The Importance of Drafting Well in the First Round

Introduction   

After completing my recent article on which teams drafted best in the first round between 1995 and 2004, I began to wonder about the impact of first round drafting success on team winning percentage.

NFL teams spend millions on scouting college players, testing prospects in every conceivable way at the combine, and bringing in potential picks for individual workouts. 

Yet every year, a few teams will draft busts—guys who are lacking motivation, are injury-prone, whose skills don’t fit the team, or who just can’t stay out of trouble.

According to Bill Polian, the general manager of the Colts, when a team drafts a bust, it loses three things it can never get back:

1) the time taken to scout the selection,

2) the money paid to the bust that eats into the team’s salary cap, and

3) the pick that could have been used to get a player who could have helped the team.

It seems intuitive that teams that draft well in the first round should enjoy success on the field, but it doesn’t always seem to work that way in real life.  For example, the Cowboys have not drafted a complete bust since 1994, but they have not won a playoff game in an equally long period of time.  On the other hand, the Packers have rarely drafted well in the first round, but have one of the best records in the league since the start of the 21^st Century.

The following report is based on two sets of data—numerical grades for each player selected in the first round between 1995 and 2004, and the winning percentage of each team in the league between 1998 and 2007. 

The lag time of three years was selected because it usually requires about three years for first round draft picks to sort themselves out between busts, average, and above average players, and to have a real impact on their teams’ success or lack thereof.

Grading System

I have graded all first round draft picks between 1995 and 2004 on a five point scale as follows:

1)      Bust—the epitome of a wasted pick

2)      Disappointment—not a total bust, but not worth the pick

3)      Starter—an average first round draft pick

4)      Pro Bowler—made the Pro Bowl at least once and generally had a good       career

5)      Canton Class—a consistently great player for many years

The grades for each first round draft pick were based on position-specific statistics, percentage of possible starts, and number of Pro Bowls to which the player was voted.  The stats were compared with all other first round picks at the same position, which allowed for a reasonably objective grading system.

Correlation

For a complete list of average draft grades from 1995 to 2004 for each NFL team, please refer to my article “Which Teams Draft Best in the First Round?”  These grades were correlated with each team’s winning percentage between 1998 and 2007.

For those who are not well-versed in statistics, the measure of correlation between two groups of data is the “r” value.  This value ranges between zero (no correlation) and one (perfect correlation) and can be either positive or negative.

The “r” value calculated for the correlation between draft grade and winning percentage is 0.383, which indicates a positive correlation between the two statistics.

But does the “r” value of 0.383 really mean anything?  Luckily, one can calculate a “t” value that can be compared to a table value to determine the significance of the correlation.  If the calculated “t” value is higher than the table value, then the correlation is significant.

The calculated “t” value for the correlation above is 2.27, which is higher than the table “t” of 2.04.  Therefore, the correlation between draft grade and winning percentage is significant.

For those who are wondering about teams that win consistently and therefore tend to pick late in the first round, I also ran a correlation between average draft position and draft grade.  The “r” value for this correlation was -0.237, which indicates a weak negative relationship between draft position and draft grade.  In other words teams that draft later tend to get less highly graded players.

However, the calculated “t” value for the correlation is 1.33, which is less than the table “t” of 2.04.  This means that draft position has no significant impact on player quality—some teams just have a knack for drafting well in spite of regularly drafting late.

Conclusion

NFL teams are right to spend so much time and money on evaluating potential first round draft picks.  Although drafting the occasional bust is not fatal, one of the surest ways for a team to win consistently is to get good value out of its first round draft picks.