How much do we tend to know after five weeks? Bill Connelly compares five-week data to full-season data to find out if we should be worried about TCU and Baylor.
19 Jan 2005
by Ned Macey
This weekend, Ben Roethlisberger will attempt to be the first rookie quarterback to lead his team to the Super Bowl. Only Shaun King in 1999 had ever before reached a conference championship game as a rookie quarterback. Roethlisberger was one of three quarterbacks taken in the top half of last year's first round, and each quarterback has seen a different amount of playing time this season. One was forced into the starting lineup almost immediately. One started only the second half of the year. One didn't throw a pass until the final game of the season.
While injuries, ineffectiveness, and other unexpected factors limit a coach's options, no common theory exists on the best way to groom a future quarterback. Some work has been done to project the future success of a quarterback based upon their rookie season statistics. But I haven't seen any work on the best way to develop a "franchise" quarterback. To fill this void, I went back to look at every quarterback drafted between 1993 and 2002, a ten year window that remains relatively modern and allows more recent quarterbacks to get some experience in the league. Roethlisberger's success notwithstanding, it seems clear to me that when an organization has to choose between playing a young quarterback sooner or later, they should err on the side of later.
From 1993-2002, 108 separate quarterbacks were drafted. Of these, 81 attempted a pass in the NFL. Of those 81, 55 attempted at least one pass their rookie season. 34 of those started at least one game as a rookie. The number of quarterbacks drafted in each round breaks down like this:
|Round 1: 19
Round 2: 9
Round 3: 10
Round 4: 17
|Round 5: 10
Round 6: 17
Round 7: 23
Round 8: 3
In this study, I want to examine the success of three groups, which we'll call the Big Ben, Eli, and Rivers groups. How successful are the careers of those quarterbacks that are trained with these three very different first year approaches?
The Big Ben group consists of those rookies who started at least 10 games or threw 250 passes. The Eli group includes those quarterbacks who started at least 5 games or threw 125 passes. The Rivers group is those quarterbacks who started fewer than 5 games and threw fewer than 125 passes.
The Big Ben group includes the following 12 players: Peyton Manning, Rick Mirer, David Carr, Chris Weinke, Tim Couch, Kerry Collins, Tony Banks, Drew Bledsoe, Joey Harrington, Charlie Batch, Heath Shuler, and Jake Plummer.
The Eli group includes the following 8 players: Ryan Leaf, Quincy Carter, Cade McNown, Donovan McNabb, Patrick Ramsey, Shaun King, Eric Zeier, and Akili Smith.
The Rivers group, obviously, includes everybody else, 88 players in all. Since of the above 20, only Eric Zeier and Chris Weinke were not chosen in the first or second round, I will list all first and second round players who would qualify.
There are 10 such players: Michael Vick, Trent Dilfer, Steve McNair, Jim Druckenmiller, Todd Collins, Chad Pennington, Daunte Culpepper, Drew Brees, Kordell Stewart, and Marques Tuiasosopo.
Here are some quick numbers comparing the success of the three groups:
On the surface, it immediately appears that the Rivers group has been the most successful, with the most Pro Bowl quarterbacks and the most quarterbacks of playoff teams. Let us look more closely at the success of each group individually:
While their playing time is roughly equal, this group can really be broken into two subgroups. Eight of the 12 were players picked in the top 5 of drafts, presumably by bad teams. Four players were drafted in the second through fourth rounds. Of the eight high first round picks, 3 have had successful careers, 3 have flopped, and 2 have had limited success but only just completed their third year. It would seem, as of today that David Carr may be successful and Joey Harrington will be out of a job, holding the success rate at about 50%. Carr, however, is far from a guarantee. This season was his best to date, and in Football Outsiders' DPAR rankings, he was the 18th best quarterback. Peyton Manning is clearly the best quarterback of any of these thirty. Drew Bledsoe has played in four Pro Bowls, while Kerry Collins has led two teams to at least the NFC Championship. Heath Shuler, Tim Couch, and Rick Mirer all had unsuccessful careers as a starter, although Mirer hangs on as a backup.
As for the other four, three failed to emerge, while 1 remains a competent starter in the NFL. Chris Weinke started fifteen games as a rookie, but only one since. Tony Banks and Charlie Batch both started for several years for mediocre teams before being replaced. They had careers that roughly mirrored the three flops from the first round group. They were starters for several seasons where they were not bad enough to be replaced and not good enough to win consistently. They remain backups today. Jake Plummer has had a similar career except he has continued to get opportunities. Over five seasons with the Cardinals, he never had a QB rating over 80. He has improved with the more talented Broncos but remains inconsistent. He is, however, in the top 50 all time in completions and has clearly had a successful career for a second round pick.
This group appears to be the least successful of the three. Donovan McNabb is a big time star, with five Pro Bowl appearances in five seasons as a starter. He is the second most successful quarterback of the 30. After McNabb, however, this group is a disaster. Ryan Leaf, Akili Smith, and Cade McNown are three of the four biggest quarterback busts of this ten year period. Only Heath Shuler can match their lack of accomplishment.
The jury is clearly still out on Patrick Ramsey, as to date he has not had enough of an opportunity to form an opinion. Eric Zeier was a third round pick who only played due to injuries to Vinny Testaverde. He showed some promise but failed to develop into a legitimate starter.
The final two Elis are Shaun King and Quincy Carter. Both were second round draft picks who were put in charge of teams in the middle of the season. King responded by leading his team to the NFC Championship game. It took several seasons, but Carter also led his team to the playoffs. Both quarterbacks were replaced after seasons in which they led their team to the playoffs. Both are now backup quarterbacks.
This group has been the most successful of the three. Six of the ten players were first round picks, and four were second round picks. Of the first round picks, five received legitimate opportunities to play. Two of these are amongst the top quarterbacks in football the last several years, Steve McNair and Daunte Culpepper. Two others, Chad Pennington and Michael Vick have already, in three full seasons as starters, led their teams to playoff wins and are back in the playoffs for the second time this year. The fifth is Trent Dilfer, who had a mixed career, making a Pro Bowl in 1997 and winning a Super Bowl in 2000. Overall, however, he has a career passer rating of 71, lower than such luminaries as Shaun King and Eric Zeier.
Two second round picks have gotten a legitimate chance to play. Kordell Stewart appeared in the Super Bowl in his rookie season as a receiver. He was a starter at quarterback for five inconsistent years with the Steelers, making the Pro Bowl and leading the Steelers to the AFC Championship in 2001 before losing the job for good in 2002. Drew Brees was groomed for the job and after a solid inaugural season as a starter appeared on his way out of San Diego. This year, he was named to the Pro Bowl. I think he is clearly somewhere in between, but he appears to be on his way to a solid NFL career.
Three players have not received extended opportunities to play. Jim Druckenmiller was a first-round pick in 1997, presumably to be Steve Young's eventual replacement. He played horribly in very limited action his rookie year and has never played again. He never developed and has now been cut or traded by three teams. Todd Collins was a second round pick drafted to replace Jim Kelly. He got a one year opportunity, struggled, and has been a backup ever since. Marques Tuiasosopo was drafted in the 2001 second round to be Rich Gannon's replacement. Last year, when Gannon went down, Tuiasosopo got only one start before injuring himself. He has only thrown 49 career passes.
A sample size of only thirty players is hardly proof of anything. The three best quarterbacks in football this year are arguably Peyton Manning, Donovan McNabb, and Daunte Culpepper. They were each brought along in a different way, and each has been amazingly successful. Each group includes a fair number of busts, from Heath Shuler to Akili Smith to Jim Druckenmiller. All quarterbacks are unique, and it is possible that no one way is the sure way to break in a quarterback.
That being said, in this time period, quarterbacks who learn on the sidelines in their first season have been wildly successful. Of the five quarterbacks taken in the first 20 picks of the draft who played sparingly in their first year, the worst player is Trent Dilfer, a Pro Bowler and Super Bowl Champion. Compare top 20 picks from the other two groups, and in the Big Ben group, after Manning and Bledsoe, none of the other six have had a career even as successful as Dilfer.
Most of the high picks were taken by bad teams, and when those teams are certain to miss the playoffs, conventional wisdom is that the highly paid player needs to get experience. A closer look indicates that a player who sits is as ready when he starts playing as someone who was getting experience is at the same point in his career.
Here is a chart comparing the first season of players drafted in the first round of the same year who were brought along at different paces. This comparison is of the player with experience's statistics in the year that the player who waited became the starter. For this comparison, we will include the stats to date of Kyle Boller, Byron Leftwich, and Carson Palmer, the three first-round quarterbacks from last year. The most successful player in any quarterback "class" is in bold.
|*Couch was injured for part of 2000, so these are 2001 stats,
his first full season where McNabb and Culpepper played.
A close evaluation of these statistics show that the only time a player who sat his first season with a lower quarterback rating in his inaugural season as a full time starter than another quarterback drafted in his class is this year, where Byron Leftwich seemingly outplayed Carson Palmer. Looking a little bit deeper thanks to DPAR, one could argue that Palmer had the more successful season, as he faced more difficult defenses and his DPAR of 39.0 surpassed Leftwich's 30.2 (not to mention David Carr's 27.5, Joey Harrington's 6.5, and Kyle Boller's 23.1).
In every other instance, the players who sat clearly hit the field at a higher level than the players in his class who played extensively their rookie season. Manning outplayed Leaf in Leaf's first full season as a starter, but Leaf was an "Eli" not a "Rivers.
As this sample size of thirty quarterbacks is relatively small, I think it would be beneficial to investigate an obvious reason that could be influencing the relative success of the Rivers group. As most Big Bens were top picks for really bad teams, and most Elis joined teams that were failing (the actual Eli being, of course, an exception), is it possible that the Rivers group is benefiting from playing with superior teams?
This other possible factor has been further emphasized by the success of Roethlisberger this year. Playing with a dominant running game, three very good receivers, and a great defense, he has had the most successful rookie season in over twenty years. Manning started for a team that was 5-4 at the time but was filled with defensive holes, a shaky offensive line, and average receiving core. It has become fashionable to attribute Roethlisberger's considerably superior rookie season to this factor.
In order to emphasize the impact on potential "franchise quarterbacks," I am going to list all quarterbacks drafted among the top 20 picks along with the team's record the year before they saw their first significant action, which I am defining as the same as the parameters for the Eli group, over 5 starts or 150 pass attempts. They are sorted from worst team to best.
This list clearly shows a trend towards quarterbacks taking over better teams having more successful careers. Of course, these are also the players who sat during their rookie season. The best record in the season before any full-time first year starter was the 4-12 posted by Washington before Heath Shuler took over in 1993. The four quarterbacks who took over the four best teams are the poster-children for the theory of sitting high draft picks.
With two competing theories working against each other, it is impossible to tell which factor is larger. This team quality theory, however, is far from perfect as it is unclear how much of a team's poor record is attributable to poor quarterback play. Some of the names of the starting quarterbacks the year before these quarterbacks took over include Hugh Millen, Stan Gelbaugh, and Bobby Hoying. In this limited sample size, no player took over a winning team and started immediately.
Over the past two seasons, two quarterbacks have taken over teams that were losers the year before but were filled with talent. Last year Kyle Boller took over a team only three years removed from the Super Bowl and two years removed from the playoffs. This year, Roethlisberger has taken over a team that was only three year's removed from an AFC title game appearance and two years from the playoffs. It is way too early to call Boller a bust or label Roethlisberger a success, but to date they have had very different results.
If you expand the list to those quarterbacks who were drafted later, the impact of a good team is obvious. Charlie Batch and Shaun King took over as full-time starters their rookie season for teams that were 10-6 and 11-5 respectively the year before. Those two had the highest quarterback ratings of any quarterback who threw more than 30 passes his rookie year. Both quarterbacks failed to develop from there, with career passer ratings well below their rookie numbers. Both players were admittedly second round picks. Even so, first year success was not a guarantee of career success.
A study of 30 players is not sufficient to determine the best way to develop a quarterback. Nonetheless, this limited evidence clearly shows that highly drafted quarterbacks that have waited through at least their rookie season have a much better track record than those who play their rookie seasons. While it is possible that a major factor in the success of those quarterbacks is the quality of the team when they became the full-time starter, it still seems inappropriate for a team to throw their high draft pick into the fire if they are not a good team.
All quarterbacks and teams are different, but I would be very hesitant to play my rookie quarterback before he or my team were ready. Between 1993 and 2002, of the twenty players who have made more than token appearances in their rookie season, only five or six have had what can be termed successful careers. Of the ten quarterbacks drafted in the first two rounds who sat their rookie season, seven have become solid NFL quarterbacks.
Next year with the success of Roethlisberger fresh in the mind of coaches everywhere a quarterback or two will be immediately put into the lineup for a bad team. When that happens, remember that the Lions will spend the offseason trying to find a quarterback to compete with Joey Harrington. Remember the Ravens who missed the playoffs in large part because of their quarterback. As Peyton Manning continues to write the record book, remember the Heath Shulers, Akili Smiths, and Tim Couchs who failed where he succeeded. You may want your favorite team to get its high pick on the field, but it may be best for him to wait until both he and the team are ready.
Indianapolis-raised, Haverford-educated Ed Macey currently lives in Australia, where he is supported by his lovely wife. As a Man of Leisure, he maintains the sports blog manofleisuresports.com. In preparation for attending University of Michigan Law School in the fall, he is devoting the next nine months to developing schemes to stop running quarterbacks. If you are interested in writing a guest column, something that takes a new angle on the NFL, please email us your idea at info-at-footballoutsiders.com.
6 comments, Last at 12 Dec 2007, 6:44am by best life ins