Taking a slightly different angle on this, think about this in terms of the "natural selection" of top league athletes from the general population.
Soccer is drawing its top-league players from a pipeline many times larger than that of hockey -- current estimates are that there are 100-200 times as many organized soccer players as hockey players in the world, and soccer also has a more efficient development system to boot, so we could be talking about even larger proportions than that.
Given the wildly different proportions, it can be assumed that there are a lot more soccer players than hockey players with a professional-level skill set. If we say the top 0.1% of the pipeline is our "most skilled" tier, we are looking at hundreds of times more soccer players who have maxed out their skill levels.
But being a professional player isn't just about skill set, it's also about athleticism and mental skills. Which sport selects more heavily for pure athleticism? The answer to that should be fairly obvious, even if we don't love it as hockey fans. Mental skills is a different question which I will leave for a different thread.
So if soccer selects more heavily for athleticism, and soccer has a prospect cohort multiple hundred times larger than hockey's, it would seem to follow that the players who make it to a top-level soccer league have a dramatically smaller margin for error athletically than their hockey counterparts. So it follows that a soccer player at the top professional level has reached a higher athletic "bar" than his hockey counterpart. It takes more work to reach a higher bar, all else equal, so we can conclude that it requires a greater level of work to attain and sustain a top-level soccer career than an NHL career.
Of course that doesn't mean that any given specific NHL'er has done less work than any given specific soccer player. But I bet this logic would prove sound if it were possible to measure how often professional players in either sport "cheat" their exercise regimens, diets, etc.