The Polacheck Method: Picking a Pool Winning March Madness Bracket

It is indeed that time of year folks. The greatest singular month of sports you’ll see all year.

The one competition where the bracket-makers who have never watched the sport have just as good a chance of winning as those who have actually played in the tournament.

Take a second and get the bracket in front of you right now. This is your map. This article will act as the building codes by which you would be wise to abide.

Your bracket will not be perfect. I guarantee it. (In case you were wondering, the odds of a perfect bracket are 1 in 9,223,372,036,854,775,808.) But it can still be beautiful.

Step 1: Pick a champion

Simply put, without a correct champion, you can’t win any pool. Most formats have the champion being just as important, if not more, than a perfect first round.

Who should you pick?

There’s two schools of thought. Pick a champion nobody else will have or pick one of the favorites and count on the rest of your bracket.

For your typical office pool, don’t get cute. Leave it to the non-know-it-all’s to make the bold picks. Pick a 1 seed.

They have won the tournament significantly more than any other seed. (21 out of 34 times to be exact). It’s really that simple.

Step 2: Pick a Runner-Up

Pick your preferred 1 seed on the other side of the bracket as your champion pick to put into the final. Consider this pick insurance, for the same reasons of picking a 1 as a champion. It is essentially a second guess at a champion, it can grab you a ton of points, and that could give you at least a chance in your pool, if the rest of your bracket was pretty close.

Step 3: The Final Four

Say no to picking all four 1 seeds. That’s only ever happened once (2008).

41% of 1 seeds all time make it to the Final Four and we have already put them in our championship game, so it is time to explore more of the bracket seeds.

Pick your favorite 2 seed. 47% of brackets all time have a 2 seed in the Final Four and 21% of all 2 seeds all time reach the Final Four, perfect for our picking needs.

Now, the final region. Time to go a calculated amount of rogue. This is where you can win your bracket by making a correct bold pick. There is no seed that has a significantly large number of representatives in this stage, but the ones with the most are the 3 and 4, each with about 30% of brackets with one of them in this round.

Step 4: Round of 64

The NCAA Tournament has had at least 64 teams in it since 1985, so we have plenty of precedent to use as our building code here.

1 vs 16

You should have two 1 seeds already penciled into the next round from your Final Four selections. The other two? Move ‘em on as well.

If you called UMBC last year, congratulations on being a wizard or a super Retriever homer, but your wizardry got you very little in terms of separation from the field.

99.3% of 1 seeds advance though the first round. Take those odds to the bank.

2 vs 15

Friday, March 16, 2012. 15^th seeded Norfolk State bounces Missouri. Just over two hours later, 15^th seeded Lehigh knocked out Duke.

Can 15 seeds beat 2 seeds? Yes. Will it happen this year? Don’t bet on it.

94% of 2 seeds make it to the next round, so make the right choice and pen them all through as well.

3 vs 14

We start to dip our toe into making some unique flourishes in our burgeoning bracket. Only 47% of brackets have all four 3 seeds advancing through. Those are spicy odds for picking a 14 over 3 upset, but the benefit just is not there. If your poll is particularly large (talking 100s of people here) and you want to be frisky, go ahead and pick a singular 14/3 upset.

4 vs 13

The 4/13 matchup follows a similar trend to the 3/14. Will all four 4 seeds advance? Probably not, since it has only happened 10 times out of 34. So go ahead and pick your favorite 13/4 upset.

5 vs 12

And now we start really putting our personal stamp on this map.

12 seeds are usually the best of the auto-bid teams, and that is why a lot of the 12 seeds are able to pull off the most famous seed upset. Given the notoriety of the 12/5 upset, it is tempting to go for broke and take them all.

I suggest you be moderate and pick 2. 65% of 5 seeds advance through here when looking since 1985, but recent brackets (last year not withstanding) have featured a higher amount of upsets (18 in the last 11 years).

If you really don’t love 2 matchups, picking only 1 of these upsets is also a viable option since it has happened 44% of the time all time.

6 vs 11

You will also be splitting this matchup 2-2. We are still looking at the best of the auto bids and the “Last 4 In” teams. Usually the “Last 4 In” teams are riding a hot streak, and that is a key to March. All time statistics tell you that picking three 6 seeds is the most common, (38%), but if you weigh the recent years more, (as basketball has evolved), picking 2-2 blows away the rest of the pick amounts, with my weighted percent gives a rating of 47%. Look at all four matchups at the same time, and pick the two you are most confident in, but this time it is wrong to pick only 1. Stick to 2 and only 2.

7 vs 10

This matchup is surprisingly more predictable than the ones above. 44% of time there is only one 10 seed that wins, but weighting recent years, they have a value of 59%. Take those odds, and pick your one favorite 10 seed to win. Use the statistical trend to your advantage in the seed matchup that feels like a toss-up.

8 vs 9

The biggest toss-up matchups possible. All time, these seeds have split a perfect 50/50. And while this may tell you to pick the matchup 2-2, maybe hold off on these until you go through the next step.

Step 5: Round of 32

1/16 vs 8/9

Since we haven’t completed the round of 64 (thanks 8 vs 9s), the first decision you have to make is whether or not the round of 32 will be the end for one of your 1 seeds.

It is not as crazy as it sounds. 50% of previous brackets have at least one 1 seed falling in this round. Only pick an upset here if you truly 100% believe in it, but know that calling an upset incorrectly is a huge killer to your points. Remember that it is better to lose with everybody else than to try to win it all on an early pick. You cannot really go super wrong by picking all 4 of the 1 seeds here.

If you pick a 1 to bite the dust this early, odds are extremely in the favor of the 8 seed beating them rather than the 9. 8 seeds win 20% of their games against the 1s, while 9 seeds only win 8% of the time. Given that the 8/9 matchup is essentially 50/50, that is a HUGE difference.

4/13 vs 5/12

Even though the opening round often brings about surprises, the majority of past brackets follows a surprisingly consistent script. The seeds that will make the Sweet 16 from here are 4-4-5-12. 53% of past brackets have a single 12 seed here, 41% have a single 5 seed, and 47% have two 4 seeds, all are clear majorities.

I advised earlier to pick a 13 over 4 upset. That 13 seed will more than likely lose in this round - they do so 78% of the time. This has to be a 5 or 12 that wins.

Once you have that filled in, you need a representative of the other seed. Look at the matchups you have of the teams that are left, and pick the best team seeded whatever 5 or 12 is required to fill the spot. From there, you should have two 4 seeds to move into the Sweet 16.

3/14 vs 6/11

If you picked a 14 over 3 upset, end that now. The 14 seed will lose. Only 6% of brackets have a 14 getting this far and the last one was in 1997.

From here, the percentages all time make things feel like a coin toss, but when I weigh the recent years more heavily. Recent trends suggest you pick 3-3-6-11.

It is similar to the way we picked the last sets of matchups. The 3 seed wins only 60% of their games in this round, the 6 seed wins 49% of the time, and the 11s win 41%.

Once you eliminate that 14 seed, get a representative for the 6 and a representative for the 11, whichever ones you need. These should just be picking the best ones you still have in the round, or pick the ones who play the worst 3 seed. This should then leave you with two 3 seeds to move on to the Sweet 16.

2/15 vs 7/10

Thus far, you have four 2 seeds, three 7 seeds, and a 10 if you have followed the steps above.

This round is intriguing, as 2 seeds only win 66% of their games here, so a top 8 team should see their run end here (For reference, they win 73% of Sweet 16 games).

Let’s decide how many 2 seeds we want in the next round.

You already have one in your Final Four that makes it through. One down.

Pick the best remaining 7 seed to win their matchup –they win 1/3 of their games against 2 seeds, and we have 3 of those matchups here. Two down.

Take those last two 2 seeds. Traditionally, this part of the bracket is just simply not where you can outwit your competition.

Step 6: Sweet 16

At this stage, you have a mixed bags of seeds.

You should have at least three 1 seeds, but likely four, and three 2 seeds.

Roll three 1 seeds onto the Elite 8. The most common occurrence (38% of brackets) features three teams. 69% of all 1 seeds reach this far, and there is an average of 2.76 teams at this stage. Weighting recent years also does not change this. Three is the right number to have.

Since we have to decide on a 1 seed to lose, and you already have 2 in your Final Four, this gives us 2 options (unless you chose one to lose earlier, then you can ignore this part, choose the three 1 seeds still rolling).

Based on what we picked above, there’s a possibility they play a 4, 5, or 12. Let’s break down all possible combinations.

If you have two matchups against 4 seeds, pick the upset you’re more confident could happen, as all percentages are the same, they’re the same seeds.

If you have a 4 and 5, pick the 4 to pull off the upset, 44% of brackets have a 4 representative winning in this round, while only 18% have a 5 seed.

12 seeds have also never beaten a 1. So the 1 seed that plays a 12 is the 1 seed you will advance, thus you should pick the other 1 seed to lose.

If you did pick an 8 seed to get this far, keep them rolling. 8 seeds that make it to the Sweet 16 win 61% of the time. They don’t have a ton of games this far, but they have beaten the 4s and 5s more times than they’ve lost to them. (They win 58% of the time against these seeds). So in an area where it’s hard to have a ton of data from, use those odds unless you truly don’t believe in that 8 seed (but you probably do as you picked them to beat a 1).

Now to the bottom half of the regions. We are down to three 2 seeds, two 3 seeds, a 6, 7, and an 11. Deciding on what teams to move through is not exactly clear. We have one 2 seed in the Final Four already. The rest is up for grabs.

53% of brackets all time have only two 2 seeds. Almost never have more than that advanced. We’re going to take the majority, 53% is an astronomical number for this late in a bracket, so we need to pick one to lose.

Since we need a 2 seed drop out, it can only possibly face a 3, 6, or 11. Versus 11s, they win 87.5% of games, and versus 6 seeds, they win 77.4% of the time. Use these odds to your advantage and make sure those matchups feature a 2 seed advancing. This means the 2 seed you will have losing will have to come from the hands of a 3 seed. 2 seeds win only 62.7% of these seed matchups, so it is not uncommon for this result to occur.

So what do we do with the 7 seed that is left? If this 7 seed is correct, this is where your bracket can truly be won. 7 seeds win 40% against 3s, 42.9% against 6s, but have never beaten an 11 (4 games played).

If you have 7 vs 11, pick the 11!

But for the other two matchups, it truly is a toss-up. For all of the seed statistics, the only numbers that are any different are the recent year weighted probabilities that teams get to the Elite 8. Most other statistics are very close to each other, if not exactly the same.

7 seeds benefit from the recent year’s bias as they have had lots of success in the last few years. Since the win percentages are almost 50-50, I suggest you take the 7.

This is where your bold pick will win you your bracket, if your Final Four is mostly correct. This sole selection has the most weight, so in the 7 vs X game, dive super deep into this matchup and make sure that the decision you are making is the right one, and that it is a team you believe in.

Step 7: Elite 8

Three of the four matchups are already decided, as you have picked 3 out of your 4 Final Four teams. That last matchup, let’s take the historical statistics for one last go around.

75% of 5 seeds that make it this far go on to the Final Four. 4 seeds win 61.9% of their times here, 7 seeds win only 30% of the time, 8 seeds win 62.5% of the time, and 11 seeds win a mere 50%. For comparison, 1 seeds win 59.6% and 2 seeds win only 45.2%. The logic of this seemingly backwards trends is that these low seeds have likely already knocked off a high seed to get here.

To make your last Final Four pick, it is easier to turn to a handy dandy chart. Below lists the win percentages of seed versus seeds (an X represents that there haven’t been 5 games played). Greater than 65%, take it to the bank. Less than 35%, get ‘em outta here. And there is your last Final Four team!