OT- Bracket picking quirks, anyone?

Kenpom

Every year is different.  

I always expect that there will be numerous upsets.  I always pick the wrong ones.

I feel dirty picking against my team, a good bracket be damned.

...Also, never more than 2 #1 seeds in the Final Four

I always have the best bracket.

outside of 3 seeds.  I also have Gonzaga losing in the second round.  

I pick a team that has a realistic, but smaller, chance of winning it all.  It puts all of your eggs in one basket but lessens the importance of the first few rounds.  If said team does win it all, or does especially well and someone else unheralded wins it, you're set.

I know this is a terrible strategy, and I've only won once (UCLA finals run 2006), but you look like a genius if you nail it.  Most years, bracket is toast after the first weekend, as it was last year (Mizzou winning it all).

In many of my brackets I have OSU in the final four.  I feel awful, everyone.  

I started doing brakets before I really started following college basketball because my friends were doing a pool and I decided to get in on it. Of course, I knew nothing beyond how teams were seeded and I didn't want to go with that, so I developed my own personal mascot challenge system. I choose every games based on which mascot would win in a fight. Sometimes it's obvious (eg. a Wolverine beats a jackrabbit 10 times out of 10) but sometimes it gets interesting and you have to do some reasearch. Does a Spartan beat a Trojan (yes. A Spartan would never fall for the old wooden horse trick) Does a commodore beat a boilermaker (again, yes. Presumably the commodore has a fleet, or at least a warship , at his disposal. The Boilermaker just has a train.)

I haven't done my bracket yet this year, so I haven't had a chance to look at any particularly interesting matchups, but if you need help deciding a game, I suggest using the mascot method. 

I've been picking against these handful of teams for years, going back to my younger years and before this notion about these teams became a bit of a cliche.  But Texas, Pitt, Notre Dame (and to some extent Florida St., especially when they were rated highly like last year) are locks to always bow out early.

I've always had a thing about the 13-4 matchups too, moreso than the 12-5 games even though those always happen.  But again, even before these past few years when an upset has happened, I've beel picking an upset amongst those 4 games for quite a few years now (usually picking a wrong one, but nonetheless).

I do at least one 12 seed over a 5 seed as well...this year I went with Oregon 

I'm avoiding a Michigan-VCU matchup entirely by picking Akron over VCU.  That should ensure Michigan gets to at least the Sweet 16, you know, because my brackets are always reliable.

Coin flips.  1 seed vs a 16 seed gets 5 flips (1/32 chance to succeed), 2s get 4  flips (1/16), 3s and 4s get 3 flips (1/8, so I'm roughly guessing a single 3 or 4 seed will get upset in every tournament), 5s and 6s get 2 flips, and the rest is a true 50/50 shot.

The Final Four and Championship is always a 50/50 shot.  It usually gives me a Cinderella team, which is always fun.  IIRC this year's Cinderella is Bucknell, with Albany upsetting Duke but not getting any farther.

I learned my lesson last year, where I actually looked into matchups and made "great" predictions but finished in last place.  Even the dog did better than I did (selected all dog mascot teams regardless of seeds, two dog matchups goes to the higher seed)

I check Vegas odds lines and if they are within 3 points or less I usually just go upset

Ohio State

Kansas

Duke

Miami

Oh, and for the record, I have Duke, Kansas, Ohio State, and Indiana. Kansas over Duke in the finals.

Without looking at the bracket, I automatically advance all four 1-seeds to the round of 32.

as well as the Vegas lines for the opening round.

The committee, honestly, isn't that great at seeding the best teams by ability 1-68, rather they may try and seed the best RESUMES 1-68.  Can be a big difference.

For example, 11 seed Minnesota is a 3.5 point FAVORITE over 6 seed UCLA.  Minnesota had a worse season and lost some games they shouldn't have, but one of UCLA's best players is out and Minnesota is probably more likely to play up to the level of their solid pts/poss differentials than play down to the level of their 12 losses.  

And Davidson/ Marquette might be the easies 14/3 upset pick in history, as Marq is only a 3.5 point fave.

I pick based on matchups only a little bit, usually when ive watched the teams a lot, but I try not to do it too much because, the longer the tourney goes on, the less likely each specific matchup is to actually happen.

I typically always pick whoever I think are the two best teams in the title game, with the best one winning.  I then look for 2-4 seeds matched up with weaker one seeds, to get a little variety in my final four picks.Also, I usually follow basic tips like advance all 1 and 2 seeds two rounds.

Depending on the size of the pool you are in, you want some variance and upsets, but not too much.  You also might want to cap your upsets at a round or two. Its usually fine to pick an 11 over a 6, but to have them beat a 3 seed too might be a bit much.

So this year, I think Indiana and Louisville are the two best teams, so theyre in my title game.  I think Gonzaga is good, but OSU is just as good and as a lower seed they have more 'bracket value', especially because they have a weak 3 seed (new mexico) in the S16.  And I have Florida as the last F4 team, their Kenpom gap over GTown, Kansas and Michigan is just too massive to ignore.  I dont think they're the best team as the rankings systems suggest, but theres just too much value picking them as a 3 seed (im already prepping for their two first wins to be massive blowouts before choking away a 1point game to Georgetown)

I always pick Wisconsin and Marquette correctly. This year I have Wisconsin losing to OSU in the Elite 8 and Marquette losing second (third) round to Butler. I hope my good luck with them continues.

Also, never pick Pitt in a close one, Belmont is an upset waiting to happen, and Louisville always goes one round further than you think.

I will be picking big 10 heavy this year.

Couple things...



Never more than 2 #1 seeds in the Final Four and at least 1 3 seed or lower.



Look at tournament history to come up with an average number of upsets based on seeding.



Use Kenpom to identify teams that may be over seeded and underdressed and come up with a list of likely contenders.

I tend to use the Massey ratings to get the estimated probability information on the Round of 64 and then build tables to get projected outcomes for the Round of 32. I typically blow an entire day of work on that essentially. I use the resultant potential matchups (and comparisons from StatSheet) to get a sense of where to go. 

From there, I tend to project no less than three brackets, the first one being fairly mundane and having only a small percentage of upsets. I then project one that has 2-3 upsets per region, and then my typical "Doom" bracket, where very little goes as expected (this one came in 4th in a pool once). I've had decent success and I've even won a few years here and there.

I try to pick the winning teams. Kenpom helps me. I usually finish in the money....especially last year.

I pick a "Worst March Ever" bracket where teams I like (Michigan and Pitt, maybe some other random places I have a bit of a connection with or good feelings towards) lose in the first round; usually the Final Four is MSU, OSU, ND and WVU. Thankfully this bracket has finished in a terrible position since I've done it. I also do one where it's all based on who I like more, again it's almost always M and Pitt in the title game/Final Four, sadly this one finishes low in the ranking as well.

Other than that, I just go with who feels stronger or has a better resume. I use Kenpom if it seems like he sees something the seeding committee doesn't, but other than that usually go by gut feeling.

Pick against Wisconsin. I put them in the final four years ago and they let me down. IIRC they never do well in the big dance. I have them loosing in my 12-5 upset this year.

I pick all Big Ten teams to do well.  Michigan wins the National Championship when they are in the tournament.  Georgetown and a 3/4 seed meet in the Championship when Michigan isn't involved.

If I have a tossup the win goes to the team with cooler uniforms.