March Madness 2019 has arrived!
Hopefully this article will prove useful to you as you make your March Madness bracket picks this week.
Our resume in brief:
In short, we provide premium technology and tools that give you the best chance to win your bracket pool -- our NCAA Bracket Picks product.
On that note, here are five tips for picking your 2019 NCAA bracket like a pro.
There are over 4,000 games in a single college basketball season.
To develop an intimate knowledge of every 2019 NCAA tournament team, a human brain would need to assimilate and process data from all of those games.
It’s simply not possible.
As a result, everyone from casual fans to the so-called experts on TV form biased opinions based on imperfect data.
Let’s say your uncle Jimbo watched a team play three games this year, and that team won all three of those games by 20 points or more. Guess what? Uncle Jimbo probably thinks they have what it takes to make a deep run in the tourney.
In addition, don't forget how a lot of commentators get on TV in the first place. Pro tip: It might have a bit more to do with their good looks and/or prior fame than their analytical skills.
Recency bias also comes into play, since a lot of people, including analysts for the big networks who usually cover NBA, don’t start paying much attention to college basketball until the conference tournaments start. Then, they tend to get overly impressed by a surprise conference tournament winner, even though there’s not much evidence that teams with "momentum" outperform expectations in the NCAA tournament.
In a season-long picking contest covering every college basketball game, only a supremely lucky or supremely skilled handful of humans would pick more winners than sophisticated computer models, or just always picking the team favored to win in the sports betting markets. So you should trust the markets and the models more than the humans.
In terms of models, there are plenty of reputable computer power ratings for college basketball that are free or cheap to get: TeamRankings (shameless plug), Pomeroy, Sagarin, Torvik, ESPN BPI, LRMC, others. Yet even great power ratings systems have blind spots, so it pays to consider how multiple systems rate every team.
In terms of the betting markets, starting Selection Sunday night, sports books begin releasing both their betting odds for all the First Round games, as well as futures odds for teams to win the NCAA tournament. With free odds conversion tools available on the web, you can translate those futures odds into implied probabilities.
Take the time to gather objective and accurate predictions, and you'll be in a much better position to win your bracket pool. It's not the only thing required to maximize your edge, but it's a key step.
The failure to recognize how your bracket pool’s scoring system should impact your picks is one of the biggest mistakes we see unskilled bracket pickers make.
In short, your pool’s scoring system can make a huge difference in determining the optimal strategy to win it. You shouldn’t even start to think about making your picks until you’ve fully analyzed this dynamic.
Even though we're in the middle of a sports analytics revolution right now, it's shocking how many people — including well-known sports analytics practitioners — completely whiff on this concept when dispensing their annual bracket pick advice.
To be considered as intelligent bracket advice, a quip like “Team X looks like a great pick to make the Final Four!” almost always needs be followed by a pool rules based qualifier, such as “...in bracket pools with traditional 1-2-4-8-16-32 scoring and no upset bonuses.” Yet you never hear that. Well, except on our site. 🙂
Let’s look at an example. The structure of the most popular 1-2-4-8-16-32 bracket pool scoring system -- which awards one point per correct pick in the First Round, on up to 32 points for picking the NCAA champion -- places a very high importance on getting your late-round picks correct.
For example, getting just one of your finalist picks right is worth 16 points, the same amount as getting 16 first round games right.
In that scoring system, in most years, you’re not likely to win your pool unless you make some late-round picks that come through. So you should focus the vast majority of your time on your Final Four picks and beyond, because agonizing over first-round upsets is almost certainly going to be a waste of time.
However, if your pool’s scoring system is flatter (say, 1-2-3-4-5-6), it’s a completely different story. In that scoring system, the first round is worth a total of 32 points, while getting an NCAA finalist pick right is only worth 5 points. Consequently, early round games are much more likely to have a big impact in determining the pool's winner, and you should concentrate on making the smartest possible picks in those high-leverage early rounds.
Finally, if your pool has upset bonuses, the strategy changes yet again. We’ve run tens of millions of computer simulations of bracket pools with upset bonuses. The results clearly demonstrate that most players in upset bonus pools aren’t nearly as aggressive as they should be when it comes to making bold picks.
In upset bonus pools, playing optimally tends to require making multiple calls that look pretty insane to a less-skilled bracket picker. The response of the masses usually goes something like this: "Five double digit seeds in the Sweet 16? That's absurd! It will never happen!"
Sure, it probably won't. But what those unenlightened players don't realize is that you may only need two of those five high-risk picks to win, in order to earn more points than you'd expect to score if you had picked all the favorites. That's the magic of upset bonuses.
Unfortunately for DIY bracket pickers, coming up with the absolute best bracket for a particular scoring system is basically impossible to do without a lot of math, solid coding skills, and lots of computers. Every March Madness, we spend over $10,000 spinning up a farm of 18 Amazon computer servers to run the necessary calculations.
Nevertheless, as long as you take the time to understand the implications of your scoring system, you'll have a level of advantage over most of your opponents. Before we built our current technology, we still won bracket pools significantly more often than expected using Excel to do most of the heavy lifting.
In bracket pools, there is no prize for getting a certain number of picks right. You win your pool if and only if you score more points than everyone else.
Put another way, you don't automatically win a bracket contest if you pick at least 70% of games correctly, or get three of your four Final Four picks right, or have a perfect 32-for-32 First Round. However, in order to win, you do need to pick at least one (and most likely several) games right that your opponents get wrong.
This is such a foundational element of bracket pool strategy that it’s amazing how many bracket pickers just don’t get it. We constantly field questions like “How many Final Four teams did you get right over the last five years?” In a vacuum, success rate picking Final Four teams is a relatively meaningless statistic, because winning a bracket pool is all relative.
Imagine you’re in a standard-rules bracket pool with 500 entries. The Elite Eight round just finished, and the Final Four consists of three No. 1 seeds, which were all the most popular picks to win their regions, plus a No. 3 seed.
Good luck winning that pool if you don’t get at least three Final Four picks correct. 499 opponents is a lot, and a bunch of them will have picked mostly favorites in the Final Four. Given the actual outcome, it's a pretty safe bet that at least a few of your opponents will have caught some luck and gotten three Final Four teams right.
Of course, that's not always how the tournament plays out. Unexpected outcomes happen fairly often, a fact that many bracket pickers seem to forget on an annual basis. As recently as 2011, for instance, the Final Four consisted of teams seeded No. 3, No. 4, No. 8, and No. 11.
In a year like that, just getting one or two Final Four picks right might be more than enough to take first place in your pool, especially if it’s a smaller pool.
So what really matters is how often you win pools, not how many correct picks you average each year.
If you got zero Final Four picks right in four out of every five years, but won a 500-person pool every fifth year, you'd sure be doing terribly picking Final Four teams right -- but your long-term profits from playing in bracket pools would be amazing.
This dynamic has huge implications for bracket picking strategy, because it means that the picks your opponents make impact the odds that your bracket has to win your pool.
Imagine you live in Chapel Hill, NC, and pick No. 1 UNC as your champion. As it turns out, 75% of your opponents make the same pick. That really sucks for you, because your odds to win a prize will be much lower than they could have been, had you picked another champion.
So you need to consider the picks your opponents are likely to make when you pick your 2019 NCAA bracket.
Back to the UNC example. Let’s say you’ve got a choice between picking Team A, likely to be a wildly popular NCAA champion pick in your pool, or Team B, which you expect to be a much less trendy pick.
Based on the trusty objective predictions you've compiled, you see that Team A has a slightly better chance to win the tournament this year.
In cases like this, you’re almost always better off picking Team B to win it all. Why? Because in the long run, your expected bracket pool prize winnings (your Expected Value, or "EV") will be higher. Of course, you’ll get your champion pick right slightly less often than your opponents will, since Team A is more popular and more likely to win.
However, in years when your unpopular champion pick wins, you'll be in a much better position to get the top score in your pool, as opposed to being just one of a bunch of people who got their champion pick right.
Game theory based decision making like this can be very difficult in practice, but it’s how the most skilled bracket pool players get their biggest edge. The simple fact is, it’s not always in your best interest to pick the team that you think is most likely to win, if they're also the most popular pick.
Like your pool’s scoring system, the number of entries in your bracket pool is another key strategy factor that should impact the picks you make.
As a general rule, the larger your pool, the more risk you will have to take to improve your chances to win a prize. We'll examine why, but the TL/DR summary is: Because you are competing against luck.
Anyone in a bracket pool, smart or dumb, has a chance to get lucky. The bigger your pool, the more people that will be expected to have a really lucky year.
This is just the phenomenon of randomness doing its thing, and some degree of it will occur every year. Sometimes, as we all know, that “clueless about college basketball / had their 5-year old fill out their bracket” person actually freakin' wins the pool. It's maddening, and it’s going to keep happening.
The bad news is, you still need to beat all lucky people in order to win your pool. The good news is, there's a strategy for that.
Our research shows that in very large bracket pools with standard 1-2-4-8-16-32 scoring, it pretty much always helps your cause to avoid picking all of the most popular teams in the higher-leverage later rounds. Instead, it’s better to make at least some highly contrarian bets (e.g. an unpopular but highly undervalued No. 6 seed making the final game), and hope that it’s somewhat of a crazy year in terms of how the tournament plays out.
The rationale here is that if your key picks in huge pools are all “safer” ones, even if most of them do end up winning, you’re still going to be competing against many, many opponents who also picked those teams. Even when you have a great luck year, this strategy is a recipe for finishing near the top of a bigger pool, but still not winning.
Remember, big pools aren't like grade school, where getting a 98% on your homework was awesome. Beating 98% of your opponents in 1,000-person pool only nets you 21st place, and probably no prize.
In comparison, taking calculated yet significant risks on highly unpopular teams will result in a lower expected score in most years, because highly unpopular teams usually have lower odds to win. It’s a boom-or-bust strategy, and in most years, you’ll finish in the bottom 50% of the pool standings.
Still, in the long term, you'll also have the best chance to actually win that huge pool at some point, and winning a big pool can deliver a life-changing payout.
Of course, another way to increase your odds to win a really big pool is to play multiple brackets, in which the key "big bets" made in each different you play are diversified. That's often the only way to give yourself a realistic chance of winning a huge pool in your lifetime.
We designed our NCAA Bracket Picks product to help with this use case, as it produces not only the optimal bracket for your pool, but also a portfolio of up to five brackets you can play together in a single bracket pool.
Creating an optimal multi-bracket portfolio also involves a whole new set of tricky math, but if you do a decent job, it significantly increases your odds to win a prize in any given year.
As you may have guessed, smaller bracket pools call for a more conservative strategy. Unlike in a 1,000-person pool, beating 98% of your opponents in a 50-person pool and taking second place as a result is typically good enough to win a prize.
In fact, you can usually get a pretty solid edge in smaller pools just by assuming that most of your opponents are going to get much too risky with their picks. As a result, you can play it safe with your picks in terms of going mostly with "chalk" (the favorites), then watch quietly as your competitors shoot themselves in the foot.
This last piece of advice is a corollary of Strategy #1 (Use Objective Data). Still, it’s worth calling out on its own, because objective data can still be misinterpreted by people who, bless their hearts, just don’t have a sound grasp of statistics.
Anytime you hear an analyst on TV preface their bracket picking advice with the phrase, “Now let me quote you this amazing stat I heard yesterday,” get your earplugs in as fast as you can.
We like to refer to these sound bites as the “Not So Golden Rules” of bracket picking, and they come in many forms. The following are made up examples, but they convey the point:
The Not So Golden Rules are almost never substantiated by enough data to conclude with high confidence that they are actually predictive. Look hard enough for a juicy sounding stat or trend, and you’ll usually find one -- but it’s often just the result of random chance. So don’t let yourself get charmed by sexy stats about bracket picking. If it sounds too automatic, it’s almost certainly bad advice.
The fact is, the characteristics of every year's NCAA tournament field are wildly different. First, the NCAA Selection Committee is made up of subjective human beings who, let’s just say, aren’t exactly renowned for being 100% consistent with their annual decisions. From year to year, the Committee also alters the criteria it uses for selecting and seeding teams, as with the recent introduction of the NET rating.
(In fairness, the Selection Committee does seem to be improving over time, in terms of "correctly" seeding teams according to how good they actually are.) Second, the distribution of the performance levels of the teams in any given NCAA tournament always varies. For example:
Consequently, blindly trusting historical data on how often No. 1 seeds make it to the Final Four, or how often No. 13 seeds teams upset No. 4 seeds, is folly without considering additional context. What really matters is what the actual expectations (i.e. objective predictions) were for each team, and how they performed in comparison.
After reading all this, it may sound like we've gotten so scientific about bracket strategy that we've taken all the fun out of NCAA pools. Au contraire. The most fun of all is watching the NCAA tournament knowing that your smart picks have given you a great shot to win your pool.
Besides, March Madness bracket pools are the most socially acceptable way to brazenly steal money from your friends and coworkers. They don't even get that mad about it! So why not make the most of that opportunity?
Good luck in your bracket pools in 2019. And if you enjoyed this post, please check out our product. Everything you just read about -- the objective data collection, the game theory, the customization of picks for your scoring system, the portfolio strategies for playing multiple brackets -- it's all in there.
In just a few minutes, you'll have an optimized bracket in your hands that gives you the best chance to win.
Printed from TeamRankings.com - © 2005-2019 Team Rankings, LLC. All Rights Reserved.