The most important position in the NFL: A Metric Analysis

I had a nagging question yesterday after reading David Pearson’s post in which he bucked conventional wisdom and suggested the Vikings should not take Matt Kalil (OT, USC). I wondered: What is the most important position in the NFL? And how can a person develop a metric to show this to be the case? When it comes down to it, I want to know what position is most critical to win games. Conventional wisdom suggests that it is a Quarterback first and then his right-hand man, the Tackle–particularly the left one–to protect him. But David looked at the past several Super Bowl winners and discovered no quality LTs in the bunch. OK, I wondered, Let’s see how this stands up to further analysis.

The first task in putting together a metric to compare players cross-positionally is to find common ground between players with very different strengths. Though I hate using peoples’ opinions at all for this purpose the All Pro lists are more than up to the task. The All Pro team consists of the best and second best players at their position across both conferences of the NFL, as voted by the Associated Press and other sports writers. Most importantly for our purposes, the All Pro team answers the question, “Who are the best players at a position in a given year?”

The next task is to decide on what time-frame to investigate. If we go back in the past too far it won’t be too fruitful, since football is a different game in the 2000s than in the 1960s. So let’s try ten years –starting with the Super Bowl between the “Greatest Show on Turf” (the Rams led by Kurt Warner) and the Patriots. This was, in many ways, the entry into a new era of pass-first football. Unlike Dave’s analysis, let’s widen the net to all playoff teams, since using only Super Bowl winners would produce a very small sample size. However, it is better to win the Super Bowl than simply make the playoffs as a Wild Card, so let’s weight the values toward those who advanced further, like so:

A player from…
A team that loses the Wildcard Round: 1 point
Divisional Round: 2 points
Conference Championship: 3 points
Super Bowl: 4 points
Super Bowl Champ: 5 points

Now, we don’t need to differentiate between players on first or second team All Pro lists since the initial hypothesis wasn’t whether a team needed “the best” player in the NFL at each position; it was whether a team needed a very good player. Any player on the team will be in the top 5% at their position, which is sufficient for our purposes.

Lastly, it should be noted that the All Pro teams have changed a little over the years. Some positions have been added over time, and each position is not equally represented. There may only be one Quarterback in a year but as many as 11 Safeties (as in 2010… Seriously, AP?!). Therefore, let’s take the total number of points accumulated by players at each position using the math above and divide that by the total number of players in each position on the All Pro teams of the same period. This gives us a raw average of the value of a position relative to the success of a team.

The Numbers
Quarterback 2.58
Center 2.28
Special Teams 2.08
Safety 1.85
Kicker 1.71
Fullback 1.67
Inside Linebacker 1.61
Outside Linebacker 1.61
Cornerback 1.53
Guard 1.49
Defensive Tackle 1.39
Punt Returner 1.36
Wide Receiver 1.35
Defensive End 1.26
Tackle 1.25
Tight End 1.14
Running Back 1.00
Kick Returner 0.70
Punter 0.29

Firstly, let’s talk a little bit about what the numbers mean. If a position rank is 1.00 that means that the player’s team, on average, makes it to the Wild Card round of the playoffs. If it is 2.00 the player’s team averages to the Divisional Round, etc. So in the numbers above, Quarterbacks in the past ten years have, on average, made it somewhere between the Divisional Round and Conference Championship. This is remarkable because it means for every QB who doesn’t make the playoffs (0 points) there needs to be corresponding top QB who wins the Super Bowl (5 points), indeed, a little more! So clearly, top Quarterbacks are not only making the playoffs with incredible regularity but advancing further in. Top Punters, by contrast, are doing roughly the equivalent of making the Wild Card every four years. This is not good.

Now there are some puzzling data here. For one, I don’t think anybody would consider Special Teams specialists the third most important position on the football field, or kickers the fifth or fullbacks the sixth. This brings us to an important point: these rankings are only useful if players have made the All Pro team on their own skill. The best kicker in the league may actually be on the worst team but due to inequitable opportunities he probably isn’t going to make the All Pro team. Likewise, a good fullback is a pure luxury that a bad team is unlikely to take on. At certain positions you gain advantages simply from being on the best teams. Hence, these numbers are not useful for non-skill positions. So let’s eliminate those and see what it looks like:

Quarterback 2.58
Center 2.28
Safety 1.85
Inside Linebacker 1.61
Outside Linebacker 1.61
Cornerback 1.53
Guard 1.49
Defensive Tackle 1.39
Wide Receiver 1.35
Defensive End 1.26
Tackle 1.25
Tight End 1.14
Running Back 1.00

This is much more handy, and it shows us some interesting trends. Firstly, Quarterbacks are the most important indicator of a successful team–that much is obvious and overwhelming. Next, however, may come as a surprise. A Center?!

Center: 2.28
Guard: 1.49
Tackle: 1.25

It may seem odd that Centers carry premium importance on the O-line, while Guards are the next in line and Tackles fall at the back. This may seem, off-hand, to invalidate the rankings altogether. However, before we give up on this exercise entirely, let’s take a moment to think why this might be:

Mediocre teams may have a single good offensive lineman. He will undoubtedly be a Tackle, because that is the position he has to play to best protect the Quarterback. So, good Tackles are spread around the league. In fact, teams that don’t have top-of-the-line Quarterbacks necessitate better Tackles, but we’ve seen that these teams are less likely to do well than teams with a strong QB. Bad teams sometimes acquire Tackles in hopes of shoring up shoddy lines, but they inevitably leave holes elsewhere. The best teams have the luxury of depth. Often that means their Guards are at the top of their position since their Tackle positions are already filled by average–but not necessarily star–players. Best of all is the ultimate luxury of a strong Center. Only the best teams end up with Centers that excel. Try and think of a great Center on a bad team… Keep thinking…

So, this is enough, I think, to answer David’s question. Should the Vikings draft Matt Kalil? Maybe. There is no single formula to win a Super Bowl. Two teams in the last ten years had no All Pro players at all and lifted the Lombardi (the 2001 Patriots and the 2007 Giants). But the numbers show that it is not necessary for the Vikings to go this route. If they go for Justin Blackmon, the WR out of Oklahoma State, and he turns out to be a top receiver then he is likely to be slightly more valuable than Kalil should he become a top Tackle–though it should be noted that Blackmon may still be less valuable than a Safety, Corner or Linebacker of the same star-power.

With this in mind, let’s compare relative values of Offensive and Defensive players:

Quarterback 2.58
Center 2.28
Guard 1.49
Wide Receiver 1.35
Tackle 1.25
Tight End 1.14
Running Back 1.00
AVERAGE: 1.58

Safety 1.85
Inside Linebacker 1.61
Outside Linebacker 1.61
Cornerback 1.53
Defensive Tackle 1.39
Defensive End 1.26
AVERAGE: 1.54

This might seem to demonstrate that it is very slightly more effective to have a player on offense who is the top at his position than it is to have a top guy on defense. However, this is itself misleading, because–as we’ve already noted–Quarterbacks are truly a category of their own. A great QB is a great predictor of success, so let’s see what happens when we remove QBs from the list:

Center 2.28
Guard 1.49
Wide Receiver 1.35
Tackle 1.25
Tight End 1.14
Running Back 1.00
AVERAGE: 1.42

Now, as you can see, Offense is considerably less important than Defense: 1.42 vs 1.54. This lends some credence to the phrase: “Defense wins Championships.” Actually, we might tweak it like this: “Great defense and great Quarterbacks win Championships.”

Conclusion

If the question is the draft, the best move is to take the best player. Of course, the problem is knowing who that is. If you believe Kalil has ability above and beyond other potential picks then it is the right move to take him. However, if you think he will be an elite Tackle, but you could also get an elite WR, CB or LB then things are different. This data can’t be used to tell you who is going to be good or bad, but what it can tell you is that good teams have certain common strengths at certain positions. A good team may have a great Punter, but it probably won’t (after all, good teams don’t have to punt a lot). It is most likely that a good team will have a great Quarterback, Center, Linebackers and/or Safeties. In the end the numbers back up what David suggested: It seems that the conventional wisdom that a team must have a great Left Tackle may, in fact, be wrong.

Advertisements
This entry was posted in Football, NFL Draft, Vikings and tagged , , , , , , , , , , , , . Bookmark the permalink.

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