NCAA Tournament Bracket Strategy: How I Learned to Stop Worrying and Pick Against Kentucky and Duke
By Jason Lisk
Kentucky is the favorite to win the NCAA Title and finish 40-0, equalling the 1976 Indiana team’s undefeated run to a title. No one denies this. So, you’re filling out your bracket, and you’ve got Kentucky winning it all.
Congratulations.
So does almost everyone else. If you like to pick your brackets based on things like “who I actually think will win” and don’t think about strategy in winning what is ultimately a contest with long odds, then read no further. You can go brag to your friends how you totally nailed Kentucky.
If you want to win your bracket pool, though, your best chance probably involves picking Kentucky not to win the title, and playing the “Don’t Pass” line, and being “that person.” Yes, that may be a hard concept to accept when making picks, but sometimes you have to play the pot odds. If Kentucky wins the title, then you almost certainly will not win. Of course, if you picked Kentucky, and they win, you almost certainly will not win either. You will have had to be better than everyone else at all the other picks. That West Virginia-Buffalo pick that you are debating and might pick based on colors? Yeah, that and a few other games like it could be the key to you being effectively blocked from winning the pool before the weekend is out, if you pick Kentucky to win it all.
Currently, Kentucky is being picked as champion in an unheard-of 59.7% of all Yahoo pools. Ken Pomeroy has them at 33.4% chance to win the title. The Vegas odds on Kentucky are 1:1, so basically even, but remember that includes the vigorish (so Vegas thinks Kentucky’s chances are less than 50%), and Kentucky is a very public team, so Vegas knows they have to account for that enthusiasm, above and beyond the real chances.
Or, think of it this way. If Kentucky plays Arizona or Wisconsin in a national championship final, they’ll be a favorite. But how much? Maybe 60%? That’s one game, and yet the public is expecting them to win six that percentage of the time.
Maybe you think, ahh, Pomeroy and those numbers, clearly he’s missing something. Kentucky is the highest rated team compared to the last eight years in his system. The highest rated team (which often coincided with public perception) won three titles in eight years, and the other five failed to reach the Final Four thanks to big upsets.
It’s March. Stuff happens. This is a tournament where the best team fails to win more often than not.
So yes, I’m saying that your best strategy is to pick someone else. Here’s a summary of all the #1 and #2 seeds, and the public percentage of picks versus Pomeroy.
Look, Kentucky is more than twice as likely as any other team to win, and that feels right. But, there is some value in going elsewhere. I would recommend picking Arizona or Wisconsin to win the title. I would lean Arizona because a) I think they are a little more balanced offensively and defensively, and can match up with Kentucky inside, and b) let’s face it, they are a 1-seed wrapped in a 2-seed, so you are getting a little seed bias discount there.
The reason I would recommend those two is because they would face Kentucky in a national semifinal. That provides a little more separation for uniqueness of your bracket from all of those that have Kentucky advancing further. Alternatively, taking Villanova or Virginia, or even a lower seed on the other side (we’ll get there in a minute) is a valid strategy.
Taking Duke, on the other hand, is not. 7.4% of the entries have Duke winning it all. That puts them as 3rd favorite. Meanwhile, a whopping 37.2% of entries have Duke in the championship game. That means there are a ton of Kentucky-Duke finals in pools, and likely in ones you are in. Duke is another national brand and people are yearning for a Kentucky-Duke final. Pomeroy, though, gives Duke only a 12.6% chance of reaching the final. Unlikely Kentucky, I’m not sure Duke merits that optimism compared to the other top seeds. There’s a ton of value in knocking Duke out of your bracket before the Final Four. Accept the fact that if Kentucky and Duke meet in the championship game, and make all the Yankees/Cowboys/Cavaliers fans happy, you aren’t winning money. Make peace with that.
Be the one to have a great chance to win if they do not.
Let’s face it. There’s a brand bias and big name bias when people make pool picks, and you have to take advantage of that. Plenty of people might even think that being a past champion makes you more likely to win. Hogwash.
Most #1 seeds are from blue-blood programs, and #1 seeds win more tourney titles. 72% of all #1 seeds from the last 25 years have been schools that had won a title previously, and 76% of all #1 seed champs have been former champs. Meanwhile, there have been more winners from the 2-3-4 seeds who were not prior champions than were (4 vs. 3) even though the “past champion” to “no titles” split was even in those seed lines.
Set aside just title winners. If you bought into the “they’ve never won” philosophy, you would have bypassed things like Butler and VCU to the Final Four (after Butler went as a 5-seed before), George Mason to the Final Four, Wichita State to the Final Four, Davidson to the Elite Eight, Northern Iowa over Kansas, and numerous 12 and 13 seeds to Sweet Sixteen runs. Some of those were unexpected. Some of them less so, even though they were not the most likely.
Know your pool. If you are in a pool with a bunch of Arizona fans, obviously, my advice would change. Figure out where the value is. But recognize the biases people have.
Here’s the breakdown of some prominent teams to reach the Final Four:
The five teams with the largest difference: Duke, Kentucky, Louisville, Wisconsin, Michigan State. If you had Michigan State as your sleeper, you are not alone in bed. More than 7 times as many entries have Michigan State to the Final Four than #5 seed Utah.
OVERVIEW
“I write the last line, and then I write the line before that. I find myself writing backwards for a while, until I have a solid sense of how that ending sounds and feels.”
— John Irving
Plenty of people like to just go through, pick by pick, starting with the first round. I like to build my picks backward, by starting with the last few lines. Figure out where the value lies for the biggest point picks, those in the Final Four. The goal–let’s say you are in a 50-person pool–is to have a 5% chance instead of a 2% chance.
Working backwards, I’m starting with Arizona or Wisconsin. Depending on the point system and size, I might have a Villanova in the final, or a Utah or Oklahoma. I’m probably taking one of these teams to the Final Four. Duke will be out. As you write backwards, you can figure out the level of risk you want to take on when to remove them.
Once I have the Final Four built, I can look at where other value lies in working backward. Wichita State looks like a potential play to the Elite Eight. Maybe Northern Iowa considering way more people like a now-depleted Louisville team. In areas where I don’t have the 4/5’s advancing far, I can look at upsets in pools that give upset bonuses. (I’m looking at you, Midwest region). But it all starts with setting up the ending.
And if you don’t listen to me about Kentucky, fer-gods-sake, don’t take Duke to the Final. That’s when you need to pick someone like Utah or Northern Iowa or Oklahoma to the final to take a calculated risk.