From one perspective, there's no particular reason it should. Drew Edwards points out that these rankings aren't issued by a league-run organization (like the NHL's Central Scouting Bureau), but are rather compiled by averaging teams' individual scouting lists. Edwards writes that "It's well known that teams lie to avoid tipping their hand and therefore the rankings often bear little resemblance to draft day," and the system's certainly set up in a way that could potentially reward that course of action. If you're Hamilton, for example, and you're drafting fifth, you can make a case that it's most valuable from an incentive point of view for you to keep the player you covet most below that on the list you submit to the league (or leave them off it altogether); that would seem likely to drop them down the league rankings, and that might convince teams with higher picks to go in a different direction, leaving your prospect available when you pick.
It's worth pointing out that that particular strategy doesn't guarantee success either, though. In fact, any lie that you know is a lie can be just as illuminating as the truth; from a CFL perspective, if a prospective player like Mitchell who's clearly talented and doesn't seem to have any obvious negatives was buried at the bottom of this list, that might tell teams that everyone is interested in him but lying about it. Thus, just reversing the order of your draft list wouldn't necessarily help you all that much. Teams invest plenty of money in scouting, so it's not like they're going to make their picks based solely on this list regardless of what it says. Oddly enough, if the central list is considered relatively unimportant to teams' decisions, that might actually make it more accurate from an incentive standpoint; if the central rankings aren't really being used by teams from either a positive or negative perspective, the incentive shifts to providing the public with the most accurate draft information out there (as that would hopefully spur league web traffic for the rankings and general interest in the draft, both of which would be positives for the CFL as a whole and for its member clubs).
How does that work from a historical standpoint? Well, the final April 2010 rankings compare reasonably well to the actual 2010 draft. The rankings got the first- (Shomari Williams), third- (Cory Greenwood), fourth- (Watkins) and 10th-(Shawn Gore) overall picks spot on, and they weren't far off on Joe Eppele (ranked fifth, picked second), Rob Maver (ranked sixth, picked fifth), or Brian Bulcke (ranked sixth, picked seventh). There were some notable misses (second-ranked offensive lineman John Bender fell to 17th overall, but part of that was thanks to growing reports of NFL interest, perhaps the most difficult element of the CFL draft to judge), but on the whole, that's a pretty good showing for the rankings, and it's one that suggests we certainly shouldn't write them off entirely (especially when you consider that they aren't explicitly taking team-specific positional needs or interests into account).
In the end, these rankings are mostly there as a discussion starter, and they accomplish that purpose admirably. They've shown in the past that they aren't completely divorced from what may actually take place on draft day, though, and that's why they're definitely worth noting. They might just predict the top overall pick for a second consecutive year, too, as Mitchell is certainly a capable prospect who would fill a good position for Winnipeg (who could use more top Canadian content on offence), he turned in a strong showing at E-Camp and he's already been linked to the Bombers. We'll see if Mitchell's name is in fact the first one called on May 8, but that possibility definitely doesn't seem all that remote.