- Joined
- Aug 22, 2005
- Messages
- 2,213
- Reaction score
- 269
The Fins will very likely be taking a QB in the draft, which raises the age-old debate as to whether it is necessary or desirable to take a QB high in the draft. Many people believe that only an elite few can play the QB position well and that those without a certain pedigree or credentials have little chance of being a successful QB in the NFL. For these people, when there is a QB available in the first round that many feel has the potential to be a quality NFL QB you must take him. When guys like Tom Brady or Kurt Warner, who were passed over for several (or all) rounds in the draft, come along, these people dismiss them as rare and unique exceptions to these inherent truths.
I have come to believe that QB success has a lot less to do with the physical and mental traits of the QB than it does the circumstances in which the QB finds himself, e.g., scheme, offensive line, receivers, play-calling, defense (is he playing from behind all the time and/or always being asked to make something happen), etc. That’s not to say that you can plug your aunt or your 52-yr. old neighbor into the QB position on a good team and expect them to do well. But once you’ve narrowed the field to guys with the basic physical requisites, some major college starting QB experience, and the other things needed to get drafted or invited to training camp as an UDFA, I think the number of guys that can succeed as starting NFL QBs in favorable circumstances if given sufficient time to develop is much larger than most people think. In fact, as discussed in greater detail below, over the past 25 years, late round picks/undrafted free agents (5th –UDFA) have, as a group, posted a higher career QB rating than 1st round picks. They have also posted a better QB rating in their first year of significant playing time (100 or more pass attempts).
Those who believe you must take the QB in the first round often point to the relatively high number of starting QBs who were drafted in the first round. At any given time, that number is usually around 50-60% of all starting QBs. That does not mean that a 1st round QB will necessarily perform better than a late round pick, or even that he is much more likely to perform better.
The reason you see so many former 1st rounders among NFL starting QBs has a lot to do with the fact that late rounders and UDFAs are rarely, if ever, given a true opportunity to “win†a starting job. While a first rounder (especially a high first rounder) will almost always be given an opportunity to start and time to develop, teams and coaches lack the cojones to hand the reins over to a late round pick and stick with him if he doesn’t have immediate and dramatic success. Teams are forced to make a massive financial commitment to First Rounders and are almost forced to put them on the field. That is why complete abortions like Ryan Leaf, Akili Smith, Cade McNown, David Klingler, Tim Couch, Todd Marinovich, etc. all got a significant number of starts before their teams gave up on them. Having made that commitment, teams will give a 1st rounder every possible chance to show that he was not a bad pick. On the other hand, guys like Brady, Warner, Delhomme, Green, etc. were only given PT because the guys in front of them (most of whom were former 1st or 2nd round picks) got hurt.
I have been unable to think of a single QB drafted after the first round that became a starter by actually winning a starting job in training camp. Few coaches have the stones to elevate an unproven late round draft pick to starting QB over an incumbent, experienced former 1st rounder.
While First Rounders at other positions will also been given a bigger chance than late round guys, the difference is not nearly the same as at QB. Teams will start rookies Week 1 that are taken after the 1st round. I'm not sure any rookie QB taken after the first round has ever started week 1. Teams are also much more hesitant to bench a struggling QB than at any other position. Conventional wisdom is that you have to maintain consistency at QB even if the QB is struggling. Combine a team's need to justify its investment in a 1st round QB, with coaches' unwillingness to simply give a later round pick a shot to win the job, and the perceived need to stick with a struggling QB to maintain consistency of leadership and you have a pretty good explanation for why there are so many former No. 1s starting at QB.
In order for a late rounder to get a starting job and sufficient time to develop, it generally requires the guy in front of them to have a season ending (or near season ending) injury. Most of the notable late round QBs likely never would have gotten a real chance to show what they could do if the guy in front of them had not gone down:
Tom Brady (injury to Drew Bledsoe)
Kurt Warner (injury to Trent Green)
Trent Green (injury to Kurt Warner)
Marc Bulger (injury to Kurt Warner)
Jake Delhomme (injury to Rodney Peete)
Etc.
There is a category of late round QBs who showed enough in limited time on one team to get traded to another team that basically handed them a starting job – Brunell, Brooks, Hasselbeck, Fiedler, Feeley, Rob Johnson, Matt Schaub, etc. But none of these guys were able to break through and become a starter for the team that drafted them.
To evaluate the relative performance of QBs of varying draft status, I’ve compiled the stats of all QBs who entered the league since 1982. Although it is an imperfect measure, I am using QB rating as the principal basis for comparison because it is readily accessible, generally accepted, and does provide a useful snapshot of a QB’s performance. I have looked at these player’s performance over their entire careers, as well as in their first year of significant playing time, which I have defined as 100 or more pass attempts. I had to draw the line somewhere and 100 pass attempts equates to roughly 3-5 games worth of passing numbers, depending on the team. In my opinion, evaluating a QB on less than 100 pass attempts or so in the NFL has little or no value. The numbers discussed below represent composite QB ratings, i.e., the total numbers for all players of a particular draft stats plugged into the QB rating formula.
While the results were consistent with my thesis, I think they may be shocking to many.
Career QB Rating By Round
1st – 78.7
2nd – 77.8
3rd – 76.2
4th – 75.9
5th – 74.4
6th – 79.0
7th – 72.7
8th – 80.0
9th – 81.6
10th – 71.7
11th – 76.3
12th – N/A
UDFA – 83.4
The numbers above show very little difference in QB rating as you move down in the draft, with late round picks as a group putting up better numbers than early rounders. Because the numbers of players included in some draft rounds, especially the later ones, is so small, I think it is more useful to look at this from the standpoint of broader draft categories. To do this, I created 3 groupings: (1) 1st rounders; (2) mid-rounders (rounds 2-4); and (3) late rounders (round 5-UDFA).
Career QB Rating By Draft Area
1st Rounders – 78.7
Mid Rounders – 76.9
Late Rounders – 79.7
As indicated above, the numbers are very close, with the Late Rounders actually having the better career performance.
Because part of my theory is premised on the notion that late rounders rarely get an opportunity to play, I also looked at how these QBs performed in their first year of significant playing time (again, more than 100 attempts in a season).
1st Year of Playing Time QB Rating By Round
1st – 69.9
2nd – 69.9
3rd – 73.4
4th – 68.1
5th – 74.4
6th – 74.7
7th – 68.6
8th – 78.2
9th – 84.5
10th – 70.3
11th – 63.3
12th – N/A
UDFA – 89.4
Again, as a group, the later round picks performed better in their first significant NFL action. Below are the numbers based on broader draft area groupings:
1st Year of Playing Time QB Rating By Draft Area
1st Rounders – 69.7
Mid Rounders – 70.4
Late Rounders – 77.8
I suspected that the dramatic performance gap between Late Rounders and 1st Rounders in this first year of playing time was likely due to the fact that, while 1st Rounders are often thrown to the wolves in their first year or early in their second year, Late Rounders often have more time to watch and learn before getting their first experience. This was borne out by the numbers. The average 1st Rounder got his first significant playing time between his first and 2nd year. The average Mid Rounder got his first shot between his second and 3rd year, and the average Late Rounder got his first meaningful PT between his third and fourth year.
Of the 1st Rounders, 13 basically “bombed†in that first year of playing experience. Some of them, e.g., Elway, Aikman, Chris Miller, went on to become very good QBs, and the jury is still out on a few other, e.g., Eli Manning and Alex Smith. For this purpose, I have defined “bombed†as posting a QB rating below 60 in that first year. Nonetheless, all but one (Art Schlichter, whose career was cut short by a compulsive gambling habit and numerous stops in prison) of those 13 were given at least 400 pass attempts to show what they could do. Only 3 others were given less than 800 NFL pass attempts. During that same time period, there were only 7 guys who “bombed†in their first NFL opportunity. Of those, only 2 ended up getting 500 NFL pass attempts – Doug Flutie and Stan Humphries. Flutie got those attempts only after being banished to Canada for a decade after 5 years of little playing time in the NFL. Washington got rid of Humphries after that miserable first season and he became a starter again in Dan Diego a few years later.
QB play is dependent on a lot of factors beyond the control of the QB: (1) pass protection; (2) quality of the receivers; (3) effectiveness of the running game; (4) the actual style of the offense; (5) play-calling and (5) effectiveness of the defense, i.e., whether playing from beyond or with a lead, among many other things. It appears that these factors, much more so than, physical talent or draft pedigree determine a QB’s success.
To evaluate the impact of these external factors on QB performance, I have looked at how backup QBs have fared when they’ve come in to replace an injured starter. Generally speaking, the play of a backup QB usually depends on the quality of the offense (taking into account all of the foregoing factors) he comes into. Historically, backups to the great QBs in NFL history have put up numbers (and winning percentages) that were pretty comparable to the Hall of Famers and MVPs they were replacing. There are some exceptions, of course, but look at the QB ratings of the following backups, with the starter’s name in parentheses:
Steve Young (Montana) 108.9
Steve Bono (Montana) 88.5
Jeff Kemp (Montana) 85.7
Scott Mitchell (Marino) 91.4
Craig Erickson (Marino) 86.3
Elvis Grbac (Young) 87.9
Jeff Garcia (Young) 89.9
Frank Reich (Kelly) 102.3
Jim Sorgi (Manning) 99.1
Matt Cassel (Brady) 89.4
Marc Bulger (Warner) 101.5
Trent Green (Warner) 101.8
Bernie Kosar (Aikman) 92.7
Rodney Peete (Aikman) 102.5
Jason Garrett (Aikman) 83.3
Billy Volek (McNair) 87.4
Some of these guys, like Young and Bulger proved to be quality starters in their own right when given the chance. Even putting aside his struggles in Tampa early in his career, it is difficult to say how much of Young’s success was due to him and how much was due to the system in SF. After all, virtually every QB who played in SF from 1982 to 2000 had a QB rating in the mid-80s or higher. Bulger has done very well in St. Louis, but that is the only offense he has ever played in. Warner and Green had similar (or greater) success in that same offense before Bulger.
Because QB play is so dependent on so many other factors, it is hard to compare QB performance of players on different teams. Although all comparisons have their limitations, the best comparison is between two QBs who play in the same offense with the same teammates. Of course, the only time we see that is when one QB is hurt or otherwise replaced by another. I looked at the last 5 years and, by just eyeballing the numbers, saw a remarkable correspondence between the play of the starter and the play of the backup. To make this meaningful, I focused on those situations where the backup had at least 100 pass attempts. Again, although far from a perfect stat, QB rating is a well-accepted measure of QB performance, so I have focused on QB rating as a measuring stick.
For the purpose of this analysis, I have focused on those situations over the last 5 years where a guy drafted after the third round (or not drafted) came in to replace a starter and got at least 100 pass attempts. This does not include notable “unheralded†backups like Brady and Delhomme because the starters they replaced were hurt so early in the season that one cannot make a meaningful comparison. The starter and his rating is listed first, followed by the backup and his rating.
2006
Bledsoe (69.2)
Romo (95.1) (undrafted)
TGreen (74.1)
Huard (98.0) (undrafted)
Leftwich (79)
Garrard (80.5) (4th)
Hasselbeck (76.0)
SWallace (76.2) (4th)
McNabb (95.5)
Garcia (95.8) (undrafted)
2005
Boller (71.8)
AWright (71.7) (undrafted)
Losman (64.9)
Holcomb (85.6) (undrafted)
Pennington (70.9)
Bollinger (72.9) (6th)
Brooks (70)
Bouman (54.7) (undrafted)
Culpepper (72)
BJohnson (88.9) (9th)
Harrington (72)
Garcia (65.1) (undrafted)
McNabb (85)
McMahon (55.2) (5th)
Bulger (94.4)
JMartin (83.5) (undrafted)
Fitzpatrick (58.2) (7th)
2004
McNair (73.1)
Volek (87.1) (undrafted)
Fiedler (67.1)
Feeley (61.7) (5th)
Grossman (67.9)
Hutchinson (73.6) (undrafted)
Rattay (78.1)
Dorsey (62..4) (7th)
2003
Boller (62.4)
AWright (72.3) (undrafted)
Couch (77.6)
Holcomb (74.6) (undrafted)
Brees (67.5)
Flutie (82.8) (undrafted)
Ramsey (75.8)
THasselbeck (63.6) (undrafted)
KCollins (70.7)
JPalmer (58.5) (4th)
Garcia (80.1)
Rattay (96.6) (undrafted)
2002
Couch (76.8)
Holcomb (92.9) (undrafted)
Fiedler (85.2)
RLucas (69.9) (undrafted)
CChandler (79.8)
JMiller (77.5) (6th)
Harrington (59.9)
McMahon (52.4) (5th)
McNabb (86)
Feeley (75.4) (5th)
Warner (67.4)
Bulger (101.5) (6th)
JMartin (71.7) (undrafted)
Dilfer (71.1)
Hasselbeck (87.8) (6th)
2001
SMatthews (72.3)
JMiller (74.9) (6th)
RJohnson (76.3)
A Van Pelt (76.4) (8th)
In 18 of these 32 instances, the low draft pick backup had the higher rating. In many of those cases, however, the difference was pretty insignificant. By my count, taking into account the number of passes and the difference in the ratings, the low draft pick significantly outperformed the starter 10 times, the starter significantly outperformed the backup 7 times and the remaining 15 times the backup’s performance was essentially equivalent to the starter’s. Of the 10 times that the starter significantly outplayed the backup, several involved situations where even the starter had a rating below 80. I’m hesitant to say that a guy with a rating of 68 has significantly outplayed another guy with a rating of 58, especially when one of the guys has only played a few games. Several of the cases where the starter’s numbers were significantly better than the backup’s involved instances in which the backups numbers were skewed by a single bad performance, usually a 3-5 interception game, that he could not overcome based on a relatively small number of attempts.
Some people say that when a starter goes down, the rest of the team elevates its play, thus enabling the backup QBs numbers to look better than they are. Of course, there is no easy way to prove or disprove that. The bottom line is that the backup QB still has to find the open man and get him the ball. In those instances where the backup has kept the starting job for an extended period of time, e.g., Brady, Delhomme, Warner, Bulger, etc., it is highly doubtful that the rest of the players on the team are still elevating their play because the starter went down a few years earlier. I personally do not believe that the “rest of the team is elevating their play†explains why backup QBs so often perform as well as, or better than, the starter.
Now, I am certainly not suggesting that backups are generally better than starters or that all low draft pick QBs can be as successful as the top starters. The data does suggest, however, that who is playing QB may be a lot less important than most people assume. Most people believe that the top QBs, such as Manning, Palmer, Brady, etc. are essentially irreplaceable. That may or may not be true. Because none of these guys have missed significant playing time, we really don’t know how someone else might perform over an extended period of time in those offenses. Since these guys have only played for 1 team, we don’t really know how well these guys would perform in another offense. The same is true of many of the HOF QBs who played on the same team their entire careers.
While we have not seen enough of Cleo Lemon to know whether he will ever be a great QB, we can’t dismiss the possibility based on his draft position (or lack thereof). We also can’t dimiss it just because he has now been a backup for several years – the same was true of Delhomme, Hasselbeck, Gannon, etc. To date, while he has not set the world on fire, he has not embarrassed himself either. Considering that the Fins’ offense was not very good last year, regardless of who has been at QB, that is something of an accomplishment in itself. Moreover, it is reasonable to assume that, with more playing time, he will improve.
With respect to the draft, I believe the difference between the top rated QBs (Quinn and Russell) and the next tier (Beck, Stanton, Edwards, Kolb, Smith) is smaller than many people believe and that the difference between that tier and the guys who will go in the 5th round or later is also smaller than many people think. I believe that the success or failure of these QBs will depend a lot more on the circumstances in which they find themselves (i.e., offensive line, receivers, running game, play calling, etc.), than on the strength of their arm or their “pedigree.†Of the 8 first rounders since 1982 who have played well (QB rating above 80) in their first significant playing time, 6 did it on teams that had won at least 16 games over the two previous seasons – Marino (18 in 25 games), Roethlisberger (16), Pennington (19), Culpepper (25), Cutler (23) and Rivers (21). Of the other two, Jim Kelly had 2 years of starting QB experience in the USFL before getting his first NFL start, and Steve McNair did it in very limited playing time (143 attempts) on an 8-8 team that had gone 7-9 the previous season. Similarly, of the guys who have "bombed" in their first playing time, virtually all of them did it on absolutely terrible teams that had been absolutely terrible for a few years before that.
I expect that the success or failure of the guys picked this year will similarly depend on the quality of the teams that draft them.
I have come to believe that QB success has a lot less to do with the physical and mental traits of the QB than it does the circumstances in which the QB finds himself, e.g., scheme, offensive line, receivers, play-calling, defense (is he playing from behind all the time and/or always being asked to make something happen), etc. That’s not to say that you can plug your aunt or your 52-yr. old neighbor into the QB position on a good team and expect them to do well. But once you’ve narrowed the field to guys with the basic physical requisites, some major college starting QB experience, and the other things needed to get drafted or invited to training camp as an UDFA, I think the number of guys that can succeed as starting NFL QBs in favorable circumstances if given sufficient time to develop is much larger than most people think. In fact, as discussed in greater detail below, over the past 25 years, late round picks/undrafted free agents (5th –UDFA) have, as a group, posted a higher career QB rating than 1st round picks. They have also posted a better QB rating in their first year of significant playing time (100 or more pass attempts).
Those who believe you must take the QB in the first round often point to the relatively high number of starting QBs who were drafted in the first round. At any given time, that number is usually around 50-60% of all starting QBs. That does not mean that a 1st round QB will necessarily perform better than a late round pick, or even that he is much more likely to perform better.
The reason you see so many former 1st rounders among NFL starting QBs has a lot to do with the fact that late rounders and UDFAs are rarely, if ever, given a true opportunity to “win†a starting job. While a first rounder (especially a high first rounder) will almost always be given an opportunity to start and time to develop, teams and coaches lack the cojones to hand the reins over to a late round pick and stick with him if he doesn’t have immediate and dramatic success. Teams are forced to make a massive financial commitment to First Rounders and are almost forced to put them on the field. That is why complete abortions like Ryan Leaf, Akili Smith, Cade McNown, David Klingler, Tim Couch, Todd Marinovich, etc. all got a significant number of starts before their teams gave up on them. Having made that commitment, teams will give a 1st rounder every possible chance to show that he was not a bad pick. On the other hand, guys like Brady, Warner, Delhomme, Green, etc. were only given PT because the guys in front of them (most of whom were former 1st or 2nd round picks) got hurt.
I have been unable to think of a single QB drafted after the first round that became a starter by actually winning a starting job in training camp. Few coaches have the stones to elevate an unproven late round draft pick to starting QB over an incumbent, experienced former 1st rounder.
While First Rounders at other positions will also been given a bigger chance than late round guys, the difference is not nearly the same as at QB. Teams will start rookies Week 1 that are taken after the 1st round. I'm not sure any rookie QB taken after the first round has ever started week 1. Teams are also much more hesitant to bench a struggling QB than at any other position. Conventional wisdom is that you have to maintain consistency at QB even if the QB is struggling. Combine a team's need to justify its investment in a 1st round QB, with coaches' unwillingness to simply give a later round pick a shot to win the job, and the perceived need to stick with a struggling QB to maintain consistency of leadership and you have a pretty good explanation for why there are so many former No. 1s starting at QB.
In order for a late rounder to get a starting job and sufficient time to develop, it generally requires the guy in front of them to have a season ending (or near season ending) injury. Most of the notable late round QBs likely never would have gotten a real chance to show what they could do if the guy in front of them had not gone down:
Tom Brady (injury to Drew Bledsoe)
Kurt Warner (injury to Trent Green)
Trent Green (injury to Kurt Warner)
Marc Bulger (injury to Kurt Warner)
Jake Delhomme (injury to Rodney Peete)
Etc.
There is a category of late round QBs who showed enough in limited time on one team to get traded to another team that basically handed them a starting job – Brunell, Brooks, Hasselbeck, Fiedler, Feeley, Rob Johnson, Matt Schaub, etc. But none of these guys were able to break through and become a starter for the team that drafted them.
To evaluate the relative performance of QBs of varying draft status, I’ve compiled the stats of all QBs who entered the league since 1982. Although it is an imperfect measure, I am using QB rating as the principal basis for comparison because it is readily accessible, generally accepted, and does provide a useful snapshot of a QB’s performance. I have looked at these player’s performance over their entire careers, as well as in their first year of significant playing time, which I have defined as 100 or more pass attempts. I had to draw the line somewhere and 100 pass attempts equates to roughly 3-5 games worth of passing numbers, depending on the team. In my opinion, evaluating a QB on less than 100 pass attempts or so in the NFL has little or no value. The numbers discussed below represent composite QB ratings, i.e., the total numbers for all players of a particular draft stats plugged into the QB rating formula.
While the results were consistent with my thesis, I think they may be shocking to many.
Career QB Rating By Round
1st – 78.7
2nd – 77.8
3rd – 76.2
4th – 75.9
5th – 74.4
6th – 79.0
7th – 72.7
8th – 80.0
9th – 81.6
10th – 71.7
11th – 76.3
12th – N/A
UDFA – 83.4
The numbers above show very little difference in QB rating as you move down in the draft, with late round picks as a group putting up better numbers than early rounders. Because the numbers of players included in some draft rounds, especially the later ones, is so small, I think it is more useful to look at this from the standpoint of broader draft categories. To do this, I created 3 groupings: (1) 1st rounders; (2) mid-rounders (rounds 2-4); and (3) late rounders (round 5-UDFA).
Career QB Rating By Draft Area
1st Rounders – 78.7
Mid Rounders – 76.9
Late Rounders – 79.7
As indicated above, the numbers are very close, with the Late Rounders actually having the better career performance.
Because part of my theory is premised on the notion that late rounders rarely get an opportunity to play, I also looked at how these QBs performed in their first year of significant playing time (again, more than 100 attempts in a season).
1st Year of Playing Time QB Rating By Round
1st – 69.9
2nd – 69.9
3rd – 73.4
4th – 68.1
5th – 74.4
6th – 74.7
7th – 68.6
8th – 78.2
9th – 84.5
10th – 70.3
11th – 63.3
12th – N/A
UDFA – 89.4
Again, as a group, the later round picks performed better in their first significant NFL action. Below are the numbers based on broader draft area groupings:
1st Year of Playing Time QB Rating By Draft Area
1st Rounders – 69.7
Mid Rounders – 70.4
Late Rounders – 77.8
I suspected that the dramatic performance gap between Late Rounders and 1st Rounders in this first year of playing time was likely due to the fact that, while 1st Rounders are often thrown to the wolves in their first year or early in their second year, Late Rounders often have more time to watch and learn before getting their first experience. This was borne out by the numbers. The average 1st Rounder got his first significant playing time between his first and 2nd year. The average Mid Rounder got his first shot between his second and 3rd year, and the average Late Rounder got his first meaningful PT between his third and fourth year.
Of the 1st Rounders, 13 basically “bombed†in that first year of playing experience. Some of them, e.g., Elway, Aikman, Chris Miller, went on to become very good QBs, and the jury is still out on a few other, e.g., Eli Manning and Alex Smith. For this purpose, I have defined “bombed†as posting a QB rating below 60 in that first year. Nonetheless, all but one (Art Schlichter, whose career was cut short by a compulsive gambling habit and numerous stops in prison) of those 13 were given at least 400 pass attempts to show what they could do. Only 3 others were given less than 800 NFL pass attempts. During that same time period, there were only 7 guys who “bombed†in their first NFL opportunity. Of those, only 2 ended up getting 500 NFL pass attempts – Doug Flutie and Stan Humphries. Flutie got those attempts only after being banished to Canada for a decade after 5 years of little playing time in the NFL. Washington got rid of Humphries after that miserable first season and he became a starter again in Dan Diego a few years later.
QB play is dependent on a lot of factors beyond the control of the QB: (1) pass protection; (2) quality of the receivers; (3) effectiveness of the running game; (4) the actual style of the offense; (5) play-calling and (5) effectiveness of the defense, i.e., whether playing from beyond or with a lead, among many other things. It appears that these factors, much more so than, physical talent or draft pedigree determine a QB’s success.
To evaluate the impact of these external factors on QB performance, I have looked at how backup QBs have fared when they’ve come in to replace an injured starter. Generally speaking, the play of a backup QB usually depends on the quality of the offense (taking into account all of the foregoing factors) he comes into. Historically, backups to the great QBs in NFL history have put up numbers (and winning percentages) that were pretty comparable to the Hall of Famers and MVPs they were replacing. There are some exceptions, of course, but look at the QB ratings of the following backups, with the starter’s name in parentheses:
Steve Young (Montana) 108.9
Steve Bono (Montana) 88.5
Jeff Kemp (Montana) 85.7
Scott Mitchell (Marino) 91.4
Craig Erickson (Marino) 86.3
Elvis Grbac (Young) 87.9
Jeff Garcia (Young) 89.9
Frank Reich (Kelly) 102.3
Jim Sorgi (Manning) 99.1
Matt Cassel (Brady) 89.4
Marc Bulger (Warner) 101.5
Trent Green (Warner) 101.8
Bernie Kosar (Aikman) 92.7
Rodney Peete (Aikman) 102.5
Jason Garrett (Aikman) 83.3
Billy Volek (McNair) 87.4
Some of these guys, like Young and Bulger proved to be quality starters in their own right when given the chance. Even putting aside his struggles in Tampa early in his career, it is difficult to say how much of Young’s success was due to him and how much was due to the system in SF. After all, virtually every QB who played in SF from 1982 to 2000 had a QB rating in the mid-80s or higher. Bulger has done very well in St. Louis, but that is the only offense he has ever played in. Warner and Green had similar (or greater) success in that same offense before Bulger.
Because QB play is so dependent on so many other factors, it is hard to compare QB performance of players on different teams. Although all comparisons have their limitations, the best comparison is between two QBs who play in the same offense with the same teammates. Of course, the only time we see that is when one QB is hurt or otherwise replaced by another. I looked at the last 5 years and, by just eyeballing the numbers, saw a remarkable correspondence between the play of the starter and the play of the backup. To make this meaningful, I focused on those situations where the backup had at least 100 pass attempts. Again, although far from a perfect stat, QB rating is a well-accepted measure of QB performance, so I have focused on QB rating as a measuring stick.
For the purpose of this analysis, I have focused on those situations over the last 5 years where a guy drafted after the third round (or not drafted) came in to replace a starter and got at least 100 pass attempts. This does not include notable “unheralded†backups like Brady and Delhomme because the starters they replaced were hurt so early in the season that one cannot make a meaningful comparison. The starter and his rating is listed first, followed by the backup and his rating.
2006
Bledsoe (69.2)
Romo (95.1) (undrafted)
TGreen (74.1)
Huard (98.0) (undrafted)
Leftwich (79)
Garrard (80.5) (4th)
Hasselbeck (76.0)
SWallace (76.2) (4th)
McNabb (95.5)
Garcia (95.8) (undrafted)
2005
Boller (71.8)
AWright (71.7) (undrafted)
Losman (64.9)
Holcomb (85.6) (undrafted)
Pennington (70.9)
Bollinger (72.9) (6th)
Brooks (70)
Bouman (54.7) (undrafted)
Culpepper (72)
BJohnson (88.9) (9th)
Harrington (72)
Garcia (65.1) (undrafted)
McNabb (85)
McMahon (55.2) (5th)
Bulger (94.4)
JMartin (83.5) (undrafted)
Fitzpatrick (58.2) (7th)
2004
McNair (73.1)
Volek (87.1) (undrafted)
Fiedler (67.1)
Feeley (61.7) (5th)
Grossman (67.9)
Hutchinson (73.6) (undrafted)
Rattay (78.1)
Dorsey (62..4) (7th)
2003
Boller (62.4)
AWright (72.3) (undrafted)
Couch (77.6)
Holcomb (74.6) (undrafted)
Brees (67.5)
Flutie (82.8) (undrafted)
Ramsey (75.8)
THasselbeck (63.6) (undrafted)
KCollins (70.7)
JPalmer (58.5) (4th)
Garcia (80.1)
Rattay (96.6) (undrafted)
2002
Couch (76.8)
Holcomb (92.9) (undrafted)
Fiedler (85.2)
RLucas (69.9) (undrafted)
CChandler (79.8)
JMiller (77.5) (6th)
Harrington (59.9)
McMahon (52.4) (5th)
McNabb (86)
Feeley (75.4) (5th)
Warner (67.4)
Bulger (101.5) (6th)
JMartin (71.7) (undrafted)
Dilfer (71.1)
Hasselbeck (87.8) (6th)
2001
SMatthews (72.3)
JMiller (74.9) (6th)
RJohnson (76.3)
A Van Pelt (76.4) (8th)
In 18 of these 32 instances, the low draft pick backup had the higher rating. In many of those cases, however, the difference was pretty insignificant. By my count, taking into account the number of passes and the difference in the ratings, the low draft pick significantly outperformed the starter 10 times, the starter significantly outperformed the backup 7 times and the remaining 15 times the backup’s performance was essentially equivalent to the starter’s. Of the 10 times that the starter significantly outplayed the backup, several involved situations where even the starter had a rating below 80. I’m hesitant to say that a guy with a rating of 68 has significantly outplayed another guy with a rating of 58, especially when one of the guys has only played a few games. Several of the cases where the starter’s numbers were significantly better than the backup’s involved instances in which the backups numbers were skewed by a single bad performance, usually a 3-5 interception game, that he could not overcome based on a relatively small number of attempts.
Some people say that when a starter goes down, the rest of the team elevates its play, thus enabling the backup QBs numbers to look better than they are. Of course, there is no easy way to prove or disprove that. The bottom line is that the backup QB still has to find the open man and get him the ball. In those instances where the backup has kept the starting job for an extended period of time, e.g., Brady, Delhomme, Warner, Bulger, etc., it is highly doubtful that the rest of the players on the team are still elevating their play because the starter went down a few years earlier. I personally do not believe that the “rest of the team is elevating their play†explains why backup QBs so often perform as well as, or better than, the starter.
Now, I am certainly not suggesting that backups are generally better than starters or that all low draft pick QBs can be as successful as the top starters. The data does suggest, however, that who is playing QB may be a lot less important than most people assume. Most people believe that the top QBs, such as Manning, Palmer, Brady, etc. are essentially irreplaceable. That may or may not be true. Because none of these guys have missed significant playing time, we really don’t know how someone else might perform over an extended period of time in those offenses. Since these guys have only played for 1 team, we don’t really know how well these guys would perform in another offense. The same is true of many of the HOF QBs who played on the same team their entire careers.
While we have not seen enough of Cleo Lemon to know whether he will ever be a great QB, we can’t dismiss the possibility based on his draft position (or lack thereof). We also can’t dimiss it just because he has now been a backup for several years – the same was true of Delhomme, Hasselbeck, Gannon, etc. To date, while he has not set the world on fire, he has not embarrassed himself either. Considering that the Fins’ offense was not very good last year, regardless of who has been at QB, that is something of an accomplishment in itself. Moreover, it is reasonable to assume that, with more playing time, he will improve.
With respect to the draft, I believe the difference between the top rated QBs (Quinn and Russell) and the next tier (Beck, Stanton, Edwards, Kolb, Smith) is smaller than many people believe and that the difference between that tier and the guys who will go in the 5th round or later is also smaller than many people think. I believe that the success or failure of these QBs will depend a lot more on the circumstances in which they find themselves (i.e., offensive line, receivers, running game, play calling, etc.), than on the strength of their arm or their “pedigree.†Of the 8 first rounders since 1982 who have played well (QB rating above 80) in their first significant playing time, 6 did it on teams that had won at least 16 games over the two previous seasons – Marino (18 in 25 games), Roethlisberger (16), Pennington (19), Culpepper (25), Cutler (23) and Rivers (21). Of the other two, Jim Kelly had 2 years of starting QB experience in the USFL before getting his first NFL start, and Steve McNair did it in very limited playing time (143 attempts) on an 8-8 team that had gone 7-9 the previous season. Similarly, of the guys who have "bombed" in their first playing time, virtually all of them did it on absolutely terrible teams that had been absolutely terrible for a few years before that.
I expect that the success or failure of the guys picked this year will similarly depend on the quality of the teams that draft them.
