Drafting an NFL QB

by Kevin Meers

On draft day, every team gets a powerful gift: hope. Each team can hope that the guy they got is going to take a franchise to new heights, bring home Superbowl trophies, and more glory than they can imagine. Every fan can hope that their team drafted not only a phenomenal first rounder, but also found a diamond in the late round picks. These feelings are especially true of the quarterback position: people hope for the next Peyton Manning and fear the next JaMarcus Russell.

Many assert that teams should stay away from first round QB because so many are busts; others argue that the first round is the best place to take a QB. Looking back on the last thirty years of the draft, it turns out that 39% of QBs taken in the first round became “elite” QBs (4,000 season passing yards, 60% completion percentage, a certain “put-the-team-on-my-back”-itude) while 39% become “busts”; the remaining 22% become middle-of-the-pack players. In fact, the first pick is the best time to take a QB: nine out of the fifteen QBs selected first overall in the past thirty years have become elite while only two have turned into busts. After the first round, the chances of selecting an elite quarterback plummet to 19% in the second, 6% in the third, and dwindle to 1% in the seventh round. Contrary to what many believe, the earlier you select a QB, the better the odds that he is actually good.

Having an elite or bust QB has a significant effect on win percentage. Teams with an elite quarterback, on average, win 55% of the time while teams with a bust have a 35% win percentage. Since the average team wins 50% of the time, the benefit of having an elite QB is +5% on win percentage while a bust performs 15% below average – a huge effect for one player to have on 53-man roster.

Knowing these stats, we can find the expected value of drafting a QB in any round in terms of win percentage added. The only time that drafting a QB has a positive expected value is the first overall pick, with an expected value of 1%. Overall, drafting a quarterback in the first round has an expected value of -5%, and the expected values quickly drop from there until reaching -14% in the fifth round. This finding confirms the idea that the earlier you draft a QB, the better.

It seems like drafting a QB might not be such a good idea after all. However, teams must gamble on QBs in the draft because elite quarterbacks stay on the teams that draft them for around six years, on average. That is far too long for franchises to wait around while wallowing in mediocre quarterback play. So teams have to take a chance on drafting quarterbacks – especially early in the first round.

It is also important to understand the magnitude of that risk. Drafting an elite quarterback pays immediate dividends, as average win percentage rises from 41% to 52% in two years. It takes teams, on average, eight years to reach 0.500 after drafting a bust QB. Drafting a quarterback is a necessary gamble that most teams lose, but they must try anyway in hopes of landing a player who will transform their franchise.

Finally, we can model win percentage based on years since drafting a quarterback. Through a quadratic regression model, we find that Win % = 0.386 + 0.054(years since drafting) – 0.006(year since drafting2)

So if your team drafts a quarterback in the first round this year, its performance will probably look something like this:


About these ads
This entry was posted in NFL Football and tagged , . Bookmark the permalink.

3 Responses to Drafting an NFL QB

  1. Pingback: Drafting an NFL QB » Stathead » Blog Archive

  2. Pingback: links for 2011-06-01 « Vier Viertel + Nachspielzeit

  3. Andy says:

    “In fact, the first pick is the best time to take a QB: nine out of the fifteen QBs selected first overall in the past thirty years have become elite while only two have turned into busts.”

    I assume JaMarcus Russell is one of these busts. Who is the other? I would consider Tim Couch, Alex Smith, and David Carr all as busts as well.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s