My first impressions are that I don't know if you meant the subject to be the man walking towards me or the man walking away. I keep looking from one to the other rather than having my eye settle on one. I think this is the biggest factor in the final score.
As a couple of the commenters noted, there is also very little else going on in the shot. if you look at the high scorers you note the following:
1. the are usually B&W - I don't know why but this seems to enhance candit shots like this
2. the subject tells a story
your shot tells a story but it is very much a short-story (man crosses road) compared to the novel that most people want, where they can engage with the subject.
Other than the man in the suit being a little interesting, I feel you've capture a crowded crosswalk, but failed to engage your viewer in the shot. It records a moment in time, but I don't feel it does much more than that.
The gentleman in the suit has a very purposeful stride and concerned appearance. He's either deep in thought or something has not gone right for him. I like that comparison against the rest of the folks, who all seem so much more casual.