Tuesday, April 1, 2014

2014 Final Four: Pomeroy + LRMC Profiles

If you recall from my "Final Thoughts" post, I thought that this tournament could likely resemble the 2011 bracket. In 2011 there was a 3, 4, 8 and 11 seed, and the group included Kentucky and UConn. This year we have a 1, 2, 7 and 8 seed, which includes Kentucky and UConn.  Go figure.

I think we end up with a Gators v Wildcats championship game with Florida cutting the nets down. 


Tuesday, March 25, 2014

2014 Bracket: Elite Eight and Final Four Picks

The Sweet 16 is here, so let's check in on the teams who made it and how it stacks up to the profiles we created back in Part 3 and the Deep Sleepers Post. The teams highlighted were a part of the original list of most likely candidates to make the Elite 8, while the last three teams were all noted as teams with a profile to be a Deep Sleeper to the Elite 8.



To read the "Change" numbers, if a team has a "Positive (+)" change, it means they improved their OE or DE, and vice versa. Note that all of the Deep Sleepers have significantly stepped up their Defensive Efficiency.

Based on this past weekend's games, and sticking the likely profiles for the Elite Eight and Final Four, here are my picks for the weekend:

  • EE: Florida and Dayton -> FF: Florida
  • EE: Arizona and Wisconsin -> FF: Arizona
  • EE: Michigan State and UConn (If Niang was playing, this would be Iowa State because the pre-tourney 80 OE for UConn is too high for the model.) -> FF: Michigan State
  • EE: Louisville vs Tennessee -> FF: Louisville

Looking forward to another great couple rounds of games and seeing how the picks hold up.



Wednesday, March 19, 2014

2014 Bracket: Picks and Final Thoughts


In the end I think this year's bracket has a really good chance of looking like 2011's.  That said, I think we should be in for some good surprises this year, and I've tried to pick a few of them.  Only taking a few gambles, but ultimately my Elite 8 field holds true to my numbers, so here goes nothing...


Key Upset Picks:

  • New Mexico St to Sweet 16. See some of my previous posts for details. 
  • Baylor to Sweet 16. 
  • Kentucky to Elite 8. KY, Louis and WSU all fit the Elite 8 profile to a "T". I was really high on Louisville before I started digging into the details.  Every Louisville loss was to a ranked opponent, and I fear their #1 LRMC is inflated due to blowouts against really weak opponents. I think the Shockers are paper champions as well, and the Wildcats get them with the size.  If Louisville wins it all, I will follow them blindly in all future years.


Random Notes:

  • Syracuse or Kansas for Elite 8.  If Embiid was playing it would be a no-brainer and I'd actually pick them to win it all. Given he's out for the whole first weekend and maybe longer, I just can't roll the dice on the Jayhawks.   
  • If not obvious below, I think it was Michigan State and not Louisville who was actually under-seeded.  The Spartans have far better wins, and won a significantly better conference tournament.  They somewhat remind me of the 2011 UConn team that won the Big East tourney and had a very similar LRMC/OE/DE at 18/21/31 (MSU: 12/11/39)
  • I'm not loving Iowa State in the Elite 8, but I hate them the least of the other options.  
  • The choices made in the Midwest bracket will ultimately make or break everyone's bracket who picks the right national champion. 


Good luck everyone!

2014 Bracket Picks: Elite 8 Deep Sleepers

This is to try and identify some potential Elite 8 sleeper teams based on historical profiles.

Going back to 2009, there have been five teams to reach the Elite 8 with an LRMC rank higher than 25: Marquette (2013), VCU, Butler and Florida (2011) and Tennessee (2010).  Of these five team, four of them had a better OE rank than DE rank, and the OE rank was less than 60.  Using that same criteria, here is a list of teams that match that profile…


Oddly enough the top three options are all in the bottom half of the West region, but many people are already talking about the Bears as a team to watch, so not sure it totally qualifies as a sleeper.

I haven't been hearing a whole lot about the Friars in the 11 seed, but they certainly have a respectable enough profile to warrant consideration in a weaker bottom half of the East.

While they won't be an Elite 8 pick for me, based on the numbers and top 10 stats, I really believe New Mexico State will be a handful for San Diego State in the first round.  I'll probably pick the upset just because this is the second time I've mentioned it in a post.

Sunday, March 16, 2014

2014 Bracket: Expert Picks (CBS, Yahoo!, SI, ESPN) (Part 4)

In short, the info below is basically like aggregating the knowledge of the top college basketball minds of ESPN, SI, Yahoo! and CBS. I tally the picks from their posted brackets. This has proven to be the most helpful for guidance in some of the 8/9, 7/10 and 5/12 match-ups.

Highlights:
  • Total Expert Brackets: 13
  • The last two years had Kentucky and Louisville as almost unanimous champions by the experts, but this year it's spread pretty evenly, with Florida having a slight edge.
  • First Round:
    • Upsets (Majority of picks for lower seeded team): Pittsburgh, OK State, North Dakota State, NCState, Iowa/Tennessee, and Arizona State
    • Syracuse, New Mexico, Michigan State, Iowa State, Creighton, Louisville and Duke each received all 13 picks.
  • Elite Eight Match-ups (Based on majority of picks): Florida/Kansas, Michigan State/Iowa State, Arizona/Creighton, and Louisville/Duke (Creighton gets tie-break because they had  a pick in the championship game)
  • Final Four: Florida/Michigan State and Arizona/Louisville 
  • Championship: Florida over Louisville


CBS Expert Picks: Here
Yahoo! Expert Picks: Here
CNN SI Expert Picks: Here
ESPN Expert Picks: Here

2014 Bracket Picks: Pomeroy + LRMC Sokol = Elite 8 (Part 3)

Per my post last year, there is a certain profile that Elite 8 teams have when looking at their composition of Ken Pomeroy's offensive and defensive efficiency (OE & DE), as well as the LRMC model.  

In the last two years I've posted on the blog, this model has helped me pick 10 of the 16 Elite 8 teams
, and I was 6 for 8 on the teams I identified as my most confident picks. 

Here is what an ideal candidate profile looks like:
  • LRMC ranking of 30 or less
  • OE and DE rankings of 40 or less
Each of the teams who fit that profile are highlighted in green:


Based on this info, and applying it to the brackets, here would be my reco for the safest Elite 8 picks. The ones that are bolded are the ones I have the most confidence in based on the other teams they'd potentially face in the list above.  :
  • South: Florida (1) and Syracuse (3)
  • West: Arizona (1) and Wisconsin (2)
  • East: Michigan State (4) and Villanova (2) - I put MSU over Virginia because of the better OE rank, as I believe MSU has a better chance to improve their DE than Virginia does in improving their OE during the tourney.
  • Midwest: Louisville (4) and Duke (3) - It will be a toss up between WSU and Louisville, so I'm giving Louisville the edge on LRMC.

Interesting Items of Note:
  • 4 of last 5 National Champions had a better OE rank than DE before the tournament. 
  • 4 of the last 5 NCs had at least an OE or DE of either 1 or 2: AZ, VCU, Duke and Creighton
  • Last year, Michigan had an LRMC of 9, OE of 2 and DE of 58. This year, the Badgers are 11/5/58.
  • Last year, Wichita State was a 9 seed with a 22/50/30 line. This year, the Zags are an 8 seed with 17/49/13.
  • Last year, Syracuse was a 4 seed with a 15/16/23 line. This year Michigan State is a 4 seed with 12/11/39.

Picking 2014 National Champion: Top 10 Stats - (Part 2)

Similar to the post I had last year, here's the quick rundown of how I narrow in on the right National Champion picks to consider:
  1. First, the champion has been a 1, 2, or 3 seed for 31 of the last 35 years (89%), so that narrows it down to 12 teams.  
  2. From there, the past 12 National Champions have finished in the top 10 in at least one of the stat categories listed below. I use ESPN's stats, and that's as far back they post stats. Overlap those two and you get the list of teams highlighted in green for who you should consider for your championship pick:
    • 1 Seeds: No top 10 stat categories 
    • 2 Seeds: Kansas: 1 of top 10 stat categories, Michigan (1)
    • 3 Seeds: Iowa St (2), Creighton (3)


This may be the year the strategy doesn't quite pan out, as I am hard pressed to put my faith into any of those four teams.  If Joel Embiid is actually healthier than the Jayhawks have been letting on, and he plays significantly, then I could buy on Kansas.  Additionally, like everyone else, I think Louisville was drastically underrated on a 4 seed, and they stand a solid chance of a repeat.

What I also like to use this list for is to help provide guidance on who not to pick. If you trust in the stats, then this rules out the likes of all 1 seeds, which could be very difficult to pick against given the number one overall seed has won the past two years in a row.

Other items of note:
  • Don't sleep on UCLA who has a solid all around game on paper, and came to play for the Pac 12 tourney.
  • With New Mexico State's high FG% and ability to get blocked shots, they could pose a serious threat to San DIego State in the first round.
  • A lot of people are already talking about it, but Kentucky's rebounding and shot blocking could very easily propel them into the Sweet 16 with an upset over the Shockers.