By Appaloosa

Come discuss this article HERE!

By Appaloosa.

By Appaloosa

Fourth Quarter Index
Now for the quantitative measure of fourth quarter ability promised at the beginning of this article. First, I reduced the data set to the 37 quarterbacks who started 125 or more games to make for a more statistically significant sample of wins and losses. Then I calculated an index based on comeback win to blown save ratio, comeback wins divided by total starts, QB ratio and points scored per start. The formula is shown below:

Index = (4QW/4QL+4QW/Starts) x QB Ratio x Pts/Start x 10

Where:
4QW = fourth quarter wins
4QL = fourth quarter losses
Pts/Start = total quarterback points divided by number of games started

The base for the index is the ratio of fourth quarter wins and blown saves, which is a good initial measure of the clutchness of a quarterback. However, some credit needed to be given for being able to have a large number of come from behind victories over the course of a career rather than a small handful of wins and losses. Therefore, the number of comeback wins divided by total starts was added to the ratio. Because this total needed to be adjusted to reflect the relative importance of the quarterback to his team, I multiplied the sum by the QB ratio and by points per start. Finally, to make the resulting index a little more user-friendly, I multiplied the total by 10, which generates a number between 10 and 100 in most cases.

The signal callers who started 125 or more games with the highest 4th Quarter indices are Peyton Manning (150.7), Jake Plummer (122.2), Joe Montana (119.2), Dan Marino (114.5), and Donovan McNabb (101.5). This list consists of two Hall of Famers, one probable and one possible future Hall of Famer, and one 4th quarter comeback specialist. The bottom five are Archie Manning (17.2), Jim Harbaugh (19.4), Jim Everett (26.0), Troy Aikman (26.5), and Dan Fouts (26.8) – an interesting grouping of two Hall of Famers and three quarterbacks who spent most of their careers struggling on bad teams. Aikman’s index is dragged down his low QB ratio and points per start, while Fouts has the worst comeback to blown save ratio of any Hall of Famer. Other indices of interest are Kerry Collins (46.0), Warren Moon (67.3), Jim Kelley (68.5), John Elway (71.3), Terry Bradshaw (72.8), Brett Favre (76.3), Steve Young (80.4), and Brad Johnson (83.4).

Comeback Kings
Quarterback Performance in the 4th Quarter 1970-2008

By Appaloosa
February 2009

Introduction
One aspect of quarterback play that really sticks in the memory of the NFL fan is the fourth quarter comeback. Nearly 20 years later, Elway’s “The Drive” is still famous. Several quarterbacks have been nicknamed “Captain Comeback” because of their fourth quarter heroics. The reputation for being “clutch” is one of the things that can get a quarterback voted into the Hall of Fame. However, there is no really quantitative way of assessing the “clutchness” of a quarterback, so the arguments over who is a comeback king versus who is merely a joker often come down to selective memory or raw totals of career comebacks. In the following article, I attempt to assign a number to the clutchness of a quarterback and try to answer some of the questions I have seen on various NFL message boards.

Before I venture into the meat of my study, I would like to mention a few caveats. As has often been stated, the glamour boy quarterbacks always get too much credit when his team wins and too much blame when the team loses. Many fourth quarter comebacks have been the result of the defense getting a timely turnover or special teams running a kick back for a touchdown. Fourth quarter losses can be due to things beyond a signal caller’s control such as fumbles by running backs or stupid penalties by the defense. Because of the scope of this study, these factors have been largely ignored. However, I believe that good luck and bad luck generally cancel each other out and that ultimately most fourth quarter comebacks and blown saves (to use a baseball term) are the responsibility of the quarterback.

Data Selection
This study of comebacks is based on the careers of all quarterbacks who entered the league after the 1970 merger and who started a minimum of 100 regular season games. Fifty-five signal-callers from Terry Bradshaw to Drew Brees meet these criteria. For each quarterback I recorded total number of starts, number of wins, number of losses, number of fourth quarter wins, number of blown saves, the number of points the quarterback contributed to his team, the total number of points scored by the team when the quarterback started, and career passer rating. All numbers are for the regular season only because most of the statistics available are for the regular season. As playoff statistics are generally kept separate from regular season statistics, including playoff statistics would have required recalculating a lot of career numbers.

For the purposes of this study fourth quarter wins were defined as victories won when the team was behind or tied at the end of the third quarter. In the case of overtime, fourth quarter efforts that tied the game followed by a victory in overtime were counted as come from behind victories. Overtime losses after game tying efforts were not counted.

Blown saves were defined as losses in which the team led at the end of the third quarter. Again, in the case of overtime, the only losses that counted against the quarterback were ones where his team blew the lead and lost in overtime.

As a rough measure of a quarterback’s contribution to the total scoring of his team, I added up each player’s career touchdown passes and rushing touchdowns and multiplied the result by 7. This number was compared to the total number of points scored by the quarterback’s team in each game in started. As described below, these numbers were used in an attempt to identify which quarterbacks were most responsible for their team’s success instead of being mainly game managers.

By the Numbers
Currently the titleholder for most fourth quarter comebacks in a career is Dan Marino with 33 based on the definition given above. Marino is followed by Warren Moon and Peyton Manning (28 each), Jake Plummer (24), and John Elway and Vinny Testaverde (23 each). The quarterback with the fewest fourth quarter comebacks to date is Kurt Warner with 6 followed by Bobby Hebert (7) and Archie Manning, Jim Everett, Richard Todd, and Matt Hasselbeck with 8 each.

The player with a minimum of 100 career starts with the fewest blown saves is Tom Brady with only a single regular season blown save to his name. Tied for second place are Neil O’Donnell and Kurt Warner with 4. Dan Pastonini, Joe Theismann, and Jake Plummer each had 9 career blown saves. The all time champion for career blown saves was Vinny Testaverde with 30, followed by Boomer Esiason with 26. Dan Fouts, Warren Moon, and Drew Bledsoe are tied for third with 24 a piece.

ESPN.com conducted a poll of the NFL head coaches to see which players are the dirtiest, the smartest offensive players (non-QB), smartest defensive players, and the best owner. Here are the tallies:

Dirtiest player:

1. Rodney Harrison (shocker)
2. Roy Williams (Mr. Horse collar himself)
T3. Antonio Pierce, Hines Ward, Kevin Mawae, Olin Kreutz

Smartest offensive player (non-QB):

1. Hines Ward
T2. Kevin Mawae, Jeff Saturday, Randy Moss
T5. Jon Runyan, Torry Holt

Smartest defensive player:

1. Zach Thomas
2. Ray Lewis
T3. Ed Reed, John Lynch
T5. Corey Chavous, Michael Strahan, Warren Sapp

Best Owner:

1. Dan Rooney (Steelers)
2. Pat Bowlen (Broncos)
3. Jerry Jones (Cowboys)
*8 other owners received votes but they are not listed.

A little side note: Patriots safety Rodney Harrison, widely regarded as the dirtiest player in the league, received 11 out of 17 votes, confirming his notorious reputation. He’s had 14 personal fouls called against him since 2001. His teammate, Richard Seymour, has had 13 in the same time span.

Those classy Pats.

Bustology – Schools

By Appaloosa

April 2006

For more information on the development of the Comprehensive Bust Rating (CBR), see Bustology 1 – Quarterbacks.

Rating the Universities

It is not fair to draw conclusions about the quality of offensive specialists from a school based on the performance of one or two players selected in the first round.    Any university or college that produced fewer than three first round quarterbacks, running backs, or wide receivers between 1970 and 2002 were not included in the analysis.  For the 39 schools remaining, the adjusted CBR for each offensive specialist was added and averaged to provide information on which universities provide the best first round values and which produce the biggest duds.

In case you were wondering, the University of Miami (The “U”) had the most total offensive specialists (13) drafted in the first round between 1970 and 2002.  Next most, with 12 each, are Florida, Penn State, and Tennessee.  Eight running backs were drafted from Penn State, which is the highest number at that position.  To paraphrase Samuel Johnson, this is another instance of the triumph of hope over experience.  The college team that produced more first round wide receivers was the Tennessee Volunteers with eight of widely varying quality.  Oregon, UNCL, and Washington State each had three quarterbacks selected in the first round.

The lowest total for any university was –276 for Pittsburgh, largely thanks to Tony Dorsett and Dan Marino and the fact that no huge ex-Panther busts have been drafted since 1970.  However, the prize for the lowest average goes Syracuse University, alma mater of Ark Monk, Marvin Harrison, and Donovan McNabb, all of whom have CBRs well into the negatives.

But you aren’t reading this article because you are interested in finding out which colleges have turned out successful NFL offensive talent.  You want to know about the failures.  The award for the highest CBR total is Texas A&M with a score of 310.  Although only Larry Stegent is the only truly epic Aggie dud, the total was helped along by the likes of Bubba Bean and Rodger Vick.  Texas A&M is doing a commendable job in producing so many busts, especially since all of them are running backs.  I guess one conclusion that can be drawn is that you should hope your favorite team never selects a running back from Texas A&M in the first round.

The highest average for any university belongs to Nebraska and Oregon, which tied with a mean of 59.  Nebraska earned this dishonor on the strength of Johnny Rodgers and Jerry Tagge.  The Cornhuskers should be grateful that Irving Fryar was such a success or Nebraska would have beaten Texas A&M for a title no one much wants.  As for the ex-Ducks, who could forget Akili Smith, though I am sure the Bengals would like to?  Joey Harrington has also made his contribution to the cause.

A few universities other than Nebraska and Texas A&M deserve dishonorable mentions for the number of busts they have imposed on the NFL.  Of the seven first round running backs and wide receivers that came from the University of Colorado, not a single one achieved a negative CBR.  Although there were no epic ex-Buffalo busts except possibly Rae Carruth, only a couple of wide receivers even came close to meeting expectations.  The total CBR for all seven Colorado draft picks is 249 with an average of 36.  Another school with more than its share of busts is Houston, the alma mater of such immortals as Elmo Wright, Willis Adams, and David Klinger.  With only five players drafted in the first round, Houston’s total CBR of 248 nearly rivals that of Colorado with eight players.

The only other school with a total CBR greater than 200 is Clemson, which had five first rounders, none of whom was better than slightly below average.  The main contributions to this total came from Terrance Flagler and Perry Tuttle, but even Steve Fuller donated his mote (2) to the pot.

Running Back Schools

Some universities might produce NFL talent at one position, but most of the players of another position might generally be duds.  For example, I have mentioned more than once that Penn State might be Linebacker U, but you do not want your team drafting a running back who played his college ball for the Nitani Lions.  The good schools for first-round running backs are Florida, Ohio State, and Texas.  The not so good schools, other than Penn State are Colorado, Michigan, Nebraska, Richmond, and Wisconsin.

Wide Receiver Schools

As stated above, more wide receivers have been selected in the first round from the University of Tennessee than from any other school.  However, the quality of these receivers has been decidedly mixed.  On one hand are Stanley Morgan (CBR –77), Willie Gault (-35), and Anthony Miller (-49), but on the other are Anthony Hancock (74) and Marcus Nash (128).  The jury is still out on Donte Stallworth who currently has a CBR of 37, but no wide receiver drafted in the 21^st Century has yet achieved a negative CBR, so there is still plenty of time to determine if he will uphold the reputation of “Wide Receiver U.”

A couple of other schools that have consistently produced good first round wide receivers are North Carolina State and the University of Southern California.  Although none of the former Wolfpack players have been spectacular there have been no busts to date either.  The only NC State alumnus not in negative figures is Koren Robinson (43), who is still relatively new to the league.  The Trojans produced four straight successful wide receivers such as Curtis Conway (-42) and Johnnie Morton (-48) until along came RJ Soward (102) who completely destroyed the perfect record.

 There are two schools that you hope your team doesn’t draft a wide receiver from in the first round.  Arizona State has had four wide-outs drafted in the first round since 1970 but only one of these was an even moderate success.  The rest are JD Hill (CBR 39), Steve Holden (79), and Aaron Cox (56).  The University of California, Los Angeles can’t even claim to have produced a moderate success.  All three of the Bruin wide receivers drafted in the first round have positive CBRs with the highest (72) belonging to Freddie Mitchell.  Freddie may have only been drafted in 2001, but he has a higher CBR than any receiver drafted in either 2001 or 2002.

Quarterback Schools

The University of Miami (AKA “The U”) has been called “Quarterback U” with reason.  The best first round draft pick quarterback who played his college ball for the Hurricanes was Jim Kelly (CBR –123), who is now in the Hall of Fame.  But even Vinnie Testaverde, who struggled at times during his long career, ended with a respectable CBR of –69 although it took him nine years in the league before his adjusted score was on the negative side of the scale.  

Stanford is the only other school with more than one quarterback drafted in the first round and a 100% success rate.  Both Jim Plunkett (CBR –34) and John Elway (-83) were selected first overall in their respective drafts and each won at least one Super Bowl before his career was over, though neither of them for the team that drafted him.  Other than Miami and Stanford, the only other schools that have had more than one quarterback selected in the first round and have a negative cumulative score are Tennessee (cumulative CBR –42), UCLA (-30), and Purdue (-1).

As for quarterback bust alma maters, dishonorable mentions should be awarded to Houston and Washington State, which both have cumulative CBRs of 142.  Both Andre Ware (CBR 38) and David Klinger (104) played for the Houston Cougars in their college days.  As for Washington State (also the Cougars), Jack Thompson’s CBR (36) is balanced by that of Drew Bledsoe (-36), but leave it to Ryan Leaf and his CBR of 142 to screw things up.

But currently the worst quarterback school is the University of Oregon, with a cumulative CBR of 256.  Chris Miller (CBR –21) might have been a modest success in his day, but his performance has been overwhelmed by the reek of Akili Smith and the taint of Joey Harrington.  As they say, “If it quacks like a Duck than it will probably be a dud.”

Bustology – Wide Receivers

By Appaloosa

April 2006 (Revised April 2007)

For more information on the development of the Comprehensive Bust Rating (CBR), see Bustology 1 – Quarterbacks. 

Development of the WR CBR

Development of the wide receiver CBR began with surveying the list of wide receivers selected in the first round of the draft since 1970.  The list of first round wide receivers was split into two groups based on the year they were drafted.  Anyone drafted prior to 1994 was placed in the retired division, and anyone drafted 1994 to 2002 was placed in the active group.  This was because 1995 is the earliest draft in which a first round wide receiver (Joey Galloway) saw significant playing time in 2006.  Although first-round wide receivers drafted as recently as 2001 are no longer in the league, prematurely retired (and therefore mostly bad) players were not added to the retired list because they would skew the data.

Once the subjects had been divided into the two groups, the following raw statistics were recorded for each wide receiver: 1) draft position, 2) number of games played, 3) career receiving yards, 4) career receptions, and 5) career touchdowns.  These data were used to calculate: 1) yards per reception, 2) touchdowns per reception, and 3) receptions per game.  Within the retired group of 67 wide receivers from Ken Burrough to Derrick Alexander, the mean and standard deviation were determined for all the raw and calculated statistics listed above.  Representative career means for a first round wide receiver drafted between 1970 and 1994 are 116 games played, 5,276 receiving yards, 34 career touchdowns, and 3 receptions per game.  However, not all statistics are useful for determining the quality of a wide receiver’s career, especially for determining the degree of “bustitude”.  For example, yards per catch is a statistic that could not be used because the standard deviation is larger than the mean, which means that in order to receive the maximum CBR, a wide receiver would have to have negative yards per catch.  Other statistics, such as career receptions and touchdowns are too dependent on career length and other statistics that would limit the independence of each aspect of the ranking.  The statistics that were chosen for determining wide receiver bustitude are career receiving yards, touchdowns per catch, and receptions per game.  In addition, career length, in number of games played, is a good indicator of a wide receiver’s success.  All four of these statistics were given approximately equal weight in calculating a raw CBR.  Then a final variable factor was added to the CBR for all offensive specialists.  This factor is based on the player’s draft position and increases on a logarithmic scale the higher in the draft he was selected.  This factor was added to account for the greater expectations that are accorded to high first round picks.  In the discussion below, this factor is referred to as the “rank bonus.”

For those who are interested, the wide receiver CBR formula is: 100(((116-G)/200)+(2.5(0.0964-(T/R)))+((2.992-(R/G))/13)+((5,267-Y)/17,500)+((1/P)/5)), where:

G = Games

P = Draft position

R = Receptions

T = Touchdowns

Y = Total Receiving Yards

Anti-Bust Wide Receivers

Once the wide receiver CBR formula was devised, it was applied to the wide receivers in both the retired and the active lists.  CBRs of less than –100 (the ultimate first-round anti-busts) were recorded for James Lofton (-108), Jerry Rice (-217), and Tim Brown (-130) from the retired list.  Lofton is already in the Hall of Fame, and Rice and Brown should be joining him in the next few years.  The only active wide receiver with CBRs less than –100 is Marvin Harrison with a current CBR of –103.  Other CBRs that readers may find interesting for non-busts are –77 for Stanley Morgan, -13 for Lynn Swann, -99 for Art Monk, -96 for Irving Fryar, -64 for Michael Irvin, -74 for Andre Rison, -46 for Keyshawn Johnson, –69 for Randy Moss, and –51 for Torry Holt. 

Lesser Busts

Wide receiver busts seem to have been less spectacular in the past than quarterback and running back busts.  Most of those who accumulated respectable CBRs are long forgotten except by the fans of the teams who drafted them and have long memories.  Here are a few examples: Elmo Wright, drafted by the Chiefs in 1971 and averaged 1.3 receptions per game and achieved a CBR of 73, an even better dud was Terry Beasley who scored 3 touchdowns in his career and a 89 CBR, and Randy Burke drafted by the Colts in 1977 and only 30 career receptions.  One year that was especially bad for wide receivers in the draft was 1973, which produced classic busts such as Steve Holden (79 CBR), Darryl Stingley (also 79 CBR), and Barry Smith (77 CBR).  1992 was an even worse year for drafting wide receivers with such gems as Anthony Hancock (74 CBR), Lindsay Scott (93 CBR), and Perry Tuttle (84 CBR).

The names of some of the more recent lesser wide receiver busts might be more familiar.  For example, Rae Carruth’s CBR is 83.  Kevin Dyson was drafted five spots ahead of Randy Moss, but has a CBR of 45.  Peter Warrick has established himself as a minor dud with a CBR of 40, but he is far from the worst wide receiver drafted in 2000.  There will be more on R.J. Soward below, but Sylvester Morris should not be forgotten since he achieved a rating of 85.

Mega Busts

Unlike quarterbacks and running backs, wide receivers do not seem to bust out quite as spectacularly, though it is interesting to note that most of the wide receiver mega busts have played in the last decade.  Only four first round pick wide receivers have accumulated the statistics worthy of a mega bust, and even their bustitude pales in comparison with the great busts among the other offensive specialists.  For example, the fourth biggest wide receiver bust of all time was R.J. Soward whose CBR of 102 would only place him 11^th on the list of quarterbacks and 7^th among running backs.

The three remaining candidates for biggest wide receiver bust since 1970 have much in common.  Each caught fewer than 20 receptions in his career, and none of those catches were for touchdowns.  All three played for major college football powers in their day and were drafted in the second half of the first round but their NFL careers were disasters.  Yatil Green played for the Hurricanes and then stayed in Miami for what passed for a pro career.  In 9 games he caught a grand total of 18 passes for 234 yards and a CBR of 115.  There’s little to choose between Green and the runner up, Johnny Rodgers who was selected by the Chargers out of Nebraska.  Although Rodgers lasted for 17 games, he only caught an average of one pass per game and coincidentally had the same number of yards as Green.  Because he caught fewer passes per game, Rodgers has a slightly higher CBR – 118.

And now to crown the king of wide receiver busts.  Marcus Nash played for “Wide Receiver U” and looked pretty good catching balls from Peyton Manning.  The Broncos apparently thought he would look just as good snagging passes from an aging John Elway and selected him with the last pick of the first round in 1998.  Unfortunately for the Broncos, Nash turned out to be a spectacular bust.  In an entire NFL career of 7 games he caught all of 4 passes for 76 yards and his CBR was 128.  The irony of Nash’s career is that this loser has two Super Bowl rings, which is one of the reasons why Super Bowls were not factored into the CBR.  Now Nash is playing for the Dallas Desperados of the AFL, hoping for a return to the NFL, but that doesn’t seem likely.

March 2006 (Revised April 2007)

For more information on the initial development of the Comprehensive Bust Rating (CBR), see Bustology 1 – Quarterbacks.

Development of the RB CBR

Development of the running back CBR began with surveying the list of running backs selected in the first round of the draft since 1970.  The list of first round running backs was split into two groups based on the year they were drafted.  Anyone drafted prior to 1994 was placed in the retired division, and anyone drafted 1994 to 2003 was placed in the active group.  This was because 1994 is the earliest draft in which a first round running back (Marshall Faulk) was still officially active in 2006.  Although first-round running backs drafted as recently as 2002 are no longer in the league, including prematurely retired (and therefore mostly bad) players to the retired list would skew the data.

Once the subjects had been divided into the two groups, the following raw statistics were assembled for each running back: 1) draft position, 2) number of games played, 3) career rushing yards, 4) career rushing attempts, 5) career touchdowns, and 6) total career yards (rushing and receiving).  These data were used to calculate: 1) yards per attempt, 2) touchdowns per attempt, and 3) total yards per game.  The mean and standard deviation were calculated for all the raw and calculated statistics listed above within the retired group of 105 running backs from Larry Stegent to Robert Smith.  Representative career means for a first round running back drafted between 1970 and 1993 are 91 games played, 3,892 rushing yards, 28 career touchdowns, 3.9 yards per attempt, and 51 total yards per game.  However, not all statistics are useful for determining the quality of a running back’s career, especially for determining the degree of “bustitude”.  For example, the difference in yards per attempt between a great running back (for example Barry Sanders at 5 yard per attempt) and a bad one (Curtis Ennis at 3.2 yards per attempt) is so small that the statistic is useless for determining running back quality.  Career touchdowns is another statistic that could not be used, in this case because the standard deviation is larger than the mean, with the result that in order to receive the maximum CBR, a running back would have to have scored negative touchdowns.  Statistics that are useful for determining running back quality are career rushing yards, touchdowns per attempt, and total yards (rushing and receiving) per game.  In addition, career length, in number of games played, is a good indicator of a running back’s success.  All four of these statistics were given approximately equal weight in calculating a raw CBR.  Then a final variable factor was added to the CBR for all offensive specialists.  This factor is based on the player’s draft position and increases on a logarithmic scale the higher in the draft he was selected.  This factor was added to account for the greater expectations that are accorded to high first round picks.  In the discussion below, this factor is referred to as the “rank bonus.”

For those who are interested, the running back CBR formula is: 100(((3,800-R)/20,000)+(((51.5-(Y/G))/100)+(20(0.0288-(T/A)))+((91-G)/200)+((1/P)/5)), where:

A = Attempts

G = Games

P = Draft position

R = Career Rushing Yards

T = Touchdowns

Y = Total Yards (rushing and receiving)

Anti-Bust Running Backs

Once the running back CBR formula was devised, it was applied to the running backs in both the retired and the active lists.  CBRs of less than –100 (the ultimate first-round anti-busts) were recorded for John Riggins (-115), Franco Harris (-117), Otis Armstrong (-126), Walter Payton (-169), Tony Dorsett (-113), Ottis Anderson (-103), Charles Alexander (-111), Marcus Allen (-157), Eric Dickerson (-121), Barry Sanders (-156), Emmitt Smith (-200), and Jerome Bettis (-120) from the retired list and Marshall Faulk (-140), Edgerrin James (-105), Shaun Alexander (-115), and LaDainian Tomlinson (-140) from the active list.  Riggins, Harris, Payton, Dorsett, Allen, Dickerson, and Sanders are already in the Hall of Fame, and others on this list are likely to follow.  The only Hall of Fame running back who did not have an CBR less than –100 was Earl Campbell, whose CBR was –67, including the 20-point maximum rank bonus.  Other CBRs that readers may find interesting for non-busts are –81 for Chuck Muncie, -74 for Gerald Riggs, -37 for Robert Smith, -85 for Eddie George, -73 for Warrick Dunn, –62 for Ricky Williams, and -85 for Larry Johnson.  Bo Jackson’s CBR was 16, due to the rank bonus for being the first overall pick and to the brevity of his career.

Lesser Busts

There are quite a few running backs whose names are mentioned when the subject turns to busts but don’t quite have the numbers to make a major splash in the world of bustitude.  Maybe you remember Alonzo Highsmith who was drafted third overall by the Oilers and gave them all of 7 touchdowns and a CBR of 67?  No?  How about D.J. Dozier from Penn State (CBR = 46)?  Or what about Blair Thomas, also of Penn State (CBR = 71), picked second overall by the Jets?

Some famous running back duds of recent years are not quite as high up on the scale as you might think.  Lawrence Phillips had a CBR of 38, Curtis Ennis scored an 80, and Ron Dayne currently holds a CBR of 25.

Mega Busts

Now that the lesser busts have been dismissed from consideration, the time has come to pay tribute to the giants of bustitude.  First the nominees:

1.                  Larry Stegent, 1970, a grand total of 12 yards for his career

2.                  Joe Moore, 1971, 23 games – zero TDs

3.                  Bill Thomas, 1972, averaged 1.5 yards per game

4.                  David Overstreet, 1981, one touchdown in 14 games

5.                  Booker Moore, 1981, scored one touchdown in 51 games

6.                  John Avery, 1998, 0.0134 touchdowns per rushing attempt

Now that the nominees are in, it’s time to examine each in order of ascending CBR.  Running back busts just aren’t what they used to be.  The most recent mega bust, John Avery had a CBR of 108, which pales in comparison with those who preceded him.  Future running backs will have to try harder if they hope to rival the busts who have gone before.  David Overstreet put forth a valiant effort after being selected 13^th overall by the Dolphins.  In his 14-game NFL career, he amassed a CBR of 111.  His draft classmate, Booker Moore (from Penn State, of course) had a longer career, but his lack of production earned him a rating of 113.

The contest is down to the final three and the suspense is palpable.  Which of these distinguished running backs will be crowned the greatest bust of his position since 1970?  First, the second runner up – Joe Moore (what is it about running backs named Moore?) who distinguished himself by producing not a single score in two years with the Bears and a CBR of 149. 

The Cowboys have had a history of drafting well, but in 1972, they pulled a boner by the name of Bill Thomas with the last pick of the first round.  This gentleman had no attempts for no yards and caught a grand total of zero passes in his one year with “America’s Team” before he was allowed to leave for Houston.  He had the most productive season of his career with the Oilers when he actually gained positive yardage and caught his single career pass.  He followed up this performance by playing an entire year for the Chiefs in 1974 and amassing a total of –3 yards.  Mercifully he and his 159 CBR were out of the league the following year.  Surprisingly, there was a first round pick running back whose career was even worse than that of Thomas, and that leads to the moment for which you have all been waiting – the crowning of the running back king of busts.  Not only was Larry Stegent the first running back drafted after the merger, but he was also the worst.  After being taken 8^th overall in the 1970 draft by the Cardinals, he did not see the field for the entire season.  In 1971, he had his single moment in the sun as an NFL player.  He caught a pass for 12 yards and then faded once more into obscurity.  For this performance, which is about as bad as possible for a running back, he achieved an impressive CBR of 171.  However, one supposes he should be grateful that three quarterbacks actually had worse CBRs than his.

Bustology – Quarterbacks

By Appaloosa

March 2006 (Revised March 2007)

Bustology – a Definition

Every year between the Combine and the Draft past draft prospects are remembered and compared with the current crop.  In retrospect some are assessed as successes and others as busts, but until now there has been no quantitative measure of “bustitude”.  A quantitative measure of relative quality allows players to be assessed objectively instead of subjectively as in the past.  The patented Comprehensive Bust Rating (CBR) system provides an objective method for comparing the relative bustitide of quarterbacks, running backs and wide receivers from 1970 to the present.

This article on quarterbacks is the first in a series that will determine the biggest bust in each of the three high-profile offensive specialist positions and ultimately crown the king of busts.  Further articles will also explore which universities have the best history of producing duds and which teams have drafted the most and biggest losers.

Introduction

The term “bust” is almost as overused as “superstar.”  Virtually any player who was drafted early in the first round who does not at least make the Pro Bowl seems to be labeled a bust.  But in many cases those who declare that a particular player is a bust may be basing their conclusion on exaggerated expectations, unfavorable comparison with a player from the same draft who had a better career, or a short memory.  The following series of articles is an attempt to make an unbiased assessment of busts unfiltered by the opinions of others or the passage of time.  Although there have been plenty of busts on defense and the less specialized offensive positions, only quarterbacks, running backs, and wide receivers amass sufficient numbers to allow their careers to be compared statistically.  Therefore the bust rating system is limited to offensive specialists.  In addition, only first round draft picks were assessed.  By definition a bust is someone of whom greatness is expected but who falls woefully short.  Greatness is generally only expected of first-round picks, so anyone picked in the second round or after was off the hook. 

Although the CBR can only be applied to offensive specialists, the construction of the ranking allows quarterbacks to be compared with running backs and running backs with wide receivers.  This is because the ratings, which are calculated from formulas specific to each position, are all based on the mean and standard deviation of four different statistics for first round draft picks whose careers ended prior to the start of the 2006 season.  All four statistics are given equal weight in calculating the CBR, which decreases the influence of any one factor and increases the comparability of players of different positions.  The CBR for a player whose career was average for a first round pick would be zero; a player whose statistics were a standard deviation worse than the average would have a CBR of 100, and a player whose statistics were a standard deviation better than the average would have a CBR of –100.  The formulas were devised based only on drafts in which all first round picks had completed their careers but they have been applied to active players who began their NFL careers as recently as 2003.  Although the CBR was invented to assess the degree of failure, a negative CBR correlates very well with success, and some significant examples have been included for comparison.

Busts by Reputation

The debate over the greatest bust in NFL draft history almost always seems to come down to a quarterback, which may not be fair.  Because the quarterback is the most visible member of a team and because quarterbacking a NFL team is the most difficult job in sports, quarterbacks are most commonly mentioned as spectacular busts.  For example, a few years ago the staff and the viewers of ESPN agreed that Ryan Leaf was the biggest bust of the last 25 years – not just in the NFL, but in all of sports.  There is no doubt that Leaf was a bust of epic proportions, but whether he was the biggest of all time is open to debate.  It is likely that part of the reason that Leaf is such an infamous dud is because: 1) he was drafted immediately after and was believed by some so-called experts to be a better NFL prospect than one of the biggest names in the game today, 2) he made himself unpopular with the media who took revenge by gleefully reporting every error he committed, and 3) Leaf’s actions are recent history, allowing earlier busts to be forgotten.  The CBR system cuts through all these mitigating factors and assesses each bust candidate solely on his merits (or lack thereof) on the field.

Development of the QB CBR

Development of the quarterback CBR began with surveying the list of quarterbacks selected in the first round of the draft since 1970.  The list of first round quarterbacks was split into two groups based on the year they were drafted.  Anyone drafted prior to 1993 was placed in the retired division, and anyone drafted 1993 to 2003 was placed in the active group.  This was because 1993 is the earliest draft in which a first round quarterback (Drew Bledsoe) amassed significant statistics in 2006.  Vinnie Testaverde was included in the retired list because he played very sparingly in 2006.  Although first-round quarterbacks drafted as recently as 1999 are no longer in the league, including prematurely retired (and therefore generally bad) players to the retired list would skew the data.

Once the subjects had been divided into the two groups, the following raw statistics were collected for each quarterback: 1) draft position, 2) number of games played, 3) career passing yards, 4) career attempts, 5) career completions, 6) career touchdown passes, and 7) career interceptions.  These data were used to calculate: 1) completion percentage, 2) yards per attempt, 3) touchdowns per attempt, 4) interceptions per attempt, and 5) passer rating.  Within the retired group of 40 quarterbacks from Terry Bradshaw to Tommy Maddox, the mean and standard deviation were calculated for all the statistics listed above.  Representative career means for a first round quarterback drafted between 1970 and 1992 are 103 games played, 17,640 passing yards, 53.6 percent completions, 105 career touchdowns, 101 career interceptions, and a career passer rating of 66.74.  However, not all of these statistics are useful for determining the quality of a quarterback’s career, especially for determining the degree of “bustitude”.  For example, raw interceptions are useless because a bad quarterback does not last long enough to throw a large number of picks compared to a quarterback who played for a normal-length career.  The interceptions per attempt statistic was not used because this value has dropped significantly since the 1970’s, and could not be used to compare quarterbacks from different eras.  As interceptions per attempt is one of the factors in passer rating, career passer rating was also deleted from formula used to calculate the CBR.  However, yards per attempt, touchdowns per attempt, and completion percentage were all good indicators of quarterback quality and have remained relatively constant among starting quarterbacks for the past 30 years.  In addition, career length, in number of games played, is a good indicator of a quarterback’s success.  All four of these statistics were given approximately equal weight in calculating a raw CBR.  Then a final variable factor was added to the CBR for all offensive specialists.  This factor is based on the player’s draft position and increases on a logarithmic scale the higher in the draft he was selected.  This factor was added to account for the greater expectations that are accorded to high first round picks.  In the discussion below, this factor is referred to as the “rank bonus.”

For those who are interested, the quarterback CBR formula is: 100((5(0.5354-(C/A)))+(((6.55-(Y/A))/2)+(20(0.036-(T/A)))+((105-G)/270)+((1/P)/5)), where:

A = Attempts

C = Completions

G = Games

P = Draft position

T = Touchdowns

Y = Yards

Anti-Bust Quarterbacks

Before exploring the quarterback busts, it might be interesting to calculate the CBR values for successful players to use to compare against the busts and to calibrate the forumla.  CBRs of less than –100 (the ultimate first-round anti-busts) were recorded for Jim Kelly (-124) and Dan Marino (-148) from the retired list and Peyton Manning (-145), and Daunte Culpepper (-130) from the active list.  Of those four, only Culpepper is not Hall of Fame caliber, and his low CBR is the result of exceptionally high completion percentage and yards per attempt, probably due to throwing to Randy Moss early in his career.  Other CBRs that readers may find interesting for non-busts are –73 for Terry Bradshaw, -75 for Phil Simms, -83 for John Elway, -61 for Troy Aikman, -73 for Steve McNair, and –47 for Donovan McNabb.  These CBRs include the rank bonus, so the CBRs of first overall picks are 20 points higher than the raw CBR.

Lesser Busts

Having dismissed the anti-busts from consideration, it is time to give out dishonorable mentions to some of the dogs in this show.  These individuals were noted busts in their day, but their stats were not execrable enough to make the Hall of Shame.  Please give a round of applause for Todd Blackledge who was drafted 20 spots ahead of Dan Marino and produced 29 touchdowns, 83 interceptions, and a CBR of 84.  Todd Marinovich wasn’t such a horrible quarterback by his stats.  After all, he completed more than 50% of his passes and threw only one more pick than touchdown pass, but due to drugs and trouble with the law, his career amounted to a grand total of 8 NFL games and a CBR of 42.  Andre Ware’s stats (37 CBR) were even a bit better than Marinovich’s in spite of his reputation as a classic bust.  But like another quarterback who is currently working his way up the bust scale, he had the misfortune to be drafted by the Lions.  Remember Rick Mirer?  Seven different teams thought they could recall the glory days of his college career, but the end result was a CBR of 76.  Heath Shuler was considered a favorite for best in show until he was surpassed by Ryan Leaf.  But in spite of completing less than 50% of his passes, Shuler’s CBR is a mere 92, which isn’t quite good enough for the pantheon of immortals of bustitude.  Tim Couch gets an occasional mention as a bust, but even with the maximum rank bonus that goes with being the first overall pick, his CBR is actually 3, which means that he was an average first-round quarterback.  His only real claim to bustitude is that he does not compare well with other quarterbacks picked first overall, although his CBR is still better than that of Michael Vick (14) or David Carr (16).

Mega Busts

Now for the stars of this whole composition – the busts of truly epic proportions.  First the nominees:

1.                  Jerry Tagge, 1972, averaged 0.0107 touchdown per attempt

2.                  Steve Pisarkeiwicz, 1977, completed 44.8% of his passes

3.                  Art Schlichter, 1982, averaged less than 5 yards per attempt

4.                  Kelly Stoufer, 1987, threw 7 touchdowns in 22 appearances

5.                  Dan McGuire, 1991, averaged 0.0136 touchdowns per attempt

6.                  Jim Druckenmiller, 1997, one (1!) career touchdown pass

7.                  Ryan Leaf, 1998, completed 48.4% of his passes

8.                  Akili Smith, 1999, a very impressive 0.0108 touchdowns per attempt

Now that the nominees are in, let us examine each in order of ascending CBR.  Ryan Leaf, thanks for stopping by.  We know you have the reputation, but a CBR of 140 just isn’t going to cut it in this crowd.  Kelly Stoufer, you can return to obscurity with a CBR of 143.  Ladies and gentlemen, we have a tie for fourth runner up.  Jerry Tagge and Steve Pisarkeiwicz, your efforts have earned you each a CBR of 154.  Commendable, but no cigar.  Nice try, Dan McGuire, but a rating of 168 isn’t even good enough for second runner up.  You can pick up your certificate on the way out the door.

Three quarterbacks remain, each distinguished in his own special way.  How can one decide between three such spectacular disappointments?  Each of these gentlemen averaged less than 5 yards an attempt, threw fewer than 0.02 touchdowns per attempt, and completed considerably less than 50% of his passes.  It requires a supreme effort to standout in such company and be crowned the king of the busts.  In the Olympic spirit (2008 being just around the corner), I award the bronze medal to Art Schlichter from THE Ohio State University whose all around lack of performance earned him a 200 CBR.  Our silver medallist is Akili Smith from the University of Oregon with a very impressive CBR of 208.  But the gold medal and the coveted title of greatest quarterback bust of all time belongs to none other than Jim Druckenmiller of Virginia Tech who overcame the handicap of being only the 26^th pick of the 1997 draft to achieve a phenomenal CBR of 232.  Only record lows for a first round quarterback in completion percentage and yards per attempt won out over nearly equally horrendous performances by higher draft picks.  “Druck” appeared in a total of 8 games for the 49ers and was last seen playing for the Memphis Maniax of the XFL.  One good bust deserves another.

Conference Call

Which Conferences Produce the Best QBs and RBs?

By Appaloosa

February 2006

            Sorting Quarterbacks and Running Backs by Conference

Certain schools and conferences have a reputation for producing high quality NFL players at specific positions.  The two positions that have the highest profile are quarterback and running back.  It seems that for quite a while the best quarterbacks have come from the PAC 10.  On the other side of the coin lately running backs from the Big Ten players have garnered a poor reputation.

            After a recent discussion on the relative merits of quarterbacks from the PAC 10 and the SEC, I decided to conduct a little study to determine which conferences have produced the highest quality quarterbacks and running backs since the AFL-NFL merger in 1970.  To compose my data set, I selected the quarterbacks and running backs (excluding full backs) from each year’s Pro Bowl roster, up to a maximum of six in each position for each year (1970 – 2005).  Where there were more than six at the position, I chose those who had been originally elected to the Pro Bowl and deleted injury replacements.  In other words, for this year I included Palmer and Brady rather than McNair and Green on the quarterbacks list.  This was done because I did not want more recent Pro Bowls with more players to skew the results, and the player originally elected to the position is likely to be considered of higher quality than an injury replacement.  Of course there were many players with multiple Pro Bowl appearances, which meant that they tended to count extra for their conference.  But as multiple Pro Bowl appearances are an extra measure of quality, I decided to count a player each time he was selected.

            Once the full list of Pro Bowl quarterbacks and running backs had been composed, I identified which school the player in question had played for prior to entering the NFL and tied that school to a conference.  In order to avoid the complication of the recent shifts in conference composition and the demise of the Southwestern Conference, I used the most recent conference memberships.  Therefore, Miami, Virginia Tech, and Boston College were included in the ACC rather than the Big East even though most of the current and past NFL players from those schools attended them while they were still part of the Big East.

            In addition to assessing the total number of Pro Bowl appearances for the entire 36 years from 1970 through 2005, I divided each group by decade to determine if there has been a shift over time in the conferences that have produced the best players at quarterback and running back.  Some of my findings were predictable but others were a little surprising.

            Best Conferences 1970 – 2005

            The following conferences produced Pro Bowl quarterbacks in significant numbers from 1970 through 2005: the six major conferences (ACC, Big East, Big Ten, Big Twelve, PAC 10, and SEC), Notre Dame, Conference USA, Mountain West, and Western Athletic Conference (WAC).  It is interesting to note that while the Mid-American Conference (MAC) has garnered a lot of attention lately for producing NFL quarterbacks (as in Chad Pennington, Byron Leftwich, and Ben Roethlesburger) no Pro Bowl quarterback to date has been from the MAC.

By far the most dominant conference for producing Pro Bowl quarterbacks is the PAC 10, which produced 49 Pro Bowl appearances out of a total of 193, or more than 25%.  The other major conferences ranged between 6 and 10 percent of the total except for the Big Twelve, which only produced three quarterback Pro Bowl appearances in 36 years, which is less than the appearances for Notre Dame (13), Conference USA (13), Mountain West (12), or even the WAC (4).

            The distribution of running backs was much more even among the major conferences.  Although the Big Twelve was the most frequent with 44 appearances out of a total of 215, or slightly over 20%, the SEC was only slightly lower, with 39 appearances (18%).  The ACC, Big Ten, and PAC 10 had totals in the 20’s, while the Big East lagged behind the rest of the majors with a total of 12.  The major conferences dominated the ranks of the Pro Bowl running backs slightly more than the ranks of the quarterbacks.  More than 69% of the running backs in the Pro Bowl attended major conference schools whereas the number of quarterbacks from major conferences was closer to 60%.  The Mountain West accounted for 12 Pro Bowl running backs, Notre Dame for 10, and Conference USA for 7.  The current WAC was completely shut out during the period of the study.

            Best Conferences by Decade

            During the decade 1970 – 1979, the quarterback position in the Pro Bowl was dominated by the SEC, with 12 appearances, followed by the Big Ten with 7.  During this era, the SEC was represented by such familiar names as Joe Namath, Fran Tarkenton, Ken Stabler, and Archie Manning.  However, with the close of the decade, the SEC appeared to drop off the map as a producer of quality quarterbacks.  It was not until Archie’s son, Peyton entered the league that the SEC was again represented at quarterback in the Pro Bowl.

            During the 1980’s and 1990’s the quarterback position was dominated by the PAC 10.  During the 80’s, the PAC produced 13 Pro Bowl quarterback appearances.  The next best major conference (ACC and Big East tie) had five.  The PAC 10 was even more dominating in the 1990’s with 25 Pro Bowl quarterbacks, while the ACC was runner up with 7.  However, since the start of the new century, the PAC 10 has been poorly represented in the Pro Bowl by quarterbacks.  The Big Ten currently holds the lead with six appearances by four different quarterbacks, followed by the SEC with five appearances, and the ACC and Big East are tied with four.  However, the totals for the SEC and the Big East are somewhat deceiving since both of these conferences have been represented by a single QB in the Pro Bowl since 2000.  The PAC 10 is falling behind with only two appearances, although those appearances were by two different quarterbacks.

            During the 1970’s the running back position was dominated by the Big Ten (Franco Harris accounted for most of these) and the Big Twelve, with 10 Pro Bowl appearances each.  With the 1980’s, the Big Ten ceased to be a major source of running backs while the number originating from the PAC 10 and SEC increased.  By the 1990’s, the running back position in the Pro Bowl began to be dominated by the SEC with 17 appearances, while the Big Twelve remained a steady producer of quality running backs with 13 Pro Bowl appearances.  In the early years of the 21^st Century, the Big Twelve and SEC are again among the leaders in producing Pro Bowl running backs, but the ACC has also accounted for 8 Pro Bowl appearances. 

In contrast to the Pro Bowl quarterbacks, who have been dominated by a few individuals, most of the major conferences have produced multiple Pro Bowl running backs.  For example, the SEC has been represented by six different running backs in Hawaii, and the ACC by four individuals.