There is only one thing more fun than watching the NCAA Tournament brackets be announced live on Selection Sunday: filling them in over the next few days to arrive at your own personal Final Four and, ultimately, national champion.
We all know that the NCAA Tournament is essentially a crapshoot. Oh sure, there are historical trends that make sense to follow (#1 seeds will make the Sweet 16, for example) and a wide spectrum of personal college basketball knowledge can help you see a Cinderella before any of the Thursday games tip-off.
But, we all see it every year: we enter into a family pool, all the guys have been watching every basketball game since January and have been poring over the numbers, and then Mom decides to pick “the underdogs and the teams with the prettiest uniforms” and ends up winning the bracket.
If something like this has never happened to you during March Madness you are either lying or Joe Lunardi. And hell, Joe Lunardi is only proven to be good at picking the brackets before they are announced. I’m sure he runs into the same problems as the rest of us when trying to predict the outcome of actual tournament.
Because of the randomness of it all, I usually just eyeball my brackets and go with first impressions. Some years I pick a lot of upsets, some years I don’t. And usually I do three, four, five different brackets in an effort to diversify my picks and give myself, theoretically, a better chance of at least winning one pool.
Well now that I have this blog, and have decided it is a good idea to publish every infinitesimal and meaningless thought I have about sports, I feel like there is more at stake with my picks this year. So I decided that I wanted to do just one bracket, and to have a specific system I used for filling it out. (Plus, I took one look at this year’s bracket and realized that so many of the games seemed like toss-ups, I needed some kind of analysis I trusted to fall back on and break the ties.)
Let me be clear right off the bat: In no way do I endorse the following system as the best NCAA Tournament bracket system possible. Nor am I wholeheartedly convinced that it is even necessarily good; nor, truth be told, did it really require a whole lot of in-depth thought to come up with.
But…after devising the system and using it to go through the entire bracket, the results were as follows:
- Higher seeds, especially the top-line seeds, won most of the games. This is usually what happens, and what I expected, so I was pleased.
- One #12 seed beat a #5 seed. This always happens, so again, I was pleased.
- A darkhorse Sweet 16 team (13th seed or lower) emerged. There is often one very low seed that makes it out of the first weekend; but rarely more than one or two.
- A #2 or #3 seed lost in the first weekend. Out of 8 such teams, there is usually one or two gone before the Sweet 16. This held true in my predictions.
- The Elite 8 and Final Four is dominated by high seeds. Again, while the first few rounds of the tournament make headlines because of upsets, when the dust settles for the final three rounds, the majority of the teams are usually the teams with the best seeds.
So, I did not fudge my system in any way during the selection process, and the results were what I “hoped” to get. By this I mean that my system did not place 4 #10 seeds in the Final Four, or just give me all favorites winning. It actually matched up pretty well with typical tournament trends that we see from year to year.
Before I jump into my actual picks, here is a description of my subjective-objective system, which, if it proves successful this year, may be my new system for picking the NCAA Tournament moving forward.
The JRod Subjective-Objective System for Picking the NCAA Tournament (Which Hopefully is More Successful Than Throwing Darts or my Mom’s Picks)
Why do I call it a “subjective-objective” system? Because I sat down and defined what kinds of quantifiable team stats mean the most to me when deciding what team will beat another. My choice of metrics was obviously very subjective. But once that choice was made, there was very little room for any other subjective decision-making, with one caveat — which I will get to in a bit.
Here are the metrics I used to measure each matchup on a head-to-head basis:
- Regular Season Road Record
- Tournament Seed
- Strength of Schedule
- Team FT%
- Team Turnovers per Game
- Experience of Guards
A few quick notes about the metrics:
- I understand that these are not perfect. Road record, for instance, does not take into account a team’s record on neutral sites, which is perhaps an even more accurate reflection of how they will do on neutral sites in the NCAA Tournament. There are plenty of other problems and holes in this system. Feel free to point them out in the comments if you wish, but just know that I already know they exist.
- Turnovers per game does not take into account assist/turnover ratios, number of possessions per game, style of play, etc. Comparing TO/game in a vacuum across teams is obviously not completely “fair” or accurate.
- To measure the experience of the the guards, I used the information available at the CollegeHoopsNet.com team tournament capsules. For each team’s projected lineup, every guard was assigned a value of 1, 2, 3, or 4, based on their class. Freshman were a 1, Sophomores a 2, Juniors a 3, and Seniors a 4. The sum was then divided by the number of guards to get an average. Obviously the higher the better. Again, this is not perfect, but I like teams with experienced, older guards, so it works for my subjective statistical metrics.
- For tournament seed, the better seed got the advantage. Why? Because more often than not, the better seed wins — except in the case of 8/9 matchups, where 9 seeds actually win 53% of the time. Hence, to play the averages, the 9 seed got the seed advantage in 8/9 games. But I thought this was important because the selection committee spent hours upon hours comparing these teams, and I generally trust their ability to tier the teams.
- In the case of a tie (i.e. each team winning three categories) the team with the better seed wins the tie-breaker and moves on. Now, here’s the caveat to complete objectivity: If there is a tie between two teams whose seeds are adjacent (a #2 seed playing a #3 seed or a #1 seed, for instance) then I use my own personal gut feelings/bias to break the tie. This only happened a few times. 95% of the games were “picked” using the objective results of the system.
If anything is unclear, leave me a question in the comments and I’ll try to clear it up. Again, I am not claiming that this system was particularly well thought out or perfect. However, the relative simplicity of the stats did allow me to compile everything into a spreadsheet in about 2.5 hours, it does take into account the traits that I personally value, and the results ended up producing a bracket I would have been happy to end with if I had just run straight through it off the top of my head.
If you are interested, here is the actual bracket breakdown spreadsheet I used so you can see the actual data. You will notice that I did not list the #16 seeds. Sorry guys, but I knew nothing would make me pick a #16 over a #1. Perhaps that will burn me, but I feel fairly confident going with history on that one.
All that said, let’s break down the regions:
Midwest Region Predictions
First Round Picks – Midwest Region
- #1 Louisville
- #9 Siena over #8 Ohio State
- #5 Utah over #12 Arizona
- #4 Wake Forest over #13 Cleveland State
- #6 West Virginia over #11 Dayton
- #14 North Dakota St over #3 Kansas
- #7 Boston College over #10 USC
- #2 Michigan State over #15 Robert Morris
Siena and Ohio State were actually pretty evenly matched up, with Siena having a slight 3-2-1 advantage. Siena has a much better record on the road and more experienced guards, so I feel comfortable about that one. But it’s an 8/9 game so who the hell knows.
The one that will jump off the page is #14 North Dakota State over #3 Kansas. NDSU went 10-4 on the road this year, makes just under 74% of their free throws, has all senior guards, and only turns the ball over 11.3 times. Their Strength of Schedule is obviously not very good, but they played outstanding against their level of competition. Considering Kansas’ poor play very late in the season, and their relative youth, I am not afraid of this upset pick. If I had to bet money, I’d obviously put it on the Jayhawks, but now I won’t be at all shocked if they go down, much to the dismay of Bill Self.
Second Round Picks – Midwest Region
- #1 Louisville over #9 Siena
- #5 Utah over #4 Wake Forest
- #14 North Dakota State over #6 West Virginia
- #2 Michigan State over #7 Boston College
I like Utah over Wake Forest because of the experience factor, the fact that Utah turns the ball over more than two times per game less, and Utah’s ability to hit free throws at a 78.2% clip. Wake Forest clearly has the more talented team though; so again, this is a game where me being wrong would not be surprising at all.
And once again, you see my tourney darkhorse North Dakota State winning again. Considering the numbers I stated above, does this surprise you? West Virginia shoots less than 70% from the line, has relative inexperience at the guard position (a 2.5 score to NDSU’s 4) and turns the ball over slightly more.
Sweet 16 Picks – Midwest Region
- #1 Louisville over #5 Utah
- #2 Michigan State over #14 North Dakota State
Louisville beats Utah across the board, except for free throw shooting. The Michigan State-North Dakota State game was actually a tie by the categories. Michigan State wins in road record, seed, and strength of schedule, but falls short in FT shooting, guard experience, and TO/game. Non-adjacent favored seeds win out though, so NDSU’s Cinderalla story comes to an end.
Elite 8 Pick – Midwest Region
#2 Michigan State over #1 Louisville
Honestly, this surprised me a little bit. When I first sat down to look at the brackets, the Spartans were a team I looked at as potentially being ripe for an early upset. But looking deeper, many of their metrics, at least with respect to my personal basketball biases, make them a tournament-ready team. Add in the extra motivation of the Final Four being in Detroit, and I can see this happening.
The Spartans win 3-2-1, with the teams tying in guard experience and Louisville besting Michigan State in seeding and TO/game. Michigan State’s other advantages were by slim margins, which is to be expected. I think this would be a great regional final that could go either way (a refrain you will probably hear repeated as we move forward), but as a Big Ten fan I would love to see Tom Izzo and the boys make it to Detroit.
West Region Predictions
First Round Picks – West Region
- #1 UConn
- #9 Texas A&M over #8 BYU
- #5 Purdue over #12 Northern Iowa
- #4 Washington over #13 Mississippi State
- #6 Marquette over #11 Utah State
- #3 Missouri over #14 Cornell
- #7 California over #10 Maryland
- #2 Memphis over #15 Cal State Northridge
No real surprises. BYU-Texas A&M was a 3-3 tie based on the metrics, but I chose Texas A&M because they have slightly more experienced guard play and played a slightly tougher schedule. Again, total toss-up, could go either way — but you have to pick one!
Second Round Picks – West Region
- #1 UConn over #9 Texas A&M
- #4 Washington over #5 Purdue
- #6 Marquette over #3 Missouri
- #7 California over #2 Memphis
The top part of the bracket goes according to seed. I actually thought Purdue would go a little farther when I first embarked on this analysis, but Washington just edges them out. The surprises come in the bottom half of the bracket.
Marquette defies seeding to beat Missouri, but I don’t really consider this an upset. Even without Dominic James, Marquette has vast experience in its backcourt, they don’t turn the ball over much, and they make 72% of their free throws.
Memphis losing this early surprised me a little, but I do not think they are nearly as good as they were last year. Cal actually bests Memphis in FT%, SOS, guard experience, and TO/game. I think Memphis’ inexperience and lower level of competition since January hurts them and they make an early exit. I have no qualms going with the stats and picking Cal here. (Side note: Is John Calipari screaming, crying, or yawning in that picture?)
Sweet 16 Picks – West Region
- #1 UConn over #4 Washington
- #6 Marquette over #7 California
UConn is very strong across the board with respect to the stats I chose, and they pretty much dominate Washington. Marquette-Cal is a relatively even matchup, but the experience and slightly better TO/game numbers give Marquette the edge.
Elite 8 Pick – West Region
- #1 UConn over #6 Marquette
An all Big East regional final in the West would be great. The problem for Marquette is that UConn is just a little bit better at a couple more categories. UConn went 10-1 on the road this year and has an all-senior backcourt that includes the great AJ Price (who I have been a big fan of ever since he dismantled Indiana last season). Plus, while this is not really taken into account statistically, having a 7′3 big man down low in Hasheem Thabeet obviously gives UConn a dimension that most teams do not have. I am comfortable sending the Huskies to the Final Four.
East Region Predictions
First Round Picks – East Region
- #1 Pittsburgh
- #8 Oklahoma State over #9 Tennessee
- #12 Wisconsin over #5 Florida State
- #4 Xavier over #13 Portland State
- #6 UCLA over #11 Virginia Commonwealth
- #3 Villanova over #14 American U.
- #7 Texas over #10 Minnesota
- #2 Duke over #15 Binghamton
Oklahoma State-Tennessee was a 3-3 tie. However, one of the categories Tennessee “won” was being a 9 seed over an 8 seed. Considering the fact that OSU has more experienced guards, turns the ball over less, and shoots better from the line, I like them to win. And for goodness sakes, 9 seeds only win 3% mo
re than 8 seeds, so it’s not like the 9 over 8 advantage should determine every one of these matchups.
Wisconsin is my #12 seed that pulls the upset. The Badgers burned me in the Big Ten Tournament (I picked them to win it) and this pick scares me a little bit considering how well Florida State played in the ACC Tournament. However, Wisconsin has more experienced guards (by class anyway) and turns the ball over a staggering 5.6 times less per game. I actually look at this game as a toss-up, but I feel comfortable going with the Badgers in the upset.
Interesting note: VCU over UCLA is a trendy first round upset pick. Led by former tournament star Eric Maynor (remember VCU’s upset of Duke a couple years ago?) a lot of people, including Jay Bilas, think VCU could/will beat UCLA. However, this was the only first round game (minus the 1/16 battles, which I did not analyze) in which one team won all six categories — and it was UCLA doing the winning. I had actually anticipated picking VCU, but based on the analysis, I now feel pretty good picking the Bruins.
Which, of course, means that VCU will still probably win. Oh well. Such is March Madness.
Second Round Picks – East Region
- #1 Pittsburgh over #8 Oklahoma State
- #12 Wisconsin over #4 Xavier
- #3 Villanova over #6 UCLA
- #2 Duke over #7 Texas
Well, here I go again banking my credibility on Bo Ryan and the Badgers. Their experience, protection of the ball, FT%, and SOS all trump Xavier. This worries me a bit because Wisconsin has been so up and down this year, but they do play the kind of basketball I respect, and have made noise in the tournament before. By no means would I be shocked to see Wisconsin lose here (or in the first round for that matter), but I’m sticking with the Badgers to move onto the Sweet 16 and help restore some cred to the Big Ten.
Sweet 16 Picks – East Region
- #1 Pittsburgh over #12 Wisconsin
- #3 Villanova over #2 Duke
Halelujah, Pittsburgh beats Wisconsin. I was really worried that I might have devised a system that would choose this game in a way I would have a really hard time justifying. But Pitt was better on the road and has more experience in the backcourt. It ended up a 3-3 tie, with Pitt’s tremendous seed advantage tipping the scales.
The Villanova-Duke matchup was 3-3, which should go to the higher seed. However, because it’s a 2-3 matchup, I gave myself the wiggle room to be subjective in my choice. Duke is playing well, but I love ‘Nova’s backcourt and I just don’t see this Duke team as being built for a long tourney run. ‘Nova moves on for another all Big East Regional Final.
Elite 8 Pick – East Region
- #1 Pittsburgh over #3 Villanova
As you can tell, my subjective choices of metrics obviously prove that I am a fan of the Big East’s brand of basketball. Through three regions, I have chosen five Big East schools to make the Elite 8. Villanova is far superior at the line, but Pittsburgh faced a tougher schedule and has more experience guards that turn the ball over less. Jamie Dixon and the Panthers move on to Detroit.
South Region Predictions
First Round Picks – South Region
- #1 North Carolina
- #9 Butler over #8 LSU
- #5 Illinois over #12 Western Kentucky
- #4 Gonzaga over #13 Akron
- #6 Arizona State over #11 Temple
- #3 Syracuse over #14 Stephen F. Austin
- #7 Clemson over #10 Michigan
- #2 Oklahoma over #15 Morgan State
A pretty much by-the-seed first round in the South. There are a few games I would not be surprised to see go the other way (WKU over Illinois, Michigan over Clemson, for example) but I have no issue
s with any of these games. Butler and LSU was close, and LSU actually has more experience and turns the ball over less, but Butler actually played a tougher schedule and was better on the road. I like rooting for the Bulldogs anyway, so I’m glad this turned out how it did.
Second Round Picks – South Region
- #1 North Carolina over #9 Butler
- #4 Gonzaga over #5 Illinois
- #3 Syracuse over #6 Arizona State
- #2 Oklahoma over #7 Clemson
Chalk reigns again as the top four seeds all advance. Illinois versus Gonzaga was very close, but I definitely would have picked Gonzaga in this one without the stats, so I feel good that it turned out this way — especially with Illinois not having Chester Frazier. Syracuse-Arizona State was won by the Orange based on their seed, but I would not be shocked to see ASU win. Syracuse turns the ball over a lot and is not good at all from the FT line. But I do love their backcourt and think Jonny Flynn is one of the ten best players in America and could lead my team anytime.
Sweet 16 Picks – South Region
- #1 North Carolina over #4 Gonzaga
- #2 Oklahoma over #3 Syracuse
Chalk again. And neither is really all that close. Surely Ty Lawson would be healthy (or healthy enough) by the Sweet 16, and I think UNC would need him to beat a really good Gonzaga team. Oklahoma beats Syracuse in everything but SOS. And honestly? It’s about time a Big East team did not make the Elite 8. With a healthy Blake Griffin all year, Oklahoma probably gets a #1 seed. This is not a surprise to see them advance so far.
Elite 8 Pick – South Region
- #1 North Carolina over #2 Oklahoma
The Tar Heels actually beat Oklahoma in every category. Before looking at the stats I wondered who I would pick just based on the eyeball test. But it’s hard to argue with a system that I devised saying North Carolina is better in every category. The Tar Heels head to the Final Four.
Final Four Predictions
- #1 UConn over #2 Michigan State
- #1 North Carolina over #1 Pittsburgh
UConn bests Michigan State in four categories, and I think the Huskies are the better and more consistent team overall. The Spartans valiant run to Detroit ends here.
North Carolina-Pitt is very close statistically, with Pitt actually holding the edge in guard experience and TO/game. I would not be surprised to see Pitt advance and make it an all Big East championship game, but North Carolina was a better team on the road and is almost a full 10% better at the FT line. This would be a GREAT Final Four game, with the slight edge going to the Tar Heels.
2009 NCAA Men’s Tournament Championship Game Prediction
#1 UConn over #1 North Carolina
Based on the metrics, UConn wins this 3-2-1. Each category is very close (save for UNC’s wide edge in FT%) but UConn’s slight advantage in backcourt experience and TO/game gives them the edge. Add in the questions surrounding Ty Lawson’s health and I’ll take this.
If I had just eyeballed the brackets, I’m not sure either one of these teams makes my title game, but it’s pretty obvious that they are proficient in the areas I value.
I’m not a big Jim Calhoun fan, although I think he’s obviously a great coach. But I have to say that banking my entire bracket on a team led by AJ Price, and that has Hasheem Thabeet down low, does not make me overly nervous. UConn was generally considered one of the top two teams in the Big East all year (with Pitt — and then Louisville came on strong at the end) and the Big East was clearly the best conference in America. It is fitting then that a Big East team ends up winning it all.
My official one-and-only bracket for the 2009 Tournament:
——————–
——————–
So there you have it — my official subjective-objective predictions for the 2009 Men’s NCAA Basketball Tournament. All this really means is that I now have more time and thought invested into getting my bracket horribly and drastically wrong. But I will say this: I have more confidence heading into the first Thursday of March Madness than ever before.
I fully expect that confidence to be completely eroded by Thursday night.
Loading ...
