Click for bigger  image

NFL Draft – CFB Development vs. Recruitment

By Dave Bartoo – The College Football Matrix @CFBMatrix
Annual Update April 15, 2013

I started adding to the profile of the modern recruiting rank era (2002 forward) and draft pick rankings since then in 2005 when the first group of recruits was eligible.  While the draft does relate back to the college football predictions it may tie together other areas of the Matrix rankings.

For those of you reading the Matrix for the first time, you will find that everything is tied together or related in some way.  It’s just one big rabbit hole with simple and understandable stats that simple help paint a picture and put you on the right side of any ‘odds’ for expected results.

For the NFL draft I am curious as to how many draft picks correlate to CFB teams recruiting rank and how many draftees come out of schools that I have identified having net negative game effect coaching.  Simply put, how good are the overall recruiting ranks and can bad coaching bring talent, even NFL level talent down.

No Surprises – Of the top 10 teams for recruiting ranking over the last decade, 9 of them are in the top 10 for total draft picks.  Like it or not, the recruiting rankings done by Scout, Rivals and ESPN are very solid even 4 year ahead of a draft.  Only non-top 10 recruiter in top 10 of total draft picks – Cal

Great Player Developers – The top teams with guys drafted totals farthest above their average recruiting ranks:

1. Cincinnati
2. Utah
3. Wisconsin
4. TCU
5. Boise
6. Iowa
7. Cal
8. Stanford

• Only Wisconsin, Iowa and Virginia have recruiting rankings above #30 from ’02-’09 and have more than 25 draft picks in the last 7 years.

Worst Top 10 Recruiter – So one team that recruits in the top 10 the last decade is not a top 10 developer of draft picks.  Bad coaches, or bad culture?  Probably both in Ann Arbor, Michigan

Weak Player Development - So we know that recruiting rank is a very good barometer of future NFL draftees.  We have posted the teams with the biggest gap between recruiting rank and draft numbers rank.  How about the other direction?  These are the teams with above average recruiting and way below average number of draft picks over the last 8 years.

1. Washington
2. UCLA
3. Ole Miss
4. Mississippi State
5. Texas A&M
6. Washington State
7. Texas Tech
8. Oklahoma State

•  Teams with less than 20 draftees in the last 7 years with recruiting better than #30? Texas A&M, Washington, Arizona State, UCLA, Ole Miss and Maryland

2012 NFL Draft Notes

• Of the 192 total picks in the 2012 draft, 166 came from teams with winning records in 2011. (86.5%)
• 19 teams had losing records in 2011 and net NEGATIVE game effect coaches (Matrix rankings).  Those teams had 19 picks or 1.05 picks per team
• 7 teams had losing records but positive game effect coaches for 1.57 picks per team
• 24 teams had winning records in ’11 but net negative game effect coaches.  77 draft picks for a 3.20 average
• 23 teams had winning records in ’11 and POSITIVE game effect coaching in the last 4 years.  84 picks of 3.65 picks per team

CFBMatrix conclusion: Recruiting produces winners and winners produce draft picks.  The wrong program and wrong coach may lessen your odds of being picked in the NFL Draft.  More research on past years is required but the results are interesting enough to give me pause to consider it.

Contact: Dave Bartoo @CFBMatrix on Twitter  Email Dave@CFBMatrix.com 971.244.3041

Current NFL Draft Pick to Recruiting Rank Ratio Matrix

Wonderlic, Shmunderlic

Let’s just put this thing to rest.  Wonderlic scores are about as useful as Roscoe giving Flash an IQ test.  But here we are every year talking Wonderlic tests.  And every year we need to point out Jim Kelly, Marino, Cunningham, McNair, Bradshaw and Vinny all posting worse scores than David Duval playing o past champion exceptions in Majors.

First, there is no proof if the numbers in the public are even right so keep that in mind.  Logically, when you look at all the available scores, it kind of makes sense.  However, the range is so wide, lets go through segments of the scores and averages so you can draw your own conclusions.

• Average score of a Hall of Fame or potential HOF QB*: 24.8
• Ave. score of current/past Pro Bowl QBs*: 25.4
• Ave score of top 5 draft picks busts*: 30.2

No enough?  Let’s keep rolling.  And BTW for reference, click this link to my Boom, Bust, Pending Spreadsheet.  It is incomplete as we did not label every guy except the HOF, All-pro and busts that were top 5 picks.
—————————————————————————————————————-

If you want a meaningful breakdown of boom/bust trends then
click here for If you can’t lead a college team, you can’t lead a pro team
or click here for Want to be a successful NFL QB? Don’t go to an Elite Program.

—————————————————————————————————————-

• All QBs:  Successful ave score: 29.2   Busts ave score: 26.4
• AQ QBs:  Successful ave score: 29.9  Busts ave. score: 25.2
• Non-AQ QBs: Successful ave score: 27.75   Busts ave score: 29.1  Ha! Not only is it close, it’s bassackwards fro what we would assume
• FCS QBs: Successful: 28.7  Busts: 27

When it come right down to it, the average successful QB drafted into the league has a Wonderlic score just slightly ahead of the Busts (except for the high bust rate non-AQs).  Simply put and in my opinion, Wonderlic scores are about as valuable as pre-season top 25 rankings.   In fact, they’re worse because they provide no direction or correlation to success or lack thereof.

Questions , concerns, comments?  I am the man behind the curtain.  Tweet me @CFBMatrix, call 971.244.3041, email dave @cfbmatrix.com  – Dave Bartoo

*I went through the list of QB with scores and pulled out guys that are in the HOF or on the voting list.  For Pro-bowls, they had to just be in one to qualify on my list.  Busts as defined on the chart put together by on-line draftniks and friends Russell Baxter and Mike’l Severe.

Want to be an NFL QB? Don’t go to an ‘Elite’ program

Dave Bartoo – The College Football Matrix
April 12, 2013

As I showed in the previous article, QBs drafted since 2004 from under performing teams are twice as likely to Bust in the NFL than guys from over performing teams.  In writing the article, I noticed there appeared to be a pattern more than their fair share of ‘Busts’ coming from the elite schools.

Since the top 16 talent adjusted recruiters since 2002 have all won and lost the last 8 National Championship games, I felt that was a place to start to see if it is better to go to an elite program for your NFL QB career or not.

• Of the teams that have been a top 16 recruiters since 2002, 18 QBs from those teams have been drafted into the NFL.  Only 4 were not labeled ‘busts’ for a failure rate of 78%.  When you look at the QBs drafted from this group AND were coming of a negative coach effect season then the numbers drop to a 11% success rate or just 1 of the last 9 guys taken.  Barkley and Manuel when drafted, with be included in this ultra high-risk group.

• On the flip side, the QBs from AQ teams outside the top 16 in elite recruiting tell a much different story.  Their success rate in the NFL is nearly double that of the elite top 16 teams at 41%.  Even more impressive is the numbers for QBs with positive coach effect as their failure rate is just 53% vs. elite team QBs at 78%.  Even the non-elite QBs from teams with a negative or zero coach effect busted less often than QBs drafted from top 16 recruiters at 73.4%, the same rate as QBs from the top 16 teams with positive coach effect.

CLICK READ & LISTEN: Dave Bartoo on The DA Show at CBS Sports Radio Talking Drafting QBs

After 8 years of draft information, I feel it is a very clear trend that QBs from these elite teams have a much higher bust rate than their ‘non-elite’ peers.  Given that a bad draft and bad choice of a quarterback can set a franchise back many years, reducing risk must be paramount in picking QB.

In my amateur opinion, QBs from these ‘elite’ teams, simply do not have the game to game pressure and learning curve to prepare for the NFL.   When you are on an ‘elite’ team, you, as a QB, are rarely in a situation where you are in a tight game, under pressure, forced to lead and carry a team every weekend.  In the NFL, that’s what you need to do every day.

Sure it is hard to judge an elite guy from and elite program but they need to at least live up to team talent expectations (Bradford).  Think about it, how many big time pressure situations (like you see in the NFL every week) happen to elite team QBs? Once, twice, in any given year?  Clausen, Quinn, Young, Leinert, Russell, Sanchez….. how many games did those guys play in to prepare them for the NFL’s most demanding position?

 2013 QB Draft Class to AVOID with 89% Bust rate: Matt Barkley, EJ Manuel and Landry Jones

Note to HS QB recruits: This tells you to win go to an elite program, to get to the NFL and be successful there get experience and prove yourself a leader and winner somewhere else.

Questions , concerns, comments?  I am the man behind the curtain.  Tweet me @CFBMatrix, call 971.244.3041, email dave @cfbmatrix.com  – Dave Bartoo

FWIW: 66%  This is the average ‘bust’ rate of QB drafted into the NFL since 2004 as defined by the spreadsheet (disagree with the spreadsheet then comment below, tweet me @CFBMatrix or email dave@cfbmatrix.com).

If you can’t lead a CFB Team, can you lead in the NFL?

Dave Bartoo – The College Football Matrix
April12, 2013

I know the NFL loves winners.  Look at the draft.  78% of all guys taken in 2012 were from college football teams with winning records.  But, as the CFBMatrix Coach Effect metric shows, it takes more than being a winner to not be a bust as a QB drafted into the NFL.

Here at the CFBMatrix, I have used the term ‘Coach Effect’ to describe a team’s seasonal performance above (positive Coach Effect) or below (negative Coach Effect) in relations to it’s talent and schedule*  However, having watched enough NFL games and hearing the Jaworski’s and Gruden’s beat it into my head that it is a QB driven league, the thought was simple: “If a QB can’t lead their college team with 18-22 year olds to wins, how in the hell are they supposed to lead a pro team.”

Click to READ & LISTEN: Dave Bartoo on The DA Show at CBS Sports Radio Talking Drafting QBs

If great QBs create wins, maybe my Coach Effect, especially when the coach has a history of under performing the talent and schedule, is dependent on good or bad quarterbacking.  Let’s look at the 2012 draft.

Take a look at the far right column.  Lots of positive Coach Effect numbers for these QBs teams in 2011 before they were drafted.

While most of these guys are still ‘pending’ to be a Boom or Bust, we can agree, the best QBs from this class were all + Coach Effect guys.

The next question should be, “What about past years?  Is it similar?  Is there a trend or pattern to +/- Coach Effect? (FCS vs non-AQ QBs in next article)

We all know that the odds are against these guys the second they are drafted and many bust in the their first couple of years.  However, if I am a GM, I want to minimize risk in the draft when ever possible. (click this link to my Boom, Bust, Pending Spreadsheet).

Since 2004, including the guys above from 2012, there have been 58 total QBs drafted from with a +/- Coach Effect number that were labeled Boom or Bust (I don’t track FCS).  Here is a quick breakdown of the results.

• QBs drafted with a + Coach Effect for their team the year before drafted:
• Total Positive + Coach Effect: 13 of 33 or a ‘bust’ rate of 60%
  - AQ QBs 9 of 23 ‘Boom” for a ‘Bust’ Rate of 60.5%
  - Non – AQ QBs 4 of 10 ‘Boom’ for a Bust rate of 40%

• Negative Coach Effect: 5 of 25 or a Bust rate of 80%
  - AQ QBs 4 of 19 successful for a 79% Bust rate
  - Non-AQ QBs 1 of 6 for a Bust rate of 83%

My take on the 2013 QB class with the trends since 2004 and the metrics of the 2013 QBs:  
Take a QB out of college from a team that underperformed it’s talent and your odds are 1 in 5 that they stick in the league. QBs that out performed the expected schedule and talent of their team have been twice as likely to be successful in the NFL.

Lowest Bust Risks:
Ryan Nassib
Mike Glennon
Matt Scott
Sean Renfree
Jordan Rodgers

Guys to avoid:
Geno Smith – 78% bust rate
Matt Barkley – 92% bust rate
EJ Manuel – 92% bust rate
Tyler Bray
Tyler Wilson
Landry Jones
Zac Dysert (plus non-AQ failure rate if drafted after Rd 3 is over 90%)
Ryan Griffin (see Dysert)

FWIW: 66%  This is the average ‘bust’ rate of QB drafted into the NFL since 2004 as defined by the spreadsheet (disagree with the spreadsheet comment below, tweet me @CFBMatrix or email dave@cfbmatrix.com).

Questions , concerns, comments?  I am the man behind the curtain.  Tweet me @CFBMatrix, call 971.244.3041, email dave @cfbmatrix.com  – Dave Bartoo

*78% of all AQ football games win/loss outcomes predicted correctly 6 months before the last 5 seasons using only CFBMatrix adjusted talent ranking and team home/away strength.

Drafting NFL QBs – Coaching vs QB Effect

By Dave Bartoo – The CFBMatrix
with contributions from Russell Baxter and Mike’l Severe

While my work here on the CFBMatrix has a focus on college football, I take liberties in reviewing ‘links’ from CFB into the NFL.   I love odds, trends and stats and when all three come together I am a very happy camper.

My inspiration for this article and subsequent research started when I saw the NFL ROY announcement a while back when the names Griffin, Luck and Wilson highlighted my screen.

I didn’t care who won, but what I saw was three guys all being drafted after their college teams (Baylor, Stanford and Wisconsin) had big ‘Coach Effect’ seasons in the CFBMatrix and lead their teams to arguably the best seasons in their programs recent history.

After watching enough NFL games and hearing the Jaworski’s and Gruden’s beat it into my head that it is a QB driven league, the thought was simple to start the article: “If a QB can’t lead their college team to wins how in the hell are they supposed to lead a pro team.”

If you love stats, and I know you do if you are reading and following on twitter, and can take an unemotional and objective approach to the numbers, the rest of this is going to be a fun read.

The first part of the review is the most subjective.  I asked my friends Russell Baxter of www.profootballguru.com and Mike’l Severe co-host of Unsportsmanlike Conduct and self-proclaimed draftnik on AM1620 in Omaha to rate each QB drafted since 2004 “boom’, ‘Bust’ or ‘Pending’.  It’s inexact, but we can all agree on most of their responses (click this link to Boom, Bust, Pending Spreadsheet).

Here is the first number to keep in mind: 66%  This is the average ‘bust’ rate of QB drafted into the NFL since 2004 as defined by the spreadsheet (disagree with the spreadsheet comment below, tweet me @CFBMatrix or email dave@cfbmatrix.com).

We all know that the odds are against these guys the second they are drafted and many bust in the their first couple of years.  However, if I am a GM, I want to minimize risk in the draft when ever possible.  This, in my opinion, is clear in the drafting trends of players from losing teams.  In the 2012 for every player picked from a team with a losing record in 2011, there were 3.56 players taking from teams above .500.  The NFL is about winning, and they look for winners and no more is that important than at the QB position.

First Stat Profile: QBs with + Coach Effect vs – Coach Effect
58 total QBs drafted from with a coach effect number (I don’t track FCS).

• Positive + Coach Effect: 13 of 33 or a ‘bust’ rate of 60%
• Negative Coach Effect: 5 of 25 or a bust rate of 80%
• Summary:  Take a QB out of college from a team that underperformed talent and your odds are 1 in 5 that they stick in the league.  Guys to avoid: Smith, Barkley, Manuel, Bray, Wilson and Jones.
• Rule #1: Hedge on drafting a QB from a team that underperformed

Second Stat Profile: Non-AQ vs FCS

• Since 2004 there have been 8 QBs drafted out of the FCS Division II.  Of those 8, only 3 have been considered ‘busts’.  When the fail rate is 66% and you are succeeding in draft successfully over half the time, while the sample size is small, it’s calling to me.
• Of the 17 non-AQ QBs only 5 were label ‘Boom’ QBs.  12 busts or a 71% bust rate.  To go deeper in to the non-AQs look at the Coach Effect rates.  For guys drafted from non-AQ programs, those with CE over +2 busted at a 60% rate, those under +2 83% bust rate.
• Of the FBS QBs taken in round 2-5 with a negative coach effect, 14 of 15 have busted.
• Summary:  Only consider non-AQs late in the draft and they better be on a team that was +2 or better coach effect the previous year.
• Rule #2: Look more to draft a QB late from FCS before a non-AQ program.  Any non-AQ QB on a negative coach effect team should be passed

Third Stat Profile:QBs from Elite vs non-Elite Recruiters

Since the top 16 talent adjusted recruiters have all won and lost the last 8 National Championship games, I felt that was a place to start to see if it is better to go to an elite program for your NFL QB career or not.   Hint: If you think an elite program is ‘too good’ and doesn’t place  QB under enough weekly pressure to win and lead then I think you are right.

• Of the teams that have been a top 16 recruiter since 2002, 18 QBs from those teams have been drafted into the NFL.  Only 4 were not labeled ‘busts’ for a failure rate at 78%.  When you look at the QBs drafted from this group AND were coming of a negative coach effect season then the numbers drop to a 11% success rate or just 1 of the last 9 guys taken.  Barkley and Manuel when drafted with be in this ultra high-risk group.
• On the flip side, the QBs from AQ teams outside the top 16 in elite recruiting tell a much different story.  Their success rate in the NFL is double that of the elite teams at 41%.  Even more impressive is the numbers for QBs with positive coach effect as their failure rate is 53% vs. elite team QBs at 78%.  Even the non-elite QBs from teams with a negative or zero coach effect busted less often than QBs drafted from top 16 recruiters at 73.4%, the same rate as QBs from the top 16 teams with + coach effect.
• Summary: QBs from ‘elite’ teams (top 16 recruiters), IMO, simply do not have the game to game pressure and learning curve to prepare for the NFL.  Sure it is hard to judge an elite guy from and elite program but they need to at least live up to team talent expectations (Bradford).  Think about it, how many big time pressure situations (like you see in the NFL every week) happen to elite team QBs? Once, twice, four times in 1-2 years?
• Rule #3: Look to draft the guy that carried a team to wins from a non-elite program.  Avoid the QBs from under performing elite talented programs.
• Note to recruits: This tells you to win go to an elite program, to get to the NFL and be successful there get experience and prove yourself a leader and winner.

Drafting QB in the 2013 NFL Draft with the Matrix

So what about the 2013 group of potential draftees at the QB position.  Well, if you got this far, you should know that this crop, by the numbers, is not very good.

Let’s just go in order of the chart on the right that lists the QBs in order of ‘perceived’ draft ranking by QBs.  If you read the article complete to this point, I’ll be just spitting out what you already know.

Geno Smith – PASS:
Not only is this guy coming off leading his team to a -1 coach effect, he did it for back to back years.  Can’t carry a Mountaineer team to wins, how you going to do it in the NFL.  Since 2004, his profile has a 80% bust rate.  Guys with similar numbers in recent drafts:  Erik Ainge, Brady Quinn, DJ Shockley, Jake Locker

Matt Barkley – PASS:  I really like this guy for standing tall for his backyard school and now alma mater.  Could have folded under those sanctions and went somewhere else.  Probably should have.  Bust rate for his profile is almost 90%.  He is a wildcard in my opinion because the winning problems may have been less him and more Kiffin and staff.  However, he also had surreal line and WR talent and still struggled to control games and win.  Great character in the media, but that doesn’t win in the NFL.  Drafted QBs with same metrics: Tannehill, Clausen, McGee and Henne

Ryan Nassib – Pick: Coming off a +4 season and a non-elite program is a green light for me.  This profile is an NFL success over 50% of the time since 2004.  He mirrors the matrix numbers of Orton, Cutler and Wilson (Kafka and D. Anderson too).  If I have to go all-in on ONE QB this draft to boom/bust it is this guy.

EJ Manuel – PASS:  This guy has been the leader of the Noles for how many years and every one has been a disappointment.  I feel FSU gets better, or at least has a better chance of improving when he leaves.  Another guy from an elite school with a negative coach effect number of -2.  The worst odds for him, like Barkley, around 11% success rate.  QBs drafted since 2004 with metric like EJ: Stafford, Yates, Compton, Greene and McPherson

Tyler Wilson:  This one is very interesting.  First, he should have come out after 2011.  Huge season and wasn’t getting any better.  My quandry is that is the -2 coach effect for the Hogs in 2012 more about the transition than his ability and leadership?  Stick to the code. PASS – But a real wildcard to make it

Mike Glennon: This guy is a real tweener in my opinion.  Sure he was coming off a +1 team, but he was also leading a -1 team in 2011.  I know, Glennon fans would say he was just learning and maybe so. Still a better risk after round 2 or 3 than most of the guys picked before him.  PICK

Tyler Bray: Maybe the third time is the charm but since Peyton, Erik and Jonathon have been busts.  Tyler was on a team that under performed every year for the last 3.  PASS

Matt Scott: QB’d a +4 team in 2012.  Wow.  Was it Matt or Rich?  Probably both but Arizona had no talent outside of Matt and Kareem.  Round 2-3 where he is slotted and based on the numbers is a solid value.  No question early round guy.  PICK

Landry Jones: Like EJ and Barkley, this is another elite school QB on a team with negative coaching effect in 2012.  He was on a -1 coach effect team in 2012 and -1 in 2011.  Numbers say he doesn’t have it.  PASS

Zac Dysert: -2 coach effect and at a non-AQ school.  Draft him and you’re flush money.  PASS

Sean Renfree: On a +3 team and led them to wins they have never seen before.  How you don’t grab this guy late if he is available and you have a spot to fill is beyond the numbers.  PICK

Jeff Tuel:  Hurt all the time and did nothing to help win under two different coaches and a ton of opportunities to elevate his team and win football games.  PASS

Ryan Griffin: -4 coach effect and at a non-AQ school.  See Dysert above.  PASS

Brad Sorensen:  All I know if he is ranked in the top 15 and a FCS QB.  Number say take him way before the non-AQ guys.  If not a late round pick great guy to sign to camp.  Late Round PICK

Jordan Rodgers: +5 team and +5 last year too?  The ultimate diamond in the rough.  Huge + coach effect on his team and played in the toughest conference.  Challenged to win every week and tons of pressure handled.  Don’t draft this guy, you are missing out on someone with excellent odds of being in NFL for more that just one year.  PICK

Questions , concerns, comments?  I am the man behind the curtain.  Tweet me @CFBMatrix, call 971.244.3041, email dave @cfbmatrix.com  – Dave Bartoo

2013 NCAA March Madness Matrix

I couldn’t resist.  For years I have been asked “Do you do a Matrix for college hoops?”  No.  I don’t have the time or the interest.  But I have to admit filling out a bracket every year.  I, like so many, just throw some darts and have fun with the tourney.

No more.  I set aside 2 days to see what I could find to correlate teams and winning tournament games.   Like the CFBMatrix, I just want simple answers to try and play the best trends and odds to have a better bracket.  It’s never going to be perfect, but my 5th stab at the numbers yielded 13 of 16 team, 6 of 8, 3 of 4 and the national title winner and loser.

So the 2013 NCAA tournament bracket a la The College Football Matrix rules are as follows:

Rule #1:  In the last 2 years there has been a group of teams that shared a common number and never made the Sweet 16.  So to start my bracket I am making sure these teams are not in my Sweet 16:  Boise, Butler, Cal, Colorado, Fla Gulf, Harvard, Iowa State, JMU, Liberty, LIU, Minnesota, New Mexico State, NC A&T, Notre Dame, Oklahoma, Oregon, Pacific, UCLA, W. Kentucky and Wisconsin.

I know…. some of these are like duhhh obvious, but it’s the little things that count.  Take out New Mexico State and Oregon and lock in Oklahoma State.  It also locks in Ohio State to the sweet 16 for me as well as Florida.  It makes Ole Miss a great upset pick and possible New Mexico a sweet 16 favorite too.

Rule #2: Take all teams in Matrix with 3-0 edge in critical metrics   (exceptions are teams with 10%+ gap in RPI/BPI)

Sub-Round 1 picks:

St Mary’s 3  Midd Tenn 0 – Pick is St. Mary’s. They dominate all 3 critical areas of winning teams in the tourney the last 2 years.  MTSU doesn’t have the RPI/BPI rank to reconsider.

Boise State 2 LaSalle 1 – Pick is BSU. I see a very close game.  No shock if LaSalle wins.

NC A&T 2 Liberty 1 – Pick is NC A&T – Bad NC offense versus worse Liberty defense.   Both of these teams have bad numbers so it could get ugly.  Neither wins round 1.

LIU 3 James Madison 0 – Pick is LIU - Like St. Mary’s above, LIU is my big favorite to win this game.

Rule #3 – Lock in the teams with the best metrics into your Final 4, Elite 8.
You either get wild, random teams no one but their own fans had or the teams with the best talent, trends, and numbers entering the tourney.  Look at 2012.  3 of the final four teams were all superior in talent, scoring and power rankings.  UK and KU were #1 and #2.  Why screw around getting to the end.  Take the best and put the odds in your favor to start.

Midwest:  Louisville is an easy #1.  Others in the mix for the Elite 8 from the lower bracket will be Duke, Creighton and Memphis.  Duke is the favorite in the metrics but they are going to have to play very well to get by both Creighton and Memphis.

South: This is the Florida and Kansas show.  I’ll get to the Elite pick between these two later

West:  This one really is driving me crazy!!  The numbers put the two best teams as Gonzaga and Pitt.  Ohio State has an EASY bracket to the elite 8.

East: Wow!  The numbers for these teams when compared to the other 3 brackets suck!  I mean bad, bad, bad.  The Hoosiers have a freakin’ cake walk to the final 4.  No one else is even close.  Any loss they suffer in that bracket is a huge upset.  Miami is clearly the talent in the other half.

So to get you to where my bracket is at after rule #3. Indiana wins the East over Miami.  Florida beats Kansas in the South.  Louisville wins the Midwest but plays Duke or Memphis in the final of that bracket.  And lastly Ohio State plays and loses to Wich State/Pitt winner.  Numbers say the Zags but I don’t buy 3 #1s in a final 4 and Gonzaga has had no tests in the last 90 days.

In the final Indiana and Louisville have the best combination of defense, offense, toughness of schedule and total overall rankings to be the picks out of the final four.  By the numbers, it is Louisville’s to lose.

Rule #4 – Teams with weak offenses can’t make runs

The numbers say stay away from high seeds that have low offense efficiency.  These are all good teams on defense but if you don’t put it into the peach basket, winning 3,4,5 in a row is much harder.  I am avoiding:  Cincinnati, Georgetown, Marquette, Michigan State, Missouri, New Mexico, Saint Louis, San Diego State, UNLV.

Rule #5 – Team with weak defense can’t make runs either

Like teams with weak defenses above, teams that allow scoring don’t make it deep in the tourney either.  Rare is the day when a team allows over a point per position in the regular season makes a Sweet 16.  The odds are strongly against some high seeds advancing: Kansas State, Michigan, and Syracuse.  Lower seeds with no D include Butler, Iowa State/Notre Dame, NC State, UNC and Oklahoma.

Round 1 Picks

You want to win a bracket pool?  Sure its great to nail the elite 8 teams and final 4s but round 1 & 2 wins add up.  Let’s try to make your bracket green.

East:
Teams having 100% of critical metrics in their favor:
Indiana over LIU
Miami over Pacific

Teams with over 70% of metrics in their favor and most critical metrics
NC State over Temple
Syracuse over Montana
UNLV over Cal

Teams with over 66% of metrics in their favor but missing at least one critical metric

Butler over Bucknell – Dang this one! Bucknell flat scores the ball and Butler is weak on both ends of the floor.  Numbers say Butler by a hair.  However, like the game below, Rule #1 says don’t take Butler past round 1.  Nothing wrong with two brackets.  One with Bucknell and one with Butler.
Colorado over Illinois – Coin toss game.  Numbers say the Buffs, but Rule #1 also says they don’t make it to round 2.
Marquette over Davidson – I still like Marquette in this one.  The talent is too much for Davidson but…what scares me is Marquette scoring the ball.  They are very inefficient and Davidson is a solid scoring team.  Davidson needs to milk the clock and keep it low scoring.  If they control tempo, they can win.

Midwest

Teams having 100% of critical metrics in their favor:
Duke over Albany
Louisville over NC A&T
Ok State Over Oregon
Saint Louis over New Mexico St

Teams with over 70% of metrics in their favor and most critical metrics
Memphis over St. Mary’s – SMC is sneaky good.  Look ahead and Memphis will get upset.  Good switch for two brackets in play
Creighton over Cincinnati

Teams with over 66% of metrics in their favor but missing at least one critical metric
Colorado St over Missouri – I don’ like this pick but every good metric points to CSU.
Michigan State over Valpo – Upset Alert folks.  State can’t score and doesn’t defend well either.  This one could be very close.  Watch the tempo.  Slow for Valpo.

South

Teams having 100% of critical metrics in their favor:
Florida over NW State
Georgetown over Fla Gulf
Kansas Over W. Kentucky
Michigan  over South Dakota State
VCU over Akron – This is the only one that looks to be a close one of these 5 teams.

Teams with over 66% of metrics in their favor but missing at least one critical metric
North Carolina  over Villanova – Coin toss game.  Low scoring = ‘Nova  High = UNC
San Diego State over Oklahoma – SDSU can’t score but SDSU plays good D.  They give up more than 65 and OU will win.
UCLA over Minnesota – Nearly identical teams but UCLA does have a starter out (not included in this draft of the tourney Matrix).  OK to go Gophers. Split in two brackets if playing multiple?

West

Teams having 100% of critical metrics in their favor:
Arizona over Belmont
Gonzaga over Southern
New Mexico Over Harvard
Ohio State over Iona

Teams with over 70% of metrics in their favor but missing at least one critical metric
K State over La Salle
Notre Dame over Iowa State – Both teams have ugly numbers in efficiency.  Neither will be a challenge to Ohio State in round 2.  This could get ugly
Wichita State over Pitt – Pitt has some of the best numbers in the bracket with only WSU and Gonzaga having better metrics.  Wrong seed for both teams.  Winners goes to elite 8.
Wisconsin over Ole Miss – Wisconsin has the metrics of a team that doesn’t get to the 2nd round.  Upset here or next round take your pick.

Round 2 Picks

EAST (i.e. cakewalk bracket)

Marquette over Butler – This is a weird game and IMO bad seeding.  Butler has the fatal flaw of a negative efficiency value and Marquette has a offense efficiency value under 1.  Both those numbers are rarely part of a Sweet 16 team.  but someone has to win and Marquette has a 3-0 winner in critical Matrix metrics.

Miami over Illinois – Numbers a lock for this game no matter who Miami plays in round 2.  Canes and Hoosiers in the cakewalk bracket.

Indiana over NC State – Assuming that NC State doesn’t blow it in round 1 because they cannot defend, Hoosiers should still smoke the Pack.

Syracuse over UNLV – If you follow rule #4, you don;t have UNLV past anyone in round 2.  They don’t have the offense to support a run to the Sweet 16.

WEST

Ohio State over Notre Dame – Doesn’t matter if OSU gets ND or ISU both get whacked.  Buckeyes have one of the easiest paths to the Elite 8.  ND and ISU both have metrics that scream no Sweet 16.

Arizona over New Mexico – Even though New Mexico has a 2-1 winner in critical metrics over Arizona, don’t take them.  They have sub 1.0 rating on offense which it makes is statistically unlikely they see the Sweet 16.

Gonzaga over Wichita State – I really don’t trust any ‘non-AQ’ team making a big run but the numbers are in favor of Gonzaga.  That’s the pick, but their lack of SOS challenges and athletes will catch up to them some time.

Kansas State over Wisconsin – Wisconsin has the fatal number of total negative efficiency AND sub 1.0 on offense rating.   Both spell death by round 2.  Badgers a good bet for and round 1 upset too.  KState also sub 1.0 on offense but less flawed than Wisconsin.

SOUTH

Florida over UCLA – UCLA/Minnesota game battle of two bad teams.  Gators best numbers in the South.  FF team all the way through

Georgetown over San Diego State – Aztecs can’t score it but neither can the Hoyas.  Better SOS/RPI/BPI to G’Town.  But the Hoyas have the metrics of a team that can’t make a deep run.  Gone in Round 3.

Michigan over VCU – Coin toss. Michigan has the defensive numbers of a team that doesn’t make a Sweet 16.  I really should be reading the numbers for VCU.

Kansas over UNC – UNC has a bad defense and the Jayhawks metrics for the season are top 5.  KU rolls easily over UNC or Villanova

MIDWEST

Louisville over Colorado St – U of L has all the numbers in their favor.  #1 average of the critical 3 metrics for winning the tourney.  They win out.

Memphis over Michigan State – MSU has 2 of 3 critical numbers in their favor but they lack a +1.0 offense rank.  With the Tiger being so balanced and the <1.0 rank of MSU time to play the odds and call the upset.

Oklahoma State over Saint Louis – Just like above, Saint Louis has a 2-1 edge in critical numbers but their offense ranking says odds very low to make a Sweet 16.  Post tourney start note: SLU/Oregon match up means a team with a metric that hasn’t made a Sweet 16 in 3 years is going to happen.

Duke over Creighton – Hell of a match up.  Duke has a tough bracket line.  Beating Creighton, then MSU/Memphis winner before elite 8 match up.  IMO not going to survive that gauntlet.

Round 3

See Rule #3 above in picking the best overall teams with the big 3 metric first.  But here is a breakdown of the Elite 8 teams and games

Florida over Kansas – IMO the strongest of the 4 games.   Expecting a great match up but the edge in efficiency goes to the Gators.  Squeaker.

Indiana over Miami – To this point neither team should have been challenged in the Cakewalk bracket.  Poor seeding in my opinion here.  Hoosiers now get the easiest Elite 8 game of the four.

Louisville over Duke – As noted above, I feel Duke has a great shot of not being in this game.  If they aren’t here it will be Creighton.  Don’t matter, both will get spanked by the Cardinals.

Ohio State over Gonzaga – The numbers say don’t love ‘non-AQ’ conferences.  The Buckeyes have one of the easiest bracket paths to this game and have played a much tougher schedule than the Zags or Wichita State (winner of that one makes it here).  Zags have the big metrics but playing the odds that they are over rated due to their schedule.

Finals

Louisville beats the Gators and Hoosiers beat the Buckeyes.

Louisville with the best efficiency and metrics combo in the country wins it all against Indiana barring unforeseen injury.

Welcome to the 2013 kickoff of College Football Season for the CFBMatrix.  It’s been a great off season for me with visits to several AQ schools to talk stats and creating better or new pro-active analytics in their football and athletic programs.    These adventures has spawned a ton of new ideas and metrics for tracking and predicting college football.  I see the CFBMatrix as a ‘Art of War’ disciple where it’s over before it’s over.

Let start by thanking the over 5 million pages read last season on this site.  All the followers on twitter @CFBMatrix and everyone that starts a conversation, gives an idea or asks a question.    In addition to adding more unique stat angles and predictions, I am going to try out video articles to get a better understanding of where I am coming from and my thinking.  Plus, I am hoping it cuts down on all my damn typing I have to do to make this bird fly.

I love this article as it was the first one I ever wrote for a CFB season and it what started my path done down college football stats.  I started with the National Championship Game as I wanted to see if there was a pattern to team making the big game.

5 years of stats and millions of data points later, the answer, at the time, was looked pretty good.  12 years later…oh hell yeah there is; only Virginia Tech has bested the formula.  24 teams, one wrong.  Good night.  it’s over before it’s over.

In the past I would just post the group without notes or cut lines but a lot of fans of the Matrix via email (dave@cfbmatrix.com), twitter or texts have asked for more breakdowns of my process.

BCS Berths 2014 – First Cut  – 83 teams with talent rank worse than #32 >>>>>>>

BCS Berths 2014 – 2nd Cut – 13 teams not in top 3 talent ranks in their conference >>>>>

BCS Berths 2014 – 3rd, 4th, 5th, 6th Cut – 7 more don’t have the metrics >>>>>

BCS Berths 2014 – Final Cut of 11 teams  >>>>>

 Ole Miss – Lining up for an Elite Run?

While Ole Miss pulling a top 10 class in recruiting this year is not common, it offers hope to programs mired outside the top 20 in recruiting for the last decade.  Every great recruiting run has to start somewhere.  Ole Miss has been on the edge before.   With classes ranked #18, #11 and #15 in the last 5 years, they have proven to be attractive to recruits.  However, most of those top guys in those classes never saw the field.  When adjusted those top 20 classes ended up in the low 40s.

Recruiting is the building blocks of long term success for nearly every team (save the immortal Bill Snyder) and is certainly no guarantee of success (when you have two of the worst game coaches in Brown and Kiffin).  While you can’t win a national title on one class, you can certainly fall behind to eliminate your opportunity.  Every winner and loser of the national title in the last 9 years was a top 20 recruiter heading into their title berth year.

By Ole Miss, being a program that has been averaging the #25 class for the last 11 years, pulling this type of class, it shows that other teams and other programs can turn the corner.  They have hope of doing the same and getting on a talent accumulation roll that only a couple of teams have made happen this decade.

How rare is Ole Miss in the Top 10?

Let’s start by looking at the top 10 composite classes since 2002.  I broke them down by average recruiting rank over that period in 10 spot increments.

The top 10 recruiters of the last decade grabbed 66% of all the top ten classes.  Add to it the 24% of the #11-#20 ranked teams and that is almost 94% of all the top 10 classes.

Ole Miss has been the #25 recruiter of the last decade.  Until this year, only 7  times in the previous 11 years had a team in their recruiting rank category of #21-#40 reeled in a top ten class (Clemson, Cal, Iowa, NC State, UVA, Stanford and Nebraska).  Sure, that’s about a once every year and a half  or one in thirty classes.  Infrequent, but it’s not unheard of in college football.

What is most interesting in the ‘pre’ numbers were the class ranks for those seven teams the year before the big top 10 class.  Until Ole Miss just went from #51 to top ten, the biggest jump was Iowas from #41 to #9 and then back to #38.

Been done before

The odds show in no way should you bet on Ole Miss to continue this level of recruiting into 2014.  Nevertheless, every trend has to start somewhere.  Let’s take a look at a few notable runs since 2000.

In 2002 and 2003, Alabama had the #34 and #45 classes in the country.  Yes, this was right smack in the middle of the Franchione, Price, Shula transition.  But poor classes by current standards nonetheless.  These classes were followed by a very semi-elite run of #17, #17, #14, #14 ending in February 2007.  Then came  along Nick Saban.  If you think it’s all about ‘Bama, you would be wrong.  In his last 3 years at LSU, Nick went #13, #1, #2.  The Tigers haven’t seen two better years of recruiting since.  When he returned to the SEC in 2007 he has now rattled off, in order, the #1, #1, #5, #2, #1 and #1 classes through 2013.  Add that to the last two years at LSU, and Saban’s average class in #1.7.  The next best 8 year average is nearly double that number for USC under Coach Pete.

From 2004-2008 the Stanford Cardinal averaged a #51 class in the country.  Then came the run.  #19, #26, #21 and the huge #5 class of 2012.  In the last 10 years, no other team has made such a big change in the last 5 years versus the previous 5 years.  Unfortunately, Stanford had great attrition rates and the 2013 class was very small and ranked outside the top 25.

What’s Next

For Ole Miss, take the under.  Of the 7 teams that had top 10 class and recruited on average at #21 or worse this decade plus, they averaged a #32 class the year before.  The year after their top 10, it was at a #27.  Not a fall back to old habits, but certainly no retention of elite recruiting.

The Rebels have averaged the #25 class in the previous decade, and the trend point to them getting back to that average.  I would expect the fall back to not reach #25 or worse.  They will still recruit very well and the past predictions the future would put them with a class of #17-#23.

I know football fans.  The CFBMatrix adjusted composite recruiting rankings mean nothing.  Unless you like National Titles.  Every winner and loser of the NCG the last 8 years has averaged better than a top 25 class and all had a top ten class.  While this class may not lead to an elite future for Ole Miss, they are closer to a national title today than nearly 100 other FBS schools.

- Dave Bartoo

How much you wanna make a bet I can throw a football over them mountains?

The National Composite Team Recruiting Ranking 2013

Note- All annual rankings ARE NOT EQUAL.  Especially year to year.  The #2 composite class in 2012 would be ranked #6 this year.  So teams slightly below 2012 rankings are adding near equal talent, those that appear to have similar classes are really adding more and those above average are way above average. 

With a key to the simplicity of the CFBMatrix being adjusted recruiting rankings, it is only right to kick off every season as I have done for the last 8 years with the Composite Ranking™ sheets. It took a few years, but after the numbers showed very little overall shifting in rankings after November 1 each year and classes ending up with a final ranking on NSD  in spite of my constant viewing I just dropped it until now.

I have been doing the the CFBMatrix Composite Rankings© for 13 years in order to reduce the bias in some ranking systems and error in evaluations.  Do I believe any one site is statistically better than another for rankings, no (proven), so I put them all together to reduce error and the potential bias in some rankings.  Each ranking site has a formula for their ranks but none has proven to be better than another including their own ‘composite ranks which, when you add up and average the number available make no sense and are misleading.

This is just the start of a bunch of new numbers, angles and predictions from the CFBMatrix.  I will update the numbers the day after National Signing Day along with breakdowns of trends of all the teams and conferences.  Until then, here is your quick reference cheat sheet.

You can use the sheet to look back on 2012 classes and 4 year non-adjusted averages of classes.  No bias to formulas, rankings or sites.  Only the stats at your personal fingertips.

Thanks for being on-board for another football season with the CFBMatrix.
- Dave Bartoo – The College Football Matrix

Another little breakdown

With 124+ teams and 4 ‘major’ ranking services, the sample size to reduce bias and evaluation error is getting better. While each outlet claims to have the best folks evaluating or the best formula for rankings players, teams or what have you, a quick review was done on the Big 4.
These numbers are based on taking the combined averages and then calculating how much each ranking system is ‘off’ from the average. In this manner we can see how far each is away from the average and the frequency that a system is significantly off (+/- 5 spots) from the group average.

With ESPN only choosing to rank #1-#40, I just looked at those 40 teams.  Quick, easy and not scientific.  Lots of points to argue but consistency in the top 40 one stands out.

Closest to Average Rank Per Team

1. Rivals – 3.025
2. 247 – 3.275
3. Scout – 4
4. ESPN – 4

Number of Occurrences of a System off +/- 5 ranking spots

1. Rivals – 10
2. 247 – 12
3. Scout – 13
4. ESPN – 16

Teams under biggest disagreements between Ranking Sites

1. Miami (FL) 27 spots – 247 #17  Rivals #44
2. Stanford 25 spots – ESPN #38  247 #63
3. Wisconsin 24 spots – ESPN #32  Rivals #56
4. Penn State 21 spots – ESPN #24  Scouts #45
5. North Carolina 21 spots- ESPN #18    Rivals #39
6. Virginia Tech 19 spots- ESPN #16  Scout #35
7. Pitt  18 spots -  Rivals #22  ESPN #40

A little breakdown

With 124+ teams and 4 ‘major’ ranking services, the sample size to reduce bias and evaluation error is getting better. While each outlet claims to have the best folks evaluating or the best formula for rankings players, teams or what have you, a quick review was done on the Big 4.
These numbers are based on taking the combined averages and then calculating how much each ranking system is ‘off’ from the average. In this manner we can see how far each is away from the average and the frequency that a system is significantly off (+/- 5 spots) from the group average.

With ESPN only choosing to rank #1-#40, I just looked at those 40 teams.  Quick, easy and not scientific.  Lots of points to argue but consistency in the top 40 one stands out.

Closest to Average Rank Per Team

1. Rivals – 3.025
2. 247 – 3.275
3. Scout – 4
4. ESPN – 4

Number of Occurrences of a System off +/- 5 ranking spots

1. Rivals – 10
2. 247 – 12
3. Scout – 13
4. ESPN – 16

Teams under biggest disagreements between Ranking Sites

1. Miami (FL) 27 spots – 247 #17  Rivals #44
2. Stanford 25 spots – ESPN #38  247 #63
3. Wisconsin 24 spots – ESPN #32  Rivals #56
4. Penn State 21 spots – ESPN #24  Scouts #45
5. North Carolina 21 spots- ESPN #18    Rivals #39
6. Virginia Tech 19 spots- ESPN #16  Scout #35
7. Pitt  18 spots -  Rivals #22  ESPN #40

Thud. Plop. You name the ugly noise and that is what these classes sound like.  Sure you can argue the greatness of the average player in a few classes but come on, look at this class versus the last 4 years!

Stanford has a small class but very good talent.  Still if attrition hits this class a huge whole is going to be ripped in the Stanford depth development pipeline.  If they follow average attrition this calls will be just 6 guys in 3 years.  That is not consistent and a potential hole in the program.

USC, while uber talented, again has a numbers issue.  Sure this can be made up with EEs next year, but the consistency is going to hurt.  compounding this issue is the massive 2011 class.  Attrition is very low in that class and right now it a ‘pig through the python’.  When the attrition hits could start another experience problem.

Boston College is a train wreck.  They used to be bad, then awful now just, well, turrible.   They are recruiting a great class for the Sun Belt conference.  Any win in the ACC should be celebrated.  If there was a team ranking for non-AQs, they would be #23.

And Georgia Tech would be #12.  Second straight year of absolute garbage classes.  Johnson has been great on the field with mid-40s talent in the past, but this, IMO is too much.  in another one to two years this team will be bottom half of the ACC.

Louisville showing once again that BCS bowl wins or even appearances don’t help much with class rankings.  Yeah, I know they only had 16 spots and they got good talent but again, it is the consistency in the program classes year over year that is a concern.  Coupled with last year’s below average class, this is not a good start for a move to the ACC (and no, there is no ‘bump’ for AQ to AQ conference transfers).

While a #14 class is still very good, it is the worst ranking for Texas, ever.  Mack can’t win with the best talent and one of the best attrition rates,  so what happens now.  I am still amazed that one LIAB national title and 2 conference titles get this much rope in a program that should consider any win total under 10 a failure.

For some reason it makes me laugh that one guy, Tuberville, is directly connected to two programs on this list in the last 90 days.  Let me know how that turns out Bearcats.  Uggghhh!

-Bartoo .