Editor's Note: This is a guest post from TeamRankings.com, a site that has provided data-driven bracket tools and analysis since 2004. They also offerpremium bracket picks.
When it comes to picking a bracket, there are no golden rules. Every tournament, every team, and every potential path to the Sweet 16 is different.
Even the usually-smart bracket advice of picking undervalued teams (e.g., teams that have a better chance to make the Sweet 16 than the public is giving them, based on pick popularity data from nationwide bracket contests) isn't all that easy in practice.
For example, it's simple enough to identify undervalued teams using win odds and pick popularity data. However, it's a whole lot harder to figure out how many of those value picks you should make, or exactly where in your bracket you should make them.
Balancing Risk vs. Reward In Your Bracket
One of the key goals of smart bracket strategy is to make sure that the risk/reward profile of your bracket as a whole makes sense for your specific pool's characteristics (e.g. its scoring system, size, and other factors). Make too many value-driven picks in a small pool, for instance, and you may end up with a bracket that is too risky overall, thus lowering your odds to win.
(We built our NCAA Bracket Picks product to figure out the optimal pick decisions for you.)
In general, though, it's always helpful to know which teams are the most overvalued teams at various stages of the bracket. That way, based on how your scoring system awards points, you can try to differentiate your bracket from the crowd where it's likely to matter most, by making sure all of your most important picks aren't on overvalued teams.
"Overvalued" Isn't Always A 4-Letter Word
Let's be clear about one thing first. In standard 1-2-4-8-16-32 bracket contest scoring, optimal strategy almost always calls for picking a bunch of overvalued teams, especially in the early rounds. Typically these picks involve the top teams in the tournament, which also tend to be very popular picks, and technically "overrated" as a result. But since top teams usually have the best chances to make a deep run, it's simply too risky to pick a bunch of them to lose early.
However, drop down a tier or two in team quality, and some better opportunities to fade the crowd start to present themselves -- teams that are being overrated by the public to make the Sweet 16, but also don't have the best chances to win it all or make the title game.
Four Popular Sweet 16 Picks Worth Reconsidering
On that note, here are four of the most overvalued Sweet 16 picks in the 2018 NCAA bracket, one from each region. In addition, none of the teams below is in the tournament's top five in terms of odds to make the title game or win it all, so depending on your pool's characteristics, they might be great picks to make an early exit from your bracket.
No. 4 Arizona (South)
Arizona is being picked by nearly 60% of the public to make the Sweet 16, but their actual odds to make it are closer to 45%. The problem for Arizona is primarily No. 5 Kentucky, a team that has been playing at a 1-seed level since having a players-only meeting back in February. There's no guarantee that Kentucky will carry its very impressive late season momentum into the NCAA tournament, but if they do, Kentucky is a legitimate sleeper pick to win the entire thing. Algorithmic projections give Kentucky and Arizona pretty similar chances to make the Sweet 16, yet the public strongly favors Arizona.
No. 3 Michigan (West)
Over 75% of the public is picking Michigan to make the Sweet 16, yet Michigan is only a modest favorite to get there, with a bit higher than 55% odds. It's not a big surprise to see Michigan being overvalued, since the public is prone to recency bias, and Michigan just made a surprise championship run in the Big Ten tournament. However, No. 6 seed Houston is no slouch, and was arguably underseeded by the Selection Committee. In addition, even a Houston upset loss in the first round would still put Michigan up against San Diego State, a relatively strong 11-seed.
No. 3 Michigan State (Midwest)
Michigan State is the most likely of all the No. 3 seeds in this year's tournament to make the Sweet 16, with around a 65% chance according to our numbers. That's over twice the odds of the next best team in its pod, No. 6 TCU. So why is Sparty on this list? Because close to 85% of public brackets are picking Michigan State to win two games. TCU, on the other hand, has about a 27% chance to make the Sweet 16. Those odds aren't great, but it's also a 27% chance to differentiate your bracket from the 85% of your competitors picking Michigan State -- and that could be a great bet to make in some upset-heavy scoring systems. Tom Izzo may well deserve his reputation as an amazing tournament coach, but stranger things have happened; let's not forget that Michigan State lost to Middle Tennessee in the first round in 2016.
See the best bracket for your pool »
No. 6 Florida (East)
Based on public picking data, No. 6 Florida is a pretty risky yet slightly trendy "upset" pick to make the Sweet 16. Florida's odds to make it there are under 35%, yet almost 37% of the public is making the pick. In contrast, the most likely winner of Florida's pod, No. 3 Texas Tech, has nearly 55% odds to make the Sweet 16 and is fairly valued to boot. So in standard rules pools, the risk-reward tradeoff of picking Florida to the Sweet 16 just doesn't make much sense. There are much better opportunities in the 2018 bracket to pick similarly risky (or even less risky) teams as Florida, yet actually have value on your side.
Should You Make All These Picks?
Let's reiterate what all this information means from a bracket strategy standpoint.
Should you avoid putting all four of the teams above in your Sweet 16 this year? No, almost certainly not. With the exception of Florida, all of the teams above are the favorites to emerge from their pods. In most pools, you'd be taking on way too much risk by not picking any of them.
However, if you're in a relatively large pool, or if your scoring system provides great rewards for picking lower-seeded teams to win, then these are some of the top opportunities to consider in terms of making early round upset picks against popular favorites.
Especially if you didn't plan on picking one of the teams above to make it much farther than the Sweet 16 anyway, you should seriously consider picking that team to lose even earlier.
If you want to see the best bracket for your pool, or just need a great bracket quickly since the tournament starts soon, check out our data-driven NCAA Bracket Picks product:
NCAA Bracket Picks from TeamRankings.com »
(If you'd like to learn more about TeamRankings, check out this WIRED magazine article and see how they did last year)