I have found a pretty big difference in timing when testing in my browser.
I used the following script:
WARNING: running this will freeze your browser a bit, might even crash it.
var n = 10000000, i;_x000D_
i = n;_x000D_
console.time('selector');_x000D_
while (i --> 0){_x000D_
$("body");_x000D_
}_x000D_
_x000D_
console.timeEnd('selector');_x000D_
_x000D_
i = n;_x000D_
console.time('element');_x000D_
while (i --> 0){_x000D_
$(document.body);_x000D_
}_x000D_
_x000D_
console.timeEnd('element');
_x000D_
<script src="https://ajax.googleapis.com/ajax/libs/jquery/2.1.1/jquery.min.js"></script>
_x000D_
I did 10 million interactions, and those were the results (Chrome 65):
selector: 19591.97509765625ms
element: 4947.8759765625ms
Passing the element directly is around 4 times faster than passing the selector.