Filling out your brackets: Any good tips?

Submitted by profitgoblue on

March Madness has us all in its grips, squeezing the picks out of us.  Every year I sweat and swear as I fill out my bracket, knowing all along that my first round picks will be decent and then my bracket will go to sh-t.  Its all I can do to have even just one team in the damn Final Four.  So I was wondering if you all have any interesting ways of picking "toss up" games or filling out your backets in general.

One interesting tidbit I found on Yahoo is this"

Make difficult picks using the "mascot fight" theory

Two of the hardest games to pick in last year's first round were UCLA-Michigan State and Vanderbilt-Richmond.  They were veritable toss-ups. But looking at these games in retrospect, they were all too easy to pick. Bruins vs. Spartans?  Hmm, a vicious bear or a person who shuns the conveniences of modern living?  I saw "Grizzly Man."  I know how that story ends.

http://sports.yahoo.com/blogs/ncaab-the-dagger/fill-2012-ncaa-tournament-bracket-024636059.html

(I deleted the reference to "Friends" in that blurb because, well, "Friends" and NCAA basketball should never be uttered in the same discussion.)

Fork over your secrets!

 

 

Wolverman

March 12th, 2012 at 11:48 AM ^

 In my experience it's always the person who knows the least about college basketball that wins lol. I got a final four of Kentucky, Syracuse , Missouri and North Carolina. I got Michigan in the sweet 16 but losing there to North Carolina, Indiana losing in the 2nd round to UVC, Ohio state losing to Cin and MSU losing to Missouri both in the elite 8. It's a tough year to make a bracket because everyone has been so inconsistant. 3-4 teams in the big ten will look like NC caliber teams one week and make you wonder how they are ranked a week later

aiglick

March 12th, 2012 at 12:03 PM ^

I think the Big Ten is head and shoulders above every league except the Big 12 with ACC a little behind the leaders. It is probably smart to pick UNC over us in Sweet 16 but we have handsome success against athletic teams (Tenessee and Memphis more recently come to mind). I also like Beilein as a coach better than Roy. Don't get me wrong they are both great coaches but I think Beilein has the x's and o's advantage.

I can't bet against Michigan but I do think we can be a sleeper. Players may underestimate us based off of our BTT performance. Generally we play up to competition so the first two rounds make me nervous a little bit. At least the Madness is almost here.

Blue-Chip

March 12th, 2012 at 12:05 PM ^

Pick M to get bounced early.

This way, if they make a deep run, you don't care about the fact that your bracket sucks. If they do go out early, you can say at least my bracket still looks good.

Mgoscottie

March 12th, 2012 at 12:05 PM ^

after that I tend to think about probabilities and competing against other brackets.  For example, last year I think I took Richmond to win 2 games as a 12 seed, at that point you want a high seed to beat them in your bracket.  If they were to win and go to the elite 8 it's unlikely someone else would have picked that and even if your high seed doesn't win, fewer brackets will get a leg up on you.

Final four I try to always have 1 really likely team and 1 unlikely team (3-5 range) so that I'm usually still interested in the final weekend as sports becomes boring around this time without it. 

brandanomano

March 12th, 2012 at 2:27 PM ^

I also had Kansas winning, and I don't feel great about picking them over Detroit. I also had Duke, Memphis, and Syracuse to round out my final 4. I don't think I'll have any of them when I eventually fill out my actual bracket.

M-Wolverine

March 12th, 2012 at 12:18 PM ^

whatever...because you may get lucky, but odds are you'll just be out the first weekend...I've just got a couple (serious) ones.

Take a look at where teams are playing.  It can influence a lot. The "oh, I'll predict North Carolina to get upset in the second round" the year they're playing in Greensboro....or when the Elite 8 is in Syracuse and....Syracuse is playing there, you have to like their changes (speaking of past years, not this year necessarily). 

And even though I may go against this one a bit this year, think of the probablities.  If there are a bunch of equal teams you think can win...who has the easiest road?  The "Duke draw", as it were.  Because you can't win it all if you can't get there.  It's harder to have to win 2 or 3 tough games than 1 (duh), and the more good teams there are in the bracket, not only is there a tougher 2 seed, but the potential that if the 2 seed gets upset, you'll STILL have to play a really good team to get into the Final Four.   Where an upset in another region means you get to play Whattsamatta U. to get to the Final Four.  Those that team has a better chance to win it all.

If you're not in a pool where the final game counts for like 64 points or something, and you have enough people in it that everyone doesn't have their own champion (odds are more than one person will have your team), it's really about who gets the most teams in the Final Four. Spend as much time concentrating on that as you do picking that awesome 1 round upset only worth 1 point.

Oh, and if you're going chalk in the 8-9 games, go 9, because they win more than the 8 seeds do.

UMGooch

March 12th, 2012 at 12:56 PM ^

I won my last three brackets. Once for a case of beer, then $50 each year after. I follow some basic rules. I know very little about basketball, and I don't really follow the NCAA season unless Michigan is doing well...

1. Mostly follow the seedings. #16 has never beated #1 in history. Don't really start looking at upsets until #4 vs #13...

2. Look at the conference. Seedings seem to take into account overall records strongly. Obviously, the ACC, B1G, and Big 12 teams have a ton of experience against big teams. Some of these middling conference teams haven't even faced any top 25 teams this year... I don't have a ton of faith in them. Sure, you have your VCU and George Masons, but rarely does anybody pick them to win more than the first rounders, and if they do, they likely won't be in your pool.

3. Check each team's performance through the year. Did they start off soft and start winning more later in the year? Did they pull out some big wins against ranked opponents? Teams will bring everything they have to the tournament. If they've pulled out some big upsets through the year, they have the ability to do it in the tournament.

4. If all else fails, pick the team that has a better basketball history. UConn, Duke, Syracuse, Kentucky and the like rarely disappoint in the tournaments.

Trebor

March 12th, 2012 at 1:00 PM ^

I like to play percentages a bit, especially in the first round where the numbers are really easy to find. For example, 1 beats 16 100% of the time in the 27 years of the 64+ team field, so I always have the 1 seeds win. 15 has beat 2 only 4 times in 108 tries, so I always advance the 2 seeds (why attempt to chase points when the downside is missing a potential Final Four team?).

3 seeds have fallen 16 times; if I'm feeling frisky and I really like a matchup, I'll consider it, but only if the 3 seed is a team I think is incredibly overrated and the matchup doesn't look good. If anyone this year, Iona over Marquette, but Iona's gotta get through BYU first.

4 seeds have fallen 23 times. This is good for a win a year on average; however, I usually advance the 4 seeds anyway unless, like above, it's a game I think is a matchup problem. This year, sadly, Michigan is probably the most likely 4 seed to lose, but I can't in good faith pick that as an upset. Wisconsin maybe...

5 seeds have fallen 36 times. This happens at least once a year; it's tough to choose sometimes, but 5 seeds don't make big runs too often so it's worth it sometimes. I'm looking between VCU-Wichita and New Mexico-Long Beach games for one to happen. 6 seeds have fallen 36 times as well. I'm thinking either Cincinnati-Texas or NC State-San Diego State.

7 seeds have fallen 43 times. This year, all of the 7-10 matchups look like toss-ups to me. I'd go with Purdue and WVU as most likely to win as 10s. 8 seeds actually lose more than they win, but it's basically 50/50. Choose two 8s and two 9s; they'll probably fall against the #1 seed anyway.

For the second round, I don't usually like taking any of my double-digit seeds here unless I match up a 12-13 game. 1 seeds win about 85% of the time in the second round, so I'll usually pick all of my 1 seeds to make the Sweet 16 (8/9 seeds don't usually make it much further past there, so again, why chase points). The 2 vs. 7/10 matchup is also slanted heavily towards the 2, and I don't think any of the 2s this year look primed for an upset. Purdue might give Kansas a good game.

The third round is where it always gets tough (points are much tougher to come by) - 1 seeds only win about 75% of their third round games, so picking one to fall here isn't a bad decision. 4s beat 1s about 30% of the time, 5s beat 1s about 20%. If you've got chalk to this point, reminder that Indiana has already beaten Kentucky. I don't think it will happen again, and I actually like the 1s to win any 4/5 matchup, but food for thought. The 2s and 3s are more or less toss-ups at this point. Elite Eight and beyond is a luck-of-the-draw. 1s still have the advantage, but only once in history have all the 1s advanced to the Final Four, while 4 times there hasn't been a single one.

Winning points on upsets is awesome, but losing points can be costly. Pick some upsets, but don't overdo it. Chalk brackets will finish top-25% in most pools but never win - so my suggestion is to start with a chalk bracket and then sprinkle in a few upsets along the way, knowing that at least one (if not more) 1 seed won't be in New Orleans.

M-Wolverine

March 12th, 2012 at 3:43 PM ^

The only thing I'd caution is taking upsets just to take one because they're probably. If you feel like it's a good shot, go for it, but don't just pick one...because if only 1 of 4 is upset annually, you might be missing your pick, AND the actual upset.  Likewise picking 2 8's and 2 9's are great if you nail them...but if you pick the wrong two, and the other two win, you get all 4 wrong and are bleeding points.

Trebor

March 13th, 2012 at 8:00 AM ^

Yeah, I totally agree - that's why I noted that in some instances I'd go against what the historical stats are (like, I don't see any bad matchups for the 3 or 4 seeds this year even though historically at least one and maybe 2 lose every year). The 8/9 thing was more of a "it's a crapshoot anyway most of the time" statement - I still take the team I think has the better matchup and on average that ends up as 2 8s and 2 9s.

However, the big thing is you HAVE to pick upsets at some point. If you don't, you can't win. A full-chalk bracket (I fill out one a year on Yahoo or ESPN just to see how well it does) will generally be top-25% (it was top 5% I think the year that Kansas beat Memphis since all four 1s made the FF, the only time in 27 years), but will never be top 1% (where the money is in most pools worth entering). The thing is to be smart with upset picks, and don't pick too many since a lot of games do go chalk, but upsets will happen and sometimes you get lucky while others you get burned. It's all part of the March Madness fun, right?

M-Wolverine

March 13th, 2012 at 12:17 PM ^

But mis-seedings. Say, you think Vandy should be higher rated than Wisconsin. Or one thinks Missouri should have had a 1 seed over MSU. Or whatever.

It used to be the equalizer was the Final Four, because there was no "chalk" there. You could go by one of the polls for the higher rated teams...but they didn't always match.  Now that they went from it being  a mystery, to telling you who was 1 overall, to telling you who was 1-4, you can really even chalk up that.  It at least used to be the variations of the Final Four you could pick. Not so much anymore.

julesh

March 12th, 2012 at 1:13 PM ^

I pick Michigan to go all the way, always, and then kind of randomize it after that. I've actually done pretty well compared to the people I play with using this method.

will

March 12th, 2012 at 4:18 PM ^

I always pick UM to lose the first game they are a toss up. This way, if we win - I'm happy. If we lose, I can at least console myself that my bracket is doing better.

I used the same strategy in vegas last year for the game against Duke...Mind you it wasn't very consoling when Darius's floater rimmed out. I proceeded to break my wedding ring on the bar at Mandalay Bay..

R Kelly

March 12th, 2012 at 1:16 PM ^

My Tips:  1.)Make only one bracket, or at least make the same picks in each pool you enter.

                 2.)  Pick the teams you like to win.  (within reason)

Its more fun to watch when you're not feeling conflicted about who to cheer for.

Sambojangles

March 12th, 2012 at 1:19 PM ^

Play a lot, for many years. Pick a lot of upsets (that are reasonable) and hope for the best.

The way I see it, the tournament is a bell curve, with chalk being the median. What actually happens will be on a tail, but you obviously don't know which one it will be ahead of time. If you want to win, pick semi-randomly and hope for the best. You might fail spectacularly by having an 8 seed in your Final Four, or it might all work out. Chalk is good for an average score almost every year, but of course that will almost never pay off.

Also, don't worry about the first round, it's mostly for fun. The money is made by getting 3 or 4 Final Four teams right, and picking the actual Champion. Everything else is just window-dressing if you are in a big enough pool. The only time I have ever won was when UConn won in 03 (somewhere around then) because I was the only one to pick them.

BeatOSU52

March 12th, 2012 at 1:58 PM ^

"Also, don't worry about the first round, it's mostly for fun. The money is made by getting 3 or 4 Final Four teams right, and picking the actual Champion"

 

That is correct if is "standard scoring" where each round it doubles up (1-2-4-8-16-32) but some pools have customized scoring like: 1-2-4-7-11-16 which makes it so the first round matters a little more but the later round games are still more important.

M-Wolverine

March 12th, 2012 at 3:40 PM ^

Has a 1, 2, 4, 6, 8, 10 point system. 32 points is way too many for 20 people to nail "Kentucky" and win over someone who may have (theoretically) had an absolutely perfect bracket up to that point. Now if you were close and nail the champion, you deserve to win...but you need to reward the more difficult accomplishments.

M-Wolverine

March 12th, 2012 at 11:18 PM ^

I think there's a huge (pool) difference between 16 and 32. One throws off the weighting of the system, where the other is just a reward.

And I don't really have a problem where people who picked the winner can't win...because it means someone else nailed a whole lot right.

legalblue

March 12th, 2012 at 1:30 PM ^

I was just discussing this with some coworkers/troglodites and we realized there are 2 to the 63 power number of ways to fill out a bracket that's 9 quintillion plus possibilities.  

ofcourse picking at random isn't a great strategy but we could then factor in the likelyhood that any particular seed will beat another seed.  Someone earlier in the thread had a good breakdown but if you look over time a 1 seed has never lost to a 16 and its exceedingly rare that a 2 seed loses to a 15 though it does happen from time to time.  Over the course of a number of these type of calculations.  Which I assure you are mind numbing and tedious (what is the likelyhood a 1 seed beats a 14th seed? has it in fact ever happened that a 1 seed has played a 14th seed?) you should get some distribution of all the possible bracket combinations that shakes out into a bell curve if you plot the various combinations over their calculated percentage to win. 

Even then a standard deviation from the best probability to win (which I think but have no emerpical evidence to support would be a strait lowest seed advances bracket) would be an astronmical number of selections the curve is likely VERY flat with the number of data points we have in combination with the very low probability that any one bracket is a winner we could have 100 trillion (or more) brackets within in one standard deviation from our mean best chance bracket. 

from a purely mathmatical point of view you have a better chance of being struck by lighting while getting eaten by a shark that is currently winning the lottery.  

That being said this isn't abotu the perfect bracket so much as it is the bear problem.  The bear problem is simple two hikers see a bear one put on the running shoes while the other chortles in a worried sort of way telling his compatriot; "Fool, you can't out run a bear".  The running shoe donning hiker replie coldly in a callow and calculating manner that he needn't out run the bear just the other hiker.

/best of luck though  

BeatOSU52

March 12th, 2012 at 2:02 PM ^

if you're going to pick the favorite to win it all such as Kentucky or UNC, that is fine, BUT, you should probably pick a few more upsets than usual due to the many amount of people picking those two teams.

 

if you pick a team almost nobody else will pick like a FSU or even a OSU, then you don't need to have as many upsets picked.

Trebor

March 12th, 2012 at 2:29 PM ^

Since you asked, a 1 seed has never played a 14 (or a 15). Each of the 4 times a 15 won an opener, they lost in the second round. Of the 16 times a 14 won an opener, only twice have they won a second round game, and are 0/2 in Sweet 16 games.

If you play smart with the historical percentages, the chances of a perfect bracket get a lot better. Of course, by a lot better, I mean going from, say, (1/2)^63 to (1/2)^55, which is still next to impossible. Maybe some day we'll see someone have a perfect bracket, but we'll probably see a 16 beat a 1 before that happens (yeah, it's FAR more likely to have a major upset like that, but there are also hundreds of thousands, if not millions, of brackets filled out every year).

Perkis-Size Me

March 12th, 2012 at 1:35 PM ^

kind of like what you touched on, there really arent any "good" tips for filling out brackets. at least 2-3 teams get upset every year that shouldn't (louisville v. morehead state, kansas v. northern iowa), despite how much of a solid pick they look like. the only picks i ever feel truly safe about are a 1 v. 16 matchup.

all you can do is pick the teams you have a good feeling about, and then hope for the best.

WingsNWolverines

March 12th, 2012 at 4:08 PM ^

Somewhere in the mix. My two upsets are Vandy knocking out Syracuse in the Elite Eight and Ohio St being knocked out in the Sweet Sixteen by FSU

will

March 12th, 2012 at 4:23 PM ^

Living in Columbus, I know my friends and coworkers are all over-zealous OSU homers.. Inevitably 25% of them will pick OSU to win it all.

Knowing that, I never pick OSU to make the final 4, because the chances I will beat them ALL out if OSU makes it that far are slim. I will often pick the 1 seed from OSU's bracket just to play the odds...

If you have alot of UM homer friends, having them lose earlier might result in some easy points..