Here's the bottom line:
Total goals in three most recent seasons: 12/13=>17; 13/14=>16; 14/15=>20 (and counting). If the trend continues, he'll score around 22 goals next season.
Goals in the Premier League only: 12/13=>17; 13/14=>15; 14/15=10 (and counting). If the trend continues, he'll score around 6 goals in the league next season.
Total goals per game (non-starter counts as 1/2 a game): 12/13=>0.56; 13/14=>0.50; 14/15=>0.46. If the trend continues, he'll score 0.42 goals per game next season.
Total goals per league game (same rule for non-starter): 12/13=>0.62; 13/14=>0.50; 14/15=>0.30. If the trend continues, he'll score 0.15 goals per game next season.
This is raw data, completely uninterpreted. What it's saying is that (i) he scored the most goals this season, and (ii) his numbers are declining in every other category. None of that is disputable.
Now we come to the controversial bit: Why are his numbers declining? Some have said it's because of a lack of effort. Others have said it's because of poor service. Others have said that under-performing colleagues result in an under-performing Lukaku. I'd like to suggest one other potential factor: The more he plays in the Premier League, the more other teams understand his approach and come up with better strategies to shut him down. This also explains why his goals per game were highest in the Europa League - because he was a new face for them.
So, if this is a factor, how do we change it? First, he must learn new tricks. We've said this about McGeady: one-trick ponies don't last that long. Second, he needs a partner upfront, because that would open up a world of other opportunities. Third, we need to be a little more savvy in coming up with strategies: In some games, that might mean keeping him on the bench.
While I like Lukaku and believe he was a good acquisition, I do see danger in those stats. The trends are mostly going the wrong way, and we have to fix that.