YoHoChecko
04-15-2007, 05:26 PM
Stats Behind the Stats:
Start ‘em or Sit ‘em?
Investigating the myths behind rookie quarterbacks
By
Max Gross
Every time a quarterback is taken in the first round of the draft, a debate takes place as to whether that quarterback should play right away or sit for a year to learn the position on an NFL level. While common sense suggests that playing a rookie quarterback is bad for the team in the short-term, there is a school of thought among many football people that the only way to truly learn the position is to play, and that the on-field experience is invaluable to the quarterback. This thinking suggests that the short-term struggles of the team and quarterback are worth enduring in the name of long-term benefits in development, and thus more wins for the franchise. An assumption is also made within this line of thinking that any first-time starter at quarterback, even after a year or two on the bench, will struggle similarly to rookie quarterbacks. Therefore, leaving a player on the bench is only delaying a year of poor play and holding back the franchise, overall.
The contrasting side of the debate claims that a quarterback is better prepared to play in the league after sitting on the bench for at least a season, and thus will not experience “rookie struggles.” This line of thinking asks why a team should endure the slow learning curve of a quarterback on the field, usually resulting in a poor season, if the benefits of doing so are negligible, or maybe even non-existent.
By looking at first round quarterbacks taken in the drafts from 1994 to 2004 and how they played in their first three years, this debate can finally be settled. All quarterbacks will be included in the study except for Jim Druckenmiller and Rex Grossman, neither of whom had enough playing time in their first three seasons to measure development or impact on their teams, leaving 23 quarterbacks in the study. The players will be broken down into three groups: those who were given the role of starter from the outset, those that got some significant starting experience during their rookie seasons and those that played only in spot back-up duty their rookie seasons, and thus gained little on-field experience.
The 10 quarterbacks that were immediate starters are Heath Shuler, Kerry Collins, Peyton Manning, Ryan Leaf, Tim Couch, David Carr, Joey Harrington, Byron Leftwich, Kyle Boller and Ben Roethlisberger. Those that had some significant starting experience were Akili Smith, Cade McNown, Donovan McNabb, Patrick Ramsey and Eli Manning. Finally, the following eight players spent most of their first year (or more) on the bench: Trent Dilfer, Steve McNair, Daunte Culpepper, Chad Pennington, Michael Vick, Carson Palmer, Philip Rivers and JP Losman.
Year One
In the first year after drafting quarterbacks in the first round, the rookie starters’ average stat line looked like this:
<table><tr><td>ATT</td><td>COMP</td><td> % </td><td>YARDS</td><td>YPA</td><td>TD</td><td>INT</td><td>RATE</td></tr>
<tr><td>199</td><td> 373 </td><td> 53.5% </td><td> 2344 </td><td> 6.29 </td><td> 12 </td><td> 14 </td><td> 67.3</td></tr></table>
It should also be noted that out of the eight non-expansion teams that put a rookie into immediate starting duty, three decreased their win total from the previous season and only two improved their win total by more than one game. In all, the ten franchises that took this route averaged six wins in year one.
Similarly, the five quarterbacks that started the year behind a veteran, but at some point during the season were inserted into the starting line-up had an average stat line as follows:
<table><tr><td>ATT</td><td>COMP</td><td> % </td><td>YARDS</td><td>YPA</td><td>TD</td><td>INT</td><td>RATE</td></tr>
<tr><td>105 </td><td> 206 </td><td> 51.1% </td><td> 1160 </td><td> 5.64 </td><td> 7 </td><td> 8 </td><td> 62.6</td></tr></table>
These five teams had only one team with declining wins from the previous season, but none improved by more than two games. On average, these five teams won six games—an improvement of one game over the previous year’s average.
Meanwhile, the leading passers for the eight teams that kept their first-round picks on the bench came out with the following average stat line:
<table><tr><td>ATT</td><td>COMP</td><td> % </td><td>YARDS</td><td>YPA</td><td>TD</td><td>INT</td><td>RATE</td></tr>
<tr><td>226 </td><td> 379 </td><td> 59.7% </td><td> 2717 </td><td> 7.17 </td><td> 18 </td><td> 12 </td><td> 84.6</td></tr></table>
Of these eight teams, only one had a decline in wins from their previous seasons (The Vikings, coming off of a 15-win year). Five of the eight improved by more than one game, and the average total wins for these teams was eight games.
This information affirms the common sensical notion that playing a rookie quarterback significantly decreases the production from that position compared to the play of veterans. Furthermore, it is more likely to cause a decline in wins and incredibly rarely leads to immediate success for the team.
Year Two
While the data from year one was hardly groundbreaking, data from year two should be the most telling. This information will allow a comparison between the play of a first-year starter who has a year on the bench to that of a rookie, while also checking the merits of the idea that a year of playing experience puts a quarterback farther along in his development.
In their second year as full-time starters, the ten quarterbacks who played immediately showed improvement in their stat lines:
<table><tr><td>ATT</td><td>COMP</td><td> % </td><td>YARDS</td><td>YPA</td><td>TD</td><td>INT</td><td>RATE</td></tr>
<tr><td>203 </td><td> 352 </td><td> 57.8% </td><td> 2422 </td><td> 6.89 </td><td> 13 </td><td> 12 </td><td> 78.1</td></tr></table>
Only two of the teams saw their win totals decline, while half of them (five) saw an improvement of more than one win. Because many teams had a decline in wins the previous season, these improvements led only to an average of eight wins apiece. It should be noted that Manning’s Colts improved from three to 13 wins in year two.
The five players who had some significant starting experience as rookies were stepping into the role of full-time starters for the first time. However, their partial on-field training did not lead to the same level of improvement, but still increased their production:
<table><tr><td>ATT</td><td>COMP</td><td> % </td><td>YARDS</td><td>YPA</td><td>TD</td><td>INT</td><td>RATE</td></tr>
<tr><td>215 </td><td> 402 </td><td> 53.5% </td><td> 2438 </td><td> 6.07 </td><td> 14 </td><td> 11 </td><td> 72.3</td></tr></table>
Their teams, likewise, did not show the same level of improvement, as two of the five teams declined, two showed significant improvement and one team stood pat. In all, they averaged seven wins.
Meanwhile, six of the eight first-rounders who sat on the bench in year one took to the field as starters in their first significant NFL action (though one of the six didn’t take over right away—Steve McNair, and one didn’t remain the team’s full-time starter—JP Losman). These six players produced as follows:
<table><tr><td>ATT</td><td>COMP</td><td> % </td><td>YARDS</td><td>YPA</td><td>TD</td><td>INT</td><td>RATE</td></tr>
<tr><td>203 </td><td> 352 </td><td> 57.5% </td><td> 2514 </td><td> 7.14 </td><td> 14 </td><td> 12 </td><td> 79.0</td></tr></table>
Only one of the six teams had a decline in wins (the Bills with Losman), and the group averaged eight wins. So not only did the teams not suffer the same dips that rookie quarterbacks provided in terms of team success, but the quarterback performance was far better than the numbers posted by rookie starters. In fact, the players who sat the bench for year one posted slightly better year two numbers than either of the quarterback groups with on-field experience.
The Jets and Chargers chose to keep Pennington and Rivers on the bench for an additional year with Vinny Testaverde and Drew Brees at the helms, respectively. Those teams were getting good play from the quarterback position and were winning games (they averaged 10 wins that year), so no change was made.
The data from year two begins to debunk the myth of the value of on-field experience. Players who saw the field as starters for the first time in their second years showed no evidence of being behind in their development, and slight evidence that it may have been even more beneficial than a year on the field. When combining the two groups that saw significant playing time in year one and comparing those totals with players that stayed on the sidelines, those who sat for a year had more completions and attempts, higher completion percentage, more yards, more touchdowns and a higher yards per attempt. Meanwhile, both groups averaged the same win total on year two, showing that either way can be successful for the team.
Year Three
By year three, all players involved in the study will have seen the field for significant playing time, and the long-term effects of the decisions made about first-round quarterbacks can begin to be gauged. However, it does get a little complicated, because in year three, some teams begin to give up on their prospects. For instance, Heath Shuler, Ryan Leaf, Akili Smith and Cade McNown had worn out their welcomes, and had limited playing time—if any. For that reason, their statistics will be excluded from averages, but their failures need to be noted. Additionally, the injury bug bites the study for the first time, as Michael Vick lost most of this season to a broken leg, and thus his statistics will also be removed from averages. Just because these players will be excluded from further statistical observations, it is necessary to remain mindful of their plights, as removing players removes the lowest performers, possibly increasing the overall average.
The eight remaining starters who had been starting since day one showed very marginal improvement from year two. With this much starting experience, it makes sense that their play may have leveled off by this time. Here’s how they fared in year three:
<table><tr><td>ATT</td><td>COMP</td><td> % </td><td>YARDS</td><td>YPA</td><td>TD</td><td>INT</td><td>RATE</td></tr>
<tr><td>252 </td><td> 428 </td><td> 58.8% </td><td> 2949 </td><td> 6.89 </td><td> 18 </td><td> 15 </td><td> 78.4</td></tr></table>
The total numbers increased, but the completion percentage and yards per attempt showed minimal change, if any. The teams all still averaged eight wins apiece, showing the numbers to be leveling off normally around what should be the league’s averages. The same could be said for the three remaining players who had some playing time as rookies, but with only three, the sample is too small to draw many conclusions. That being said, here’s how the remaining players performed in year three:
<table><tr><td>ATT</td><td>COMP</td><td> % </td><td>YARDS</td><td>YPA</td><td>TD</td><td>INT</td><td>RATE</td></tr>
<tr><td>252 </td><td> 429 </td><td> 58.7 </td><td> 2714 </td><td> 6.33 </td><td> 20 </td><td> 14 </td><td> 79.3</td></tr></table>
These numbers are very similar to the first-year full-time starters, and their teams also averaged eight wins.
The five players that had been held out of their first year but started in their second years continued to improve, albeit somewhat marginally. They finished year three with the following average stat line:
<table><tr><td>ATT</td><td>COMP</td><td> % </td><td>YARDS</td><td>YPA</td><td>TD</td><td>INT</td><td>RATE</td></tr>
<tr><td>266 </td><td> 440 </td><td> 60.5% </td><td> 3005 </td><td> 6.83 </td><td> 18 </td><td> 14 </td><td> 81.3</td></tr></table>
This group offers the first 3,000 yard, 60 percent, 80+ QB rating of the study thus far, aside from the veteran starters in year one. However, the win average dropped slightly to seven.
There is only one remaining group, and it contains only two players. Again, this is a small sample, but stats cannot be ignored just for that reason. Starting for the first time as third-year players, Chad Pennington and Philip Rivers averaged the following stat line:
<table><tr><td>ATT</td><td>COMP</td><td> % </td><td>YARDS</td><td>YPA</td><td>TD</td><td>INT</td><td>RATE</td></tr>
<tr><td>280 </td><td> 430 </td><td> 65.2% </td><td> 3254 </td><td> 7.58 </td><td> 22 </td><td> 8 </td><td> 97.8</td></tr></table>
These two teams also averaged 12 wins. Naturally, the Chargers and Jets had much more going for them than their strong quarterback play, but the fact that these players performed so well in their first year as starters certainly goes to debunk the idea that on-field experience is the most important aspect of developing a quarterback. Now in the third year, it might seem appropriate to condense the study down to two groups: those with significant first-year playing experience, and those without. In such an analysis, you get the following results:
<table><tr><td></td><td>ATT</td><td>COMP</td><td> % </td><td>YARDS</td><td>YPA</td><td>TD</td><td>INT</td><td>RATE</td><td>WINS</td></tr>
<tr><td>With First-year:</td><td> 252 </td><td> 428 </td><td> 58.9% </td><td> 2885 </td><td>6.73</td><td>18 </td><td> 15 </td><td> 78.7 </td><td> 8</td></tr>
<tr><td>W/O First-year: </td><td> 236 </td><td> 383 </td><td> 61.6% </td><td> 2691 </td><td>7.03</td><td> 17 </td><td> 11 </td><td> 85.5 </td><td> 8</td></tr></table>
Again, a reminder that the group of first-year starters is absent four “draft busts,” and the group without first year experience is absent the pass-challenged Michael Vick. Those omissions are likely to have artificially inflated both of these averages a little, but especially the first-year starters. While that is not enough evidence to definitively claim that sitting for a year (or two) actually enhances the development of a quarterback moreso than getting onto the field, it is certainly enough to disprove the converse. There is absolutely no evidence that backs up the notion that a quarterback is better off long-term if he gets onto the field quickly to “take his lumps.” There is also no evidence that suggests that the teams that make that decision fare better in the long-term, as the winning averages even out around year two and stay approximately even. In fact, the only apparent advantage of playing a quarterback right away is that if it is not working out, teams are able to cut the ties by year three, possibly earlier than those quarterbacks who have seen less time on the field. Of course, that also brings up the point that the number of “draft busts” are significantly higher among quarterbacks forced onto the field during their first season, lending circumstantial evidence to the notion that playing in year one might actually decrease the chances for success among quarterbacks.
Additionally, with no evidence that any long-term benefits exist for playing rookie quarterbacks, it seems a good time to re-state the evidence from year one showing that rookie starters fare far worse than veterans, and often lead to a one-year decline for the team. This begs the question as to why a team would sacrifice some measure of success one season if it bears no long-term benefits. Indeed, it could be said to be a fool-hearty decision, unless the team is hoping for an improved draft pick the next year without the appearance of “tanking” a season.
Finally, it should be noted that the two quarterbacks who sat for two seasons combined to show better production (by QB rating) than all but four quarterbacks in the study showed in any single year (Roethlisberger, P. Manning, Palmer and Culpepper). Such a fact combined with the other data provided in this analysis suggests that each year of development on the bench may be worth slightly more than a year of development on the field, and indicates that the best chance to ensure that there is never a drop-off in team or quarterback performance is to let early picks at quarterback sit.
Start ‘em or Sit ‘em?
Investigating the myths behind rookie quarterbacks
By
Max Gross
Every time a quarterback is taken in the first round of the draft, a debate takes place as to whether that quarterback should play right away or sit for a year to learn the position on an NFL level. While common sense suggests that playing a rookie quarterback is bad for the team in the short-term, there is a school of thought among many football people that the only way to truly learn the position is to play, and that the on-field experience is invaluable to the quarterback. This thinking suggests that the short-term struggles of the team and quarterback are worth enduring in the name of long-term benefits in development, and thus more wins for the franchise. An assumption is also made within this line of thinking that any first-time starter at quarterback, even after a year or two on the bench, will struggle similarly to rookie quarterbacks. Therefore, leaving a player on the bench is only delaying a year of poor play and holding back the franchise, overall.
The contrasting side of the debate claims that a quarterback is better prepared to play in the league after sitting on the bench for at least a season, and thus will not experience “rookie struggles.” This line of thinking asks why a team should endure the slow learning curve of a quarterback on the field, usually resulting in a poor season, if the benefits of doing so are negligible, or maybe even non-existent.
By looking at first round quarterbacks taken in the drafts from 1994 to 2004 and how they played in their first three years, this debate can finally be settled. All quarterbacks will be included in the study except for Jim Druckenmiller and Rex Grossman, neither of whom had enough playing time in their first three seasons to measure development or impact on their teams, leaving 23 quarterbacks in the study. The players will be broken down into three groups: those who were given the role of starter from the outset, those that got some significant starting experience during their rookie seasons and those that played only in spot back-up duty their rookie seasons, and thus gained little on-field experience.
The 10 quarterbacks that were immediate starters are Heath Shuler, Kerry Collins, Peyton Manning, Ryan Leaf, Tim Couch, David Carr, Joey Harrington, Byron Leftwich, Kyle Boller and Ben Roethlisberger. Those that had some significant starting experience were Akili Smith, Cade McNown, Donovan McNabb, Patrick Ramsey and Eli Manning. Finally, the following eight players spent most of their first year (or more) on the bench: Trent Dilfer, Steve McNair, Daunte Culpepper, Chad Pennington, Michael Vick, Carson Palmer, Philip Rivers and JP Losman.
Year One
In the first year after drafting quarterbacks in the first round, the rookie starters’ average stat line looked like this:
<table><tr><td>ATT</td><td>COMP</td><td> % </td><td>YARDS</td><td>YPA</td><td>TD</td><td>INT</td><td>RATE</td></tr>
<tr><td>199</td><td> 373 </td><td> 53.5% </td><td> 2344 </td><td> 6.29 </td><td> 12 </td><td> 14 </td><td> 67.3</td></tr></table>
It should also be noted that out of the eight non-expansion teams that put a rookie into immediate starting duty, three decreased their win total from the previous season and only two improved their win total by more than one game. In all, the ten franchises that took this route averaged six wins in year one.
Similarly, the five quarterbacks that started the year behind a veteran, but at some point during the season were inserted into the starting line-up had an average stat line as follows:
<table><tr><td>ATT</td><td>COMP</td><td> % </td><td>YARDS</td><td>YPA</td><td>TD</td><td>INT</td><td>RATE</td></tr>
<tr><td>105 </td><td> 206 </td><td> 51.1% </td><td> 1160 </td><td> 5.64 </td><td> 7 </td><td> 8 </td><td> 62.6</td></tr></table>
These five teams had only one team with declining wins from the previous season, but none improved by more than two games. On average, these five teams won six games—an improvement of one game over the previous year’s average.
Meanwhile, the leading passers for the eight teams that kept their first-round picks on the bench came out with the following average stat line:
<table><tr><td>ATT</td><td>COMP</td><td> % </td><td>YARDS</td><td>YPA</td><td>TD</td><td>INT</td><td>RATE</td></tr>
<tr><td>226 </td><td> 379 </td><td> 59.7% </td><td> 2717 </td><td> 7.17 </td><td> 18 </td><td> 12 </td><td> 84.6</td></tr></table>
Of these eight teams, only one had a decline in wins from their previous seasons (The Vikings, coming off of a 15-win year). Five of the eight improved by more than one game, and the average total wins for these teams was eight games.
This information affirms the common sensical notion that playing a rookie quarterback significantly decreases the production from that position compared to the play of veterans. Furthermore, it is more likely to cause a decline in wins and incredibly rarely leads to immediate success for the team.
Year Two
While the data from year one was hardly groundbreaking, data from year two should be the most telling. This information will allow a comparison between the play of a first-year starter who has a year on the bench to that of a rookie, while also checking the merits of the idea that a year of playing experience puts a quarterback farther along in his development.
In their second year as full-time starters, the ten quarterbacks who played immediately showed improvement in their stat lines:
<table><tr><td>ATT</td><td>COMP</td><td> % </td><td>YARDS</td><td>YPA</td><td>TD</td><td>INT</td><td>RATE</td></tr>
<tr><td>203 </td><td> 352 </td><td> 57.8% </td><td> 2422 </td><td> 6.89 </td><td> 13 </td><td> 12 </td><td> 78.1</td></tr></table>
Only two of the teams saw their win totals decline, while half of them (five) saw an improvement of more than one win. Because many teams had a decline in wins the previous season, these improvements led only to an average of eight wins apiece. It should be noted that Manning’s Colts improved from three to 13 wins in year two.
The five players who had some significant starting experience as rookies were stepping into the role of full-time starters for the first time. However, their partial on-field training did not lead to the same level of improvement, but still increased their production:
<table><tr><td>ATT</td><td>COMP</td><td> % </td><td>YARDS</td><td>YPA</td><td>TD</td><td>INT</td><td>RATE</td></tr>
<tr><td>215 </td><td> 402 </td><td> 53.5% </td><td> 2438 </td><td> 6.07 </td><td> 14 </td><td> 11 </td><td> 72.3</td></tr></table>
Their teams, likewise, did not show the same level of improvement, as two of the five teams declined, two showed significant improvement and one team stood pat. In all, they averaged seven wins.
Meanwhile, six of the eight first-rounders who sat on the bench in year one took to the field as starters in their first significant NFL action (though one of the six didn’t take over right away—Steve McNair, and one didn’t remain the team’s full-time starter—JP Losman). These six players produced as follows:
<table><tr><td>ATT</td><td>COMP</td><td> % </td><td>YARDS</td><td>YPA</td><td>TD</td><td>INT</td><td>RATE</td></tr>
<tr><td>203 </td><td> 352 </td><td> 57.5% </td><td> 2514 </td><td> 7.14 </td><td> 14 </td><td> 12 </td><td> 79.0</td></tr></table>
Only one of the six teams had a decline in wins (the Bills with Losman), and the group averaged eight wins. So not only did the teams not suffer the same dips that rookie quarterbacks provided in terms of team success, but the quarterback performance was far better than the numbers posted by rookie starters. In fact, the players who sat the bench for year one posted slightly better year two numbers than either of the quarterback groups with on-field experience.
The Jets and Chargers chose to keep Pennington and Rivers on the bench for an additional year with Vinny Testaverde and Drew Brees at the helms, respectively. Those teams were getting good play from the quarterback position and were winning games (they averaged 10 wins that year), so no change was made.
The data from year two begins to debunk the myth of the value of on-field experience. Players who saw the field as starters for the first time in their second years showed no evidence of being behind in their development, and slight evidence that it may have been even more beneficial than a year on the field. When combining the two groups that saw significant playing time in year one and comparing those totals with players that stayed on the sidelines, those who sat for a year had more completions and attempts, higher completion percentage, more yards, more touchdowns and a higher yards per attempt. Meanwhile, both groups averaged the same win total on year two, showing that either way can be successful for the team.
Year Three
By year three, all players involved in the study will have seen the field for significant playing time, and the long-term effects of the decisions made about first-round quarterbacks can begin to be gauged. However, it does get a little complicated, because in year three, some teams begin to give up on their prospects. For instance, Heath Shuler, Ryan Leaf, Akili Smith and Cade McNown had worn out their welcomes, and had limited playing time—if any. For that reason, their statistics will be excluded from averages, but their failures need to be noted. Additionally, the injury bug bites the study for the first time, as Michael Vick lost most of this season to a broken leg, and thus his statistics will also be removed from averages. Just because these players will be excluded from further statistical observations, it is necessary to remain mindful of their plights, as removing players removes the lowest performers, possibly increasing the overall average.
The eight remaining starters who had been starting since day one showed very marginal improvement from year two. With this much starting experience, it makes sense that their play may have leveled off by this time. Here’s how they fared in year three:
<table><tr><td>ATT</td><td>COMP</td><td> % </td><td>YARDS</td><td>YPA</td><td>TD</td><td>INT</td><td>RATE</td></tr>
<tr><td>252 </td><td> 428 </td><td> 58.8% </td><td> 2949 </td><td> 6.89 </td><td> 18 </td><td> 15 </td><td> 78.4</td></tr></table>
The total numbers increased, but the completion percentage and yards per attempt showed minimal change, if any. The teams all still averaged eight wins apiece, showing the numbers to be leveling off normally around what should be the league’s averages. The same could be said for the three remaining players who had some playing time as rookies, but with only three, the sample is too small to draw many conclusions. That being said, here’s how the remaining players performed in year three:
<table><tr><td>ATT</td><td>COMP</td><td> % </td><td>YARDS</td><td>YPA</td><td>TD</td><td>INT</td><td>RATE</td></tr>
<tr><td>252 </td><td> 429 </td><td> 58.7 </td><td> 2714 </td><td> 6.33 </td><td> 20 </td><td> 14 </td><td> 79.3</td></tr></table>
These numbers are very similar to the first-year full-time starters, and their teams also averaged eight wins.
The five players that had been held out of their first year but started in their second years continued to improve, albeit somewhat marginally. They finished year three with the following average stat line:
<table><tr><td>ATT</td><td>COMP</td><td> % </td><td>YARDS</td><td>YPA</td><td>TD</td><td>INT</td><td>RATE</td></tr>
<tr><td>266 </td><td> 440 </td><td> 60.5% </td><td> 3005 </td><td> 6.83 </td><td> 18 </td><td> 14 </td><td> 81.3</td></tr></table>
This group offers the first 3,000 yard, 60 percent, 80+ QB rating of the study thus far, aside from the veteran starters in year one. However, the win average dropped slightly to seven.
There is only one remaining group, and it contains only two players. Again, this is a small sample, but stats cannot be ignored just for that reason. Starting for the first time as third-year players, Chad Pennington and Philip Rivers averaged the following stat line:
<table><tr><td>ATT</td><td>COMP</td><td> % </td><td>YARDS</td><td>YPA</td><td>TD</td><td>INT</td><td>RATE</td></tr>
<tr><td>280 </td><td> 430 </td><td> 65.2% </td><td> 3254 </td><td> 7.58 </td><td> 22 </td><td> 8 </td><td> 97.8</td></tr></table>
These two teams also averaged 12 wins. Naturally, the Chargers and Jets had much more going for them than their strong quarterback play, but the fact that these players performed so well in their first year as starters certainly goes to debunk the idea that on-field experience is the most important aspect of developing a quarterback. Now in the third year, it might seem appropriate to condense the study down to two groups: those with significant first-year playing experience, and those without. In such an analysis, you get the following results:
<table><tr><td></td><td>ATT</td><td>COMP</td><td> % </td><td>YARDS</td><td>YPA</td><td>TD</td><td>INT</td><td>RATE</td><td>WINS</td></tr>
<tr><td>With First-year:</td><td> 252 </td><td> 428 </td><td> 58.9% </td><td> 2885 </td><td>6.73</td><td>18 </td><td> 15 </td><td> 78.7 </td><td> 8</td></tr>
<tr><td>W/O First-year: </td><td> 236 </td><td> 383 </td><td> 61.6% </td><td> 2691 </td><td>7.03</td><td> 17 </td><td> 11 </td><td> 85.5 </td><td> 8</td></tr></table>
Again, a reminder that the group of first-year starters is absent four “draft busts,” and the group without first year experience is absent the pass-challenged Michael Vick. Those omissions are likely to have artificially inflated both of these averages a little, but especially the first-year starters. While that is not enough evidence to definitively claim that sitting for a year (or two) actually enhances the development of a quarterback moreso than getting onto the field, it is certainly enough to disprove the converse. There is absolutely no evidence that backs up the notion that a quarterback is better off long-term if he gets onto the field quickly to “take his lumps.” There is also no evidence that suggests that the teams that make that decision fare better in the long-term, as the winning averages even out around year two and stay approximately even. In fact, the only apparent advantage of playing a quarterback right away is that if it is not working out, teams are able to cut the ties by year three, possibly earlier than those quarterbacks who have seen less time on the field. Of course, that also brings up the point that the number of “draft busts” are significantly higher among quarterbacks forced onto the field during their first season, lending circumstantial evidence to the notion that playing in year one might actually decrease the chances for success among quarterbacks.
Additionally, with no evidence that any long-term benefits exist for playing rookie quarterbacks, it seems a good time to re-state the evidence from year one showing that rookie starters fare far worse than veterans, and often lead to a one-year decline for the team. This begs the question as to why a team would sacrifice some measure of success one season if it bears no long-term benefits. Indeed, it could be said to be a fool-hearty decision, unless the team is hoping for an improved draft pick the next year without the appearance of “tanking” a season.
Finally, it should be noted that the two quarterbacks who sat for two seasons combined to show better production (by QB rating) than all but four quarterbacks in the study showed in any single year (Roethlisberger, P. Manning, Palmer and Culpepper). Such a fact combined with the other data provided in this analysis suggests that each year of development on the bench may be worth slightly more than a year of development on the field, and indicates that the best chance to ensure that there is never a drop-off in team or quarterback performance is to let early picks at quarterback sit.