Colorado Recruiting Class Comparison: 2006-2015

On the boards, the debate never ends.  Do stars even matter?  JJ Watt and Aaron Rodgers would tell you no, but are those names exceptions to the rule, or the faces of a flawed system?

This piece will hopefully help solve the riddle when it comes to Colorado football, as we break down each class from 2006-2015 and try to determine what exactly we can expect from each recruit that joins the program moving forward.

The first breakdown I wanted to show was an average of RR rating for each class.  For those who don’t know, RR is how rates their prospects. I chose to use this metric rather than average star rating because it gives a more exact and true breakdown of each individual in each recruiting class.  Why?

  • 2* on Rivals are rated anywhere from 4.9 to 5.4, with 5.4 being the highest possible.
  • 3* on Rivals are rated anywhere from 5.5 to 5.7
  • 4* on Rivals are rated anywhere from 5.8 to 6.0.
  • 5* on Rivals are rated at 6.1.

This further illustrates the point that not all 3* are created equally.  A 5.7RR 3* recruit should not be viewed as a similar commit as a 5.5RR guy, and those who review strictly average star rating will not see that distinction in their metrics.

Here are how the Colorado Buffaloes’ recruiting classes rank over the past 10 seasons.

Average RR:

  • 5.633 – 2008
  • 5.524 – 2012
  • 5.522 – 2007
  • 5.500 – 2009
  • 5.494 – 2015
  • 5.488 – 2010
  • 5.452 – 2014
  • 5.443 – 2013
  • 5.439 – 2011
  • 5.336 – 2006

A few interesting notes breaking down this data:

— The bottom 3 classes are all “transition” classes, or classes that newly hired head coaches had to quickly put together following their arrival in Boulder.  This would suggest that some continuity among head coaches would serve the program well in the long run in order to obtain and retain quality talent.

  • 2013 was the class transitioning from Embree to MacIntyre
  • 2011 was the class transitioning from Hawkins to Embree
  • 2006 was the class transitioning from Barnett to Hawkins

— MacIntyre’s transition class was the highest among the three, but his other classes have not, thus far, had as much initial quality as Hawkins or Embree.  Of the 7 “non-transitional” classes in the past decade, Mac’s classes rank 5th (2015) and 7th (2014) respectively.

Of course, this data doesn’t take into account how those recruits actually performed on the field, and whether or not they actually completed their college career in Boulder.

Therefore, the next step is to track the percentage of recruits in each class who both “contributed” and “graduated” for the Buffaloes.  Of course, “contribution” is tough to measure and is somewhat subjective, so I tried to include anyone that started games in multiple years or had solid  stats in at least one year while on campus.

For “graduated” players, it’s really whether or not the player exhausted their eligibility with the program.  Therefore, recruits who graduate transferred to finish their career elsewhere are excluded, while players who left early for the NFL Draft are included.

Here are the recruiting classes ranked by “Contribution %” and “Graduating %”:

       Contribution %:                                                           Graduation %

  • 2008 – 66.67%                                                             2008 – 57.14%
  • 2009 – 55.00%                                                             2011 – 56.52%
  • 2007 – 48.15%                                                             2006 – 50.00%
  • 2010 – 45.83%                                                             2010 – 45.83%
  • 2006 – 45.45%                                                             2007 – 44.44%
  • 2011 – 43.48%                                                             2009 – 35.00%

A few interesting notes breaking down this data:

— Of course, any of the classes from 2012 forward aren’t included in this comparison because we don’t have full data as many of the prospects are still eligible and playing.

— Of the 137 players we signed in the classes from 2006-2011, 69 (50.36%) of them contributed for Colorado, and 66 (48.18%) exhausted their eligibility with the program.

— The 2011 class is an interesting data point, as they had the fewest percentage of contributing players, while graduating the second highest percentage of players.  This realization suggests that this class had a lot of talent lingering around the program that didn’t really contribute to the team’s overall success.

— The 2008 class, easily the program’s highest ranked class of the past decade, has received a lot of scrutiny from fans as being “overrated” because big names like Bryce Givens, Lynn Katoa, Darrell Scott, and Max Tuiti-Mariner failed to meet expectations in Boulder.  However, this data shows that despite those issues, it still remains the most productive class from 2006-2011, with the highest contribution and graduation percentages.  Some names who helped the program:  Curtis Cunningham, Ryan Deehan, Tyler Hansen, Jon Major, Shaun Mohler, Will Pericak, Ray Polk, Doug Rippy, and Rodney Stewart.

— You’ve heard the phrase “leaving the cupboard bare” associated with this program a lot in the past 4-5 years.  This data is your proof that it was, in fact, true.  The last three complete classes (2009-2011) shows that just 32/67 (47.8%) contributed on the field at Colorado, and just 32/67 (47.8%) exhausted their eligibility.  The 2012 class has already seen 13 early exits from the program without contribution, so in a best case scenario, that class will see 16/29 (55.2%) in both contribution and exhausted eligibility.  Those type of numbers make it absolutely impossible to build quality depth and sustain success through injury, which is inevitable in the game of football.

— The one thing that Coach MacIntyre has seemingly been able to improve in his short time here is player retention.  As noted, the data shows that the classes from 2006-2011 had less than 50% of their players exhaust eligibility as Colorado Buffaloes.  The 2012 class might show slight improvement there, but it will be slight at best.  The classes that Mac is responsible for, 2013 and forward, have thus far seen just two players exit the program.  Of course, those classes are still fairly young, and those numbers will undoubtedly rise, but it does seem reasonable to assume that the losses in those classes will not match the previous two regimes.  Coach MacIntyre’s 2013 class would need to see 10 defections in the next few years to match the jail break averages of the past six years.

Now, to help solve the old debate about whether or not stars actually do matter.  Some of the best Buffs over the past decade (David Bahktiari, Addison Gillam, Jalil Brown, and Rodney Stewart) did not receive high marks during the recruiting process, which has led some to believe the system is flawed and means little when it comes to success on the field.  Overall, though, what does it generally mean when Colorado earns the commitment of a 2* versus a 3* versus a 4*?  Find out below:

RR — Prospects — Contributor — Cont % — Graduate — Grad % — Drafted — Draft %

4.9             10                  3             30.00%            4               40.00%         1           10.00%

5.0              2                   1             50.00%            1               50.00%         0            0.00%

5.1              1                   0              0.00%             0                 0.00%         0            0.00%

5.2              8                   3             37.50%            2               25.00%         0            0.00%

5.3              9                   6             66.67%            5               55.56%         1          11.11%

5.4             19                  8             42.11%            7               36.84%          0           0.00%

2*              49                 21             42.86%          19              38.78%          2            4.08%

RR — Prospects — Contributor — Cont % — Graduate — Grad % — Drafted — Draft %

5.5              33                15             44.45%         13               39.39%        0             0.00%

5.6              26                15             57.59%         17               65.38%        2             7.69%

5.7              13                 7              53.85%           9                69.23%       0             0.00%

  3*               72               38             51.39%         40               54.17%        2            2.78%

RR — Prospects — Contributor — Cont % — Graduate — Grad % — Drafted — Draft %

5.8                7                 5              71.43%          4                 57.14%       1           14.29%

5.9                4                 3              75.00%          2                 50.00%       0             0.00%

6.0                3                 1              33.33%          1                 33.33%       1            33.33%

6.1                2                 2              100.00%        1                 50.00%       1            50.00%

4*/5*             16              11               68.75%         8                50.00%       3             18.75%

TOTAL           137          70              50.36%        67                48.18%       7              5.11%

So what does this mean?

  • A 2* historically contributes at Colorado roughly 43% of the time, graduates 39% of the time, and has a 4% chance to get drafted into the NFL on average.
  • A 3* historically contributes at Colorado roughly 51% of the time, graduates 54% of the time, and has a 3% chance to get drafted into the NFL on average.
  • A 4/5* historically contributes at Colorado roughly 69% of the time, graduates 50% of the time, and has a 19% chance to get drafted into the NFL on average.

—  A Colorado 4/5* recruit was more than 5.6x more likely to be drafted than a 2/3* recruit over the past decade.

— A Colorado 4/5* recruit was more than 60% more likely to “contribute” than a 2* recruit and more than 30% likely to “contribute” than a 3* recruit.  A Colorado 3* was roughly 20% more likely to “contribute” than a 2* recruit.

— A Colorado 3* recruit was roughly 40% more likely to exhaust their eligibility at Colorado than a 2* recruit.

— The real break in the data seems to appear between the 5.5RR 3* and the 5.6RR 3*.  If you break up the data as such:

  • Of the 82 players at or below 5.5RR 3* rating, the data shows we’ve had 43.9% contribute, 40.2% graduate, and 2.4% of the players were drafted into the NFL.
  • Of the 55 players at or above 5.6RR 3* rating, the data shows we’ve had 58.2% contribute, 61.8% graduate, and 9.1% of the players were drafted into the NFL.

Just because, if you apply those realities to the 2015 class, which has just 3 players ranked 5.6RR or higher (Patrick Carr, Tim Lynott, and NJ Falo) it becomes clear that MacIntyre still needs to find ways to attract better talent to Colorado if we expect to improve exponentially on the field.

In conclusion, the numbers are clear.  Sure, every once in awhile you’ll find a diamond in the rough, but more often than not, getting more highly ranked players leads to higher contribution on the field, and a higher percentage chance those players will take their talents to the next level.

It is true that it is far from a certainty that a high ranked player will pan out at Colorado (just look at Lynn Katoa, Bryce Givens, or Darrell Scott for proof), but that doesn’t change the fact that its far more likely that they will succeed than a 2* player, as the data has proven.

Recruiting is a numbers game, not on an individual level, but on an overall level.  This is a game that Colorado has been losing over the past decade.  If you are able to bring in highly ranked players consistently, some of those players will pan out.  Maybe not necessarily the ones you project off the bat, but enough that your overall production will have some quality.  For example, once again refer to the 2008 class, the most productive class of the decade besides several high-end busts.  If Colorado can improve their retention percentages and find a way to recruit better players to Boulder, they just may find themselves opening the door to that Pac-12 basement.


Bubble Math 2015 v2.0


As we alluded to in the first version of our 2015 Bubble Math, this year’s version of the bubble is among the weakest we’ve seen this decade, and the parody is top-notch.  Our second version of the analysis has seen an incredible 10 new teams added to the data set since last week, making our data set a total of 30 teams.

The new teams added to the data set are:  Boise State, BYU, Illinois, LSU, Oregon, Stanford, Texas A&M, Tulsa, UCLA, and UMass.

As always, only 16 teams in the data set are grouped into Lunardi’s “bubble”… so the remaining 14 have worked their way into the field or played their way out of contention.  This week, Iowa, Oklahoma State, Ole Miss, St. John’s, Temple, and Xavier have earned their way safely into the field.  Clemson, Florida, George Washington, Kansas State, Michigan, Richmond, Tennessee and Wyoming have stumbled their way off the bubble according to Lunardi’s latest bracketology.

To clarify, Lunardi’s top two groupings (labeled Last 4 Byes and Last 4 In) represent the 8 teams he projects to make the Tournament.

– If RPI metrics were used solely to determine the last 8 teams getting in from the data set, these teams would make the field:  Oklahoma State, Colorado State, Temple, UMass, Ole Miss, UCLA, Xavier, and St. John’s.

– If KenPom metrics were used solely to determine the last 8 teams getting in from the data set, these teams would make the field:  Oklahoma State, Xavier, Ole Miss, LSU, BYU, Florida, St. John’s, and Stanford.

– Kansas State leads all teams in the data set with an impressive 4 victories over RPI top 25 opponents.  Unfortunately, they are also just 8-13 vs. RPI top 200 opponents, worst among the teams in the data, and thus are not really being considered for the Tournament at this point.

– The only other team to own multiple wins vs. RPI top 25 opponents and not be considered “on the bubble” at this point?  Tennessee.

– The least impressive resume among the “Last 4 Byes” category has to belong to Texas A&M at this point.  The Aggies are 0-5 vs. RPI top 50 opponents, and their only win vs. a team expected to be in the field is @ LSU.  They have no bad losses, which is more than can be said for nearly everyone on the bubble, but the committee tends to favor quality wins over the lack of bad losses.

– Compare that resume with say, NC State, who is currently projected among the “Last 4 In”.  NC State is 3-7 vs. the RPI top 50, including wins over Duke, @ Louisville, and Boise State.  They also own two more RPI top 100 victories than Texas A&M.  A&M has one more road win and a better winning % vs. RPI top 200 opponents, and NC State owns a bad loss @ RPI #128 Wake Forest, but wouldn’t you rather see a team in the Tournament capable of beating big names?

–  George Washington is the only team in the data set to avoid multiple losses vs. opponents ranked outside the top RPI that isn’t projected to be in the field.  Why?  They are just 3-7 on the road, own just 9 wins vs. RPI top 200 opponents (t-24th in the data set), and just 3 wins vs. RPI top 100 opponents (next to last in the data).

– The 30 teams in the data set are a combined 112-146 on the road, a winning percentage of 43.4%.

– The 30 teams in the data set are a combined 75-137 (35.4%) vs. RPI top 50 opponents

– The 30 teams in the data set are a combined 163-205 (44.3%) vs. RPI top 100 opponents

– A team not among the data set that has a chance to make some late noise?  UTEP.

The Miners own solid non-conference victories over Xavier, Washington State, and Kent State, and also played Colorado State, Arizona, and Washington very tough.  If they can find a way to win @ La Tech next week, you may start to see them pop up on some people’s radar.  They aren’t deserving of a bid as it stands today, but they aren’t being talked about enough either.

Be on the lookout for Bubble Math v3.0 early next week!


We’ve reach February, which means we’ve officially walked through the doors of bubble watch season in college basketball.  This is the fourth season I’ve spent time trying to decipher which teams on the bubble are worthy of inclusion in the NCAA Tournament, and without a doubt, it remains one of my favorite projects.  This season, however, has the potential to be the craziest one to date.  The bubble is extremely weak overall, and there are dozens and dozens of teams piled up trying to fight through the muck and earn one of the last eight or so spots into the Big Dance.

First of all, let me talk about what it is, exactly, that I track, and how I determine which teams to include in my data set.  As I’m sure you know, ESPN’s Joe Lunardi puts together bracketology updates throughout the year as we approach March, and he has teams around the bubble broken into these specific categories:  Last 4 Byes, Last 4 In, First 4 Out, Next 4 Out.  Any team that appears in one of those four categories starting February 1 through the end of the season gets included into my data set, because those teams are determined to be “on the bubble”.  As mentioned above, I expect the number of teams included in the data set to be the highest ever, given all of the parody.  Once a team is included in the data set, they are not removed for any reason.  Should teams work their way securely into the field, or off the bubble completely, they still remain in the data and can be used as data points for the other teams to strive for (or against).

I break out each bubble teams resume in various categories, including RPI top 25, top 50, top 100, and top 20 W-L records, RPI 200+ losses (bad losses), strength of schedule, and road W-L records, among others.  Combined, these factors create an image of each team’s resume that can be compared to one another to determine who is worthy of inclusion into the Tournament.

The first version of my data set includes 20 teams, 16 of which sit in the categories we discussed before, and 5 total who moved out of those categories within the past week.  Oklahoma State has worked their way safely into the field, while Michigan, Florida, and Davidson have since dropped off the bubble.  The 16 teams still sitting on the bubble are as follows:  Clemson, Colorado State, George Washington, Iowa, Kansas State, Miami FL, NC State, Old Dominion, Ole Miss, Purdue, Richmond, St. John’s, Temple, Tennessee, Wyoming, and Xavier.

Let’s break down the numbers for this first group:

— RPI considers Colorado State to be the “best” team in the data set, ranked 28th by their metric.

— KenPom loves Oklahoma State, ranked 26th, but of the teams still on the bubble, Xavier is the highest ranked team, at 27th overall.

— No team in the data set owns more than 2 victories against RPI top 25 opponents, in fact, just 4 teams own more than 1 win against such opponents, Tennessee, Kansas State, St. John’s, and Xavier.

— Kansas State and Oklahoma State both own 4 RPI top 50 wins, the most amongst the data set.  Kansas State has also played the most games vs. RPI top 50 opponents (10), while Old Dominion has only played 2 such opponents (should be noted that ODU is 2-0 in those matchups).

— Colorado State, sitting at 19-4, has earned 11 of their victories over opponents ranked outside the RPI top 200.  Outside of Wyoming and Temple, who has 10 and 8 such wins respectively, no other team in the data set has more than 6 such wins against “weak” opponents.  This realization can further be highlighted by their strength of schedule numbers, where Colorado State is just one of three teams in the data to own a SOS outside the top 150 in the country.  Wyoming’s SOS is ranked 306th nationally, among the worst of any team in the nation, and thus, sits last in our data set as well.

— There are just 7 teams in the data set with above .500 road records.  Colorado State, Temple, and Davidson lead the way with 5 victories away from home.  Of the teams in high major conferences, only Miami FL (4-2) and Tennessee (3-2) own winning records on the road.

— Who is overrated?  To me, Iowa and Temple are leading the way right now.

Temple, of course, has a resume built entirely on their win over RPI #1 Kansas.  Of course, you can’t ignore that victory, but does one win make a resume?  That’s up to you, I suppose, but in my eyes, the rest of their body of work is not worthy of inclusion right now.  Outside of that win, they’re next three resume pegs are wins @ Uconn and over La Salle and LA Tech.  Not great.

Iowa’s resume is also incredibly mediocre.  Lunardi currently has them among the “Last 4 Byes” which means they are almost safely into the field, yet, their resume suggests they are overrated.  They are just 2-7 vs. the RPI top 50, and just 8-9 vs. the RPI top 200.  They’ve swept Ohio State, sure, but they’re best win OOC is currently Pepperdine.  Ohio State remains the only team they’ve beaten all year expected to make the Tournament.

Be on the lookout for the next version of Bubble Math next week!