Happy March Madness season, everyone! Even if you aren’t the biggest basketball fan, there is no denying that March is a special time of the year because of all the fun that comes with the tournament. As you are filling out your brackets, I’m sure you will be reading a dozen articles, that toss out all kinds of odd stats. You can use stats to make a case for or against any team in the field, but today I am going to talk about the one stat with the highest predictive return of investment and how to utilize it to maximize your odds of winning your bracket challenge.

Sync custom cheat sheets to your fantasy baseball draft with our draft software

Overall efficiency is the name is the game. There are metrics built by KenPom.com, TeamRankings.com and FiveThirtyEight.com among several others, that take offensive efficiency, defensive efficiency, schedule of strength and luck then mix them up into a power ranking. These efficiency measures have been substantially more reliable than RPI, seeds, and anything along those lines to predict March winners. One team that stands out is Wichita State, who is ranked #8, #11 and #14 by these three sites, but somehow ended up a 10-seed in the tournament. On the opposite end of the spectrum, Arizona is a 2-seed, but ranked #10, #20 and #20 by the same measures. I took these numbers, threw them into the bracket and simulated 10,000 tournaments to determine the odds each team would win the title. After that, I compared those odds to the percentage of Yahoo and ESPN brackets that have each team winning the title. The reason I do this is that the trick to maximizing your odds of winning your contest is to find the teams with a chance to win the title who are overlooked.

If you don’t get your championship pick correct, you have virtually zero chance of winning your bracket because that game is weighted so heavily in bracket challenge scoring systems. That is, unless some team like Oklahoma State somehow wins it all. Let’s say you are in a contest with 199 other entries. That gives you and everyone else a 0.5% chance of winning before you even fill out your bracket. If you are one of 20 people to pick the winner, your odds of winning become 10 times as high, so you are now looking at a 5% chance of winning. That is all great, but we don’t know who will win, only the odds each team has to win. Instead, think about it in reverse. If you pick Villanova to win, you are among the 16.3% of brackets across ESPN and Yahoo that did the same thing. That means in your contest with 200 entries, you are 1 of about 33. In 200 simulations, however, Villanova just wins 29 times, so even though they have the highest odds in the field of winning, your odds of winning your bracket actually go backward if you fill them in as your champs. Here is all the data:

Brk% = Percent of brackets that selected the team as champions
Champ% = Percent of 10,000 sims that the team ended up champions
Entries = How many brackets you should expect with them as winners in a contest of 200
Sims = How many times in 200 simulations each team would be expected to win

So how does this change your odds? Remember, your odds before filling out your bracket were just 1 in 200, so picking some of these teams as your winner would be a major step in the right direction. Below are the odds your bracket would have in a 200 entry contest if you pick a team as your champion, and how much that increases or decreases your initial 1 in 200 odds:

Now, the rest of the bracket requires a different approach. It is all about not shooting yourself in the foot. Of course, there are going to be upsets, but even if you manage to pick 10 out of the 14 first-round upsets correctly, it won’t give you much separation from the rest of the field. On the flip side, if you try to pick 16 first-round upsets and get just two right, then have a handful of those teams in the Sweet 16, that could kill your bracket. Play it cool by taking a Rhode Island here, a Wichita State there, and maybe toss in a Middle Tennessee, but beyond that, you are risking too much. This concept also applies to the Sweet 16 and Elite 8, but less so than the first round, because correctly predicting St. Mary’s in the Elite 8 does actually help a fair amount.

If your contest is more than 200 players, strategies obviously change a bit. For instance, in a 1,000 person bracket contest, 24 people would be expected to pick Louisville as their champions, so you may need additional separation in your final four if you choose them. Or in a smaller 10 person bracket contest, you will often be the only one that picks Gonzaga, so there would be no need to go bold with a West Virginia or Virginia. Also, keep in mind that if your contest’s scoring system is different than standard settings, this may not all apply. Lastly, please realize, this is not any kind of sure-fire winner. Rather, it only enhances your odds of winning and by a great amount, mind you, but still gives you just a 6% chance at best in a 200 bracket contest.

Thanks for reading and good luck!

Subscribe: iTunes | Stitcher | SoundCloud | Google Play | TuneIn | RSS