The best DNA parentage studies find that breeding is spread far more evenly across all ages of bucks than what was originally assumed, but there is still some selection by age going on. Yearling bucks are the largest single age-class of buck in the woods, but these studies find they are the smallest percentage of successful breeders. And the fewer the does available, the lower the percent of yearlings successfully breeding. In the best study I know of, I saw the breeding data broken down by week of the breeding process. In the first and last weeks, when the fewest estrus does are available, all successful breeding was conducted by bucks 3 1/2 and older. 2 1/2s and yearlings do most of their successful breeding during the peak of the rut, when the highest number of estrus does are available at one time.
It also appears the most aggressive breeders may be the 3 1/2 year-olds instead of the most mature bucks. Mature bucks are successful, but generally not on multiple does.
In addition, the strangest finding of all is that some bucks appear to never successfully breed, no matter how old they get. And it's not a small percentage either.