Bracket Contest Breakdown

My bracket is currently in 1st out of 19, 94th percentile in the country, with 430 points. 

In the Elite 8, I can get four teams right: Alabama, Houston, Texas, Gonzaga.

Final 4, I have three alive: Alabama, Houston, Gonzaga. 

And the title game of Houston vs Alabama is still alive. 

So good position, but still only four teams alive.

Niraj trails me by 40 points, but he has a chance to get an extra 40 points because in addition to the exact bracket I have, he also has UConn advancing to the Elite 8. If UConn beats Arkansas, Niraj and I will end up with the same score.

Despite having tiebreakers, ESPN does not calculate the tiebreaker. Instead saying that it total score relative to the actual total score. My tiebreaker total was 150. Niraj put 146. If the national championship scores 148 points, I guess we're still tied. 

But that's just Niraj. 

Harrison has 5 teams alive: SDSU, Houston, Texas, Arkansas, Gonzaga. He also has Gonzaga winning it all. So he has a path to victory.

Brit has the same 5 teams alive as Niraj but with Texas winning it all. So she has a path to victory. 

Adam's wife KC has Xavier beating Texas but has no teams in the championship alive. Her path is very slim and relies on Xavier and chaos. Seems almost impossible. 

Savan is the highest score that has Alabama winning it all. So he has a path.

Mark can pass Savan as Mark has Kansas State in the Final Four. Savan has Gonzaga so Mark needs Kansas State to keep winning, root against Gonzaga, and wants Alabama to win it all. But it's possible. 

Eve has some unique teams alive, so it seems like she also has a path. Her Elite 8 teams alive are Alabama, Miami, Xavier and Tennessee. With Tennessee in the Championship game. So it's possible. 

Adam has Xavier and UConn alive, along with Alabama, Houston, Gonzaga and Houston winning it all. So if everything fell his way, probably has a path.

Jonah could possibly tie or pass Brit with Tennessee and UCLA wins and then be in prime position if Texas wins it all. So still alive I think.

Grant despite intentionally making a bunch of bozo picks, has San Diego State beating Gonzaga in the championship. Still alive. 

Matt despite being at the bottom, has UCLA winning it all with Xavier also in the Final Four. Still a chance.

Cora, Leela, Missy, Jay, Bridget, Kirat, and Sid cannot pass people ahead of them. 

The 11 people with a path to victory are in bold.

How To Build a Winning Bracket (Advice for Next Year)

When it was time to make a bracket, I searched in vain through my own archives for a post that would help me build a winning bracket. I found no post. I lost. 

So this year, I made 12 different brackets, partially for fun, partially to experiment and see what it takes to build a winning bracket. Currently with the title game yet to be played, my brackets range from the 4th percentile to the 96th. 

I'm going to go round by round and look at the best bracket I made, compare it to what I submitted as my primary bracket, and see how I can replicate the success in the future. 

Best First Round: 538 chalk: 240 points

Upsets picked: one 9 over an 8, one 10 over a 7

Results of upset picks: 2/2

Results of chalk picks: 22/30

Total for first round: 24/32

My First Round: Backyard Dunks: 190 points

Upsets picked: 9, 13, 10, 12, 11, 12, 10

Results of upset picks: 2/7

Results of chalk picks: 17/25

Total for first round: 19/32

It's hard to pick upsets. It's so tempting when you know there will be some. And it's fun when you get it right. I had UCLA over BYU and Ohio over Virginia. But I also missed on 5 upsets, plus all the upsets I didn't pick. It's best to go conservative early. 

Best Second Round: (tie) 160 points

Mostly BPI  2nd round picks:

1532 1462 1562 1532

Home Cookin' 2nd round picks

1562 1432 1462 1432

Backyard Dunks 2nd round picks

1532 1432 1462 1462

So my Backyard Dunks bracket got 6/16 correct, while the top two both got 8/16.  All the brackets in question picked the 1 seeds and 2 seeds to get to the sweet 16. So it really came down to which 4/5 and which 3/6 to go with. 

Best Sweet 16: 538 chalk 160 points

This bracket picked all 1 seeds and 2 seeds to advance to the elite eight and it was 4/8, which is the best I could do. Harrison used a combination of 1-3 and 1-2 to get 5/8 of correct. 

I used 1-2, 1-3, 1-2, 1-6 and only got 2/8 correct. Going with 1-2 to make every elite eight is more boring, but is going to give you the best chances here. 

Best Elite 8: I had two brackets that got Gonzaga-Baylor-Houston correct in this round, 3/4. 

Their final fours were 1-3-1-2 and 1-2-1-2. The correct answer was 1-11-1-2. 

My final four of 1-3-2-1 was in the ballpark but only got one correct. 

In the top quadrants, it's makes sense to go with the 1 seeds to make the final four as they're the two best teams overall. 

In the bottom, might as well pick the 2 seed.

Let's test that quick theory: The last 5 years, the 1 seeds on top of the bracket are 5/10 to reach the final four. 

The last 5 years, the 1 seeds on the bottom of the bracket are 3/10 to reach the final four. 

Best Final Four: 

In the last five years, every title game has featured at least one 1 seed, twice it's been both. 

How to fill out my bracket in the future. 

Start by filling out an Elite Eight that is 1-2 in every region. 

Final Four should probably have the 1 seeds on top advancing, probably a 2 from the bottom. 

Picking your title game and winner, stick to the team that was strongest all year, not someone who got hot recently.

In the Sweet Sixteen should stick to the 4/5 and the 3/6, and lean towards 3s. 

The Second Round should have hardly any upsets, literally just look at the 8-9s probably. 

Now just to keep these notes secret so only I can see them.

Bracket Philosophy: Sweet 16 Upsets

I've studied this at length and here's the long and short of it: I think people rely too much on chalk when making their Sweet 16.

As defined for Hoagie Central, an upset in the Sweet 16 is any team not seeded 1-4. Here's the history dating back to when we started college for number of total upsets, as well as times that a region went by chalk 1-4.

Sweet 16 History
2001: 6 upsets, 1 perfect region
2002: 6 upsets, 0 perfect regions
2003: 7 upsets, 0 perfect regions
2004: 7 upsets, 1 perfect region
2005: 8 upsets, 0 perfect regions
2006: 6 upsets, 0 perfect regions
2007: 5 upsets, 1 perfect region
2008: 5 upsets, 1 perfect region

It would seem recent history would suggest that you should have between 5 and 8 upsets in your Sweet 16, with either 1 or 0 perfect regions. So let's look at how we've picked this year.

2009 Group Summary
Brittany: 0 upsets, 4 perfect regions
Savan: 2 upsets, 2 perfect regions
Mark: 2 upsets, 2 perfect regions
Obama: 2 upsets, 2 perfect regions
Laura: 3 upsets, 1 perfect regions
Matt: 4 upsets, 2 perfect regions
Steve-O: 4 upsets, 1 perfect region
Niraj: 4 upsets, 1 perfect region
Me: 4 upsets, 0 perfect regions
Bobby: 5 upsets, 0 perfect regions


I admire Bobby as he was able to do what I couldn't. Although he always sucks at brackets. I mean, Ohio State to knock off Louisville? They won't get past Siena.

Of course, even if you anticipate 5 or 6 upsets, it only counts if you pick the right ones. So I understand the argument that if you're not sure where the upset will be, not to pick it. But come on, this is March Madness. Not March play it safe and hope everyone else screws up. Of course, my wife was the one who went all chalk. Whatever.