"My rule was I wouldn't recruit a kid if he had grass in front of his house.
That's not my world. My world was a cracked sidewalk." —Al McGuire

Tuesday, March 22, 2011

Bracket formula has me in ESPNs 97th percentile - says MU virtually even with UNC but no Final 4

Based on the formula I posted before the tournament and emailed to 7,000+, three of the four teams with the most NBA talent are in Newark with us (only Kansas is elsewhere). Of the Sweet 16, we also get three of the five Sweet 16 teams who had a bracket rating of more than 100 (Duke and Kansas the only two that are elsewhere).

However, due to UNCs narrow escapes throughout the ACC and NCAA tourney games so far, the formula now gives MU almost a 50-50 chance of stunning UNC - but almost no chance against the Ohio State-Kentucky winner Sunday. Here is what the formula says:




KentuckyMarquetteN. CarolinaOhio St.Notes
Sagarin90868996MU, OSU up 1
Hot/Cold22-23UNC drop off
Experience-20-22more than K,NC
Home adv.-13-11shift to BE country
NBA talent10088All outflank MU
Bracket rating999192110Elite 8 possible


In this post I want to cover; A) just how accurate the formula has been in picking winners this tournament, B) why the formula now shows UNC-MU as a toss-up, and C) the fact that if you throw out having to play a 3rd consecutive night in the Louisville game, MARQUETTE HAS BEEN A BETTER TEAM IN EVERY CONFERENCE AND NCAA GAME THAN UNC HAS BEEN IN ANY OF THEIR CONFERENCE OR TOURNAMENT GAMES. If that trend continues Friday, Marquette wins.

A. Formula has been EXTREMELY accurate – 97th percentile in ESPN

The formula rundown I posted and emailed to 7,000 people around the country has been right on. So far:

1. I usually try not to post self-serving items, but ... My bracket on ESPN (yes I only posted one) has me in the 97th percentile with Florida and the rest of my Final 4 intact.

2. I met a client between the OSU and MU game that said hundreds of people in their offices in Akron, OH have my formula picks taped up over their desks because the picks have been so accurate, and

3. Of the numerous people who have emailed me to tell me they are winning their pools based mainly on my picks, this was my favorite: “Pud, i'm in 4 pools, and in first place in all 4 of them. I haven't watched a college basketball game since Maryland won the finals back in '03. And before that, it was maybe 10 years. I based most of my picks off of your ridiculous emails! Thank you!” (My note, while this friend has great season tickets to the Washington Wizards, I’d say he really doesn’t follow college basketball in light of the fact that he didn’t know his alma mater actually won the title in 2002).

No one including me followed the formula exactly. For example, many subbed Michigan in when the AD basically announced they were firing Pearl. As outlined in my post, I picked Marquette to go Sweet 16 despite the formula having them losing narrowly in the first two match-ups, and I also picked Richmond for the Sweet 16 in my ESPN entry because the formula showed them beating Vandy and virtually even with Louisville, and I just didn’t believe Louisville could beat the Spiders.

B. Formula calculates UNCs edge of MU has slipped from 101-87 to just 92-91.

The bracket I did before the tournament has UNC beating Marquette Friday, with the only slight edge for Marquette being experience. The formula showed UNC as the better team (Sagarin 89-85), the hotter team (+4 to 0) and giving a huge edge in NBA talent (+8 to 0).

The NBA talent is still the factor that could let UNC run away with the game, as they get points here for the www.nbadraft.net projections of with 5th pick Harrison Barnes (6-foot-8); 60th, Kendall Marshall, 6-3; 66th, John Henson, 6-10; 73th, Tyler Zeller, 6-11. Basically, four of our guys will be matching up with someone who should be in the NBA within two years, and that could lead to another Roy Williams blowout of MU.

But what has improved is the venue, as UNC moves from five tournament games in their home state to Big East country in New Jersey, and the fact that UNC is no longer the hotter team coming in.

C. Postseason – MU 10 points a game better than UNC this post season

As for the Louisville game, I believe MU has proved that its reliance on speed and quickness to overcome other weaknesses makes it a poor team when playing a third consecutive day. The only three times this has happened in the last two years were the blown 18-point lead to FSU, and the blowout losses to Georgetown and Louisville in the 3rd game of the last two Big East tournaments.

However, if you throw out that one Louisville game, Marquette has played a better game in EVERY post season game this year than UNC has played in ANY post season game. Here are the game-by-game Sagarin ratings:




MU opponentMUUNCUNC opponent
Providence9982Miami
West Virginia9388Clemson
Louisville6378Duke
Xavier9488Long Island
Syracuse9390Washington
Post season (ex. Lville)9585Post season


Amazingly, that means if these were the only games of the season, Sagarin would calculate Marquette as 10 points better than UNC. In fact, even if we include getting blown out by Louisville back into the averages, Sagarin would calculate Marquette as a 3-point favorite over UNC (88-85 rating) based just on the five post season games for each.

As you will note, the formula shows the Kentucky-Ohio State winner blowing out whoever wins the Marquette-UNC game. As I have covered over the past could of years, it almost always takes three players who will be in the NBA within the next two years to make the Final Four, and Marquette is the only team in Newark this weekend who doesn't have them.

However, the Elite 8 is not impossible.

No comments: