the past 10 optimal lineups averaged to have a total seed of 29.5, well over the minimum cutoff of 20.
Also, the past nine optimal lineups included at least three Final Four teams. All 10 included the eventual national champion.
Exploring Past Optimal Trends
Optimal, of course, implies the top score possible, but you don’t necessarily need the top score in order to win on FanDuel. You need to aim for at least 400 points, and if you score at least 2,500 total points, you'll split $20,000 with the other entries who do the same. The top score will get (or split) $5,000.
What trends can help you put up a big score?
This table displays seed and frequency for combinations that scored at least 2,500 points since 2000, which we'll call a "winning score." (This means, for example, 62.9% of lineups to score at least 2,500 points had at least one 1-seed.)
| Seed Frequency | At Least 1 | At Least 2 | At Least 3 |
|---|---|---|---|
| 1 Seed | 62.9% | 12.0% | 0.3% |
| 2 Seed | 56.9% | 11.4% | 0.5% |
| 3 Seed | 69.2% | 19.3% | 1.6% |
| 4 Seed | 42.5% | 7.0% | 0.5% |
| 5 Seed | 46.2% | 7.8% | 0.5% |
| 6 Seed | 30.2% | 3.1% | 0.1% |
| 7 Seed | 31.2% | 3.5% | 0.1% |
| 8 Seed | 31.3% | 2.9% | |
| 9 Seed | 17.5% | 0.4% | |
| 10 Seed | 14.2% | 0.1% | |
| 11 Seed | 11.1% | ||
| 12 Seed | 6.6% | ||
| 13 Seed | 4.2% | ||
| 14 Seed | 2.6% | ||
| 15 Seed | 1.5% | ||
| 16 Seed | 0.8% |
The most common "trend" is actually having at least one 3-seed (69.2% of winning combinations since 2000 had at least one 3-seed). That’s followed by having at least one 1-seed (62.9%), one 2-seed (56.9%), one 5-seed (46.4%), and one 4-seed (42.5%).
Rostering multiple 9-seeds or lower yielded a winning score less than 0.5% of the time over the past 18 NCAA Tournaments, which makes sense. Firstly, 9-seeds face 1-seeds in the Round of 32, and taking two or more double-digit seeds already gets you to the minimum total of a combined seed tally of 20.
The data says that you never want to go with four of the same seed, even four 1-seeds (which would require a 16-seed paired with them). No combination since 2000 would've earned at least 2,500 points if you select all four teams that are a certain seed.
Even three 1-seeds has been a weak combination (1.6% of winning lineups had three 1-seeds). The reason for that is that a non-championship appearance from a 1-seed will leave your scores lacking, due to the absence of a multiplier (1-seeds get their scores multiplied by 1).
But nearly 20% of combinations with two 1-seeds met the point cutoff (19.3% to be exact) because 1-seeds have won 12 of the past 18 tournaments, and only Connecticut in 2014 (a 7-seed) won with a seed lower than 3.
Sticking with 5-seeds and lower (as in, 1-4) could be the key to putting up a winning score.
Region Stacking
With the need to roster five teams and only four regions in the tournament, you must select at least two teams from the same region at least once. Should you do everything you can to spread out your picks, or is there merit to stacking up a particular region to try to ensure an Elite Eight appearance?
Here’s the statistical breakdown, again based on our sample of combinations scoring at least 2,500 points.
| Number of Teams in Region | Frequency |
|---|---|
| Maximum of 2 Teams | 65.4% |
| Maximum of 3 Teams | 31.0% |
| Maximum of 4 Teams | 3.5% |
| Maximum of 5 Teams | 0.1% |
This means that 65.4% of lineups scoring at least 2,500 points maxed out at two teams in a region. Take note of that. There’s still some merit to going with three teams in the same region, which has yielded a 31.0% success rate in our sample, but spreading out your picks can maximize your lineup's upside -- and its floor.
You probably don’t want to mess with four teams in the same region, and certainly avoid going with all five teams from the same quadrant.
Takeaways
Even though 1-seeds don't benefit from a big multiplier, there's safety in selecting them, as they can put up a big score (935 points) with a championship. No matter how hard we root for the underdogs, top seeds win the tournament quite often, but 2- and 3-seeds are very vital picks, given the multiplier.
Loading up with 5-seeds and lower (better) could be the sweet spot, though. You also want to spread your picks out to different regions if you're looking for 2,500 points.
Just don't get too cute while trying to find a double-digit seed you love. The best combinations are full of Final Four teams, champions, and low seeds.