Here is my theory:
The girls were never at the lake. The dogs picked up their scent at the bikes and the woods just off the bike path because their bikes (with the girls' scent) were put there by someone else. So the dogs just ran to the bikes and then to the wooded area and then stopped. That was it. If the girls had been on that road (on bikes, walking, or by car) wouldn't that dogs have kept going? Someone planted those bikes/purse there.
SBM
Your theory about the dogs could be true but it is hard to tell one way or the other. The dog handlers are experts and they keep meticulous training logs that can stand up to scrutiny in court proving that their dogs are reliable when used under the conditions encountered. A report from the dog's handler can be considered fairly likely to be accurate.
I want to say, though, that reading dogs is a combination of experience and long term training with that particular dog. So an onlooker's interpretation of the dog's actions may not mean what that person thought it meant. The only person who knows to any degree of certainty is that particular dog's handler.
If someone else planted the bikes and purse by the lake, the dogs would have indicated on the bikes and then gone... essentially nowhere. Typically, a dog in that situation will circle the object starting close up and spiralling outwards until the handler gives up and either starts the dog all over again, moves to a different spot in the same general area to try again or just gives up for that area.
Dogs are trained to follow the freshest scent of the victim that is available. If the freshest scent was only on the bicycles and all other scent was older, the dog would stick to the bicycle and ignore the older scent.
If the girls were by the bikes at any point that afternoon, the dogs would start by circling the bikes to find the freshest trail from whichever point they approached them. Then the dog would move along the same line that the victim moved along (in this case, theoretically from the bikes into the woods).
If the victim made a sharp turn, the dog may overshoot the turn by anywhere from a couple inches to several feet and then stop, circle back to the last spot they could smell the victim and then cast around as above until they find the new direction.
If the victim is riding a bicycle or is being carried, that usually leaves plenty of scent for a dog to find.
Dogs can be trained to trail victims that are in vehicles but it's pretty hit or miss because there are so many variables. For instance, some vehicles are quite airtight (for instance, a vehicle with windows up and a/c set on recirculate) and some vehicles are quite leaky. Usually, the newer and higher cost the vehicle, the more likely it is to be very airtight. The older the vehicle and the cheaper, the more likely it is to be relatively leaky. Road conditions can kill off scent, particularly heavy gasoline or diesel fumes.
So if the dog keeps tracking the victim and you know they were in a car (because the track leads onto an interstate, for instance), that's a positive indicator that the victim travelled that way. But if the dog stops tracking, it doesn't mean that the victim was picked up in a helicopter, it just means that particular dog under those particular conditions was not able to find any more scent.
Now, what did the FBI Bloodhounds find? I have no idea because I'm not either dog's handler.
So in the end, I don't think any firm conclusion can be drawn at this time based on the evidence given. The only report we're getting is from someone who is not either dog's handler and probably has never handled a tracking dog herself.
