The Packers returned 47.56% of the kickoffs last season, ranking 19th in the league.
I guess kickoff return TDs aren't included in the data. Three of the five teams which scored one are ranked behind the Packers.
As far as I can tell there was only a single penalty during any of the Packers' kickoff returns and that one was an offensive holding penalty against Brandon Bostick.
The Packers recovered three onside kicks during the 2014 season, tied for second most in the league.
It would take way too long to figure that one out but it's possible it factors in quite a bit.
It's looking like the disconnect between the Packers KO return average and average starting field position might be a collection of small, distorting factors.
KO TDs counting for nothing, even if there were only 6 of them, is one small distortion. It stands to reason those should count as a starting position at the opponents zero yard line. Even if such a thing does not exist, a program could be made to think it does even if a canned query tool cannot. That's not a criticism; we use the tools we have available to us.
On-side kicks are an interesting case. Each one recovered adds a zero yard return, or something close to it, deflating the KO return average, while those plays yield decent field position.
Somebody needs to do a study of KO returner performance that accounts for player judgement. A KO returner's average isn't very relevant if he's padding his numbers with bad returns, such as taking the ball 8 yards deep and running it back to the 19 for a 27 yard return. It seems to me a returner should be judged on the average field position he yields, not the misleading return average. Maybe PFF has something in the archive?
At the extreme the Jets returned 58 KOs, twice as many as the Packers. That disparity is not accounted for by the Jets surrendering 49 more points than the Packers. It would appear the Jets are return happy, running everything out of the end zone. I'd be interested to see how they rank in starting field position. With all those returns and a not very impressive return average, it would stand to reason their starting field position would not be that great.
One thing you might do is compare each team's KO return average ranking to their average starting field position ranking. We know what perfect correlation looks like. Crudely put, 100% uncorrelated (random) would mean the average spread between the two rankings would be 16.