There are two phases to getting your dog to work off leash, reliably:
1. Making your dog think he’s on-leash (that you can correct him consistently for unwanted behavior, like running away or not responding to the come command).
and
2. Building a conditioned response to off leash commands, through repetition and proofing.
It’s the first phase that most people have trouble understanding. The trick is to let your dog wear only the tab (a one-foot short leash) when you’re close enough to him to grab him and correct him (like in the house or backyard) and then working with the long in the exact same way that you use the tab, when you’re in other environments.
This means that you’ll be replicating the way you would work with the tab… even though your dog is wearing a long line. So, if you yell “Down!” to make your dog go into a down-stay from 25 feet away, YOU DO NOT tug on the leash from 25 feet away, but rather walk to your dog and correct him as if he were wearing the tab. If he runs, you’re in a position to step on the leash and then go all the way to him and correct him for running away– then put him in the down position.
After correcting him a couple of times for running away, in a few different environments, he won’t run away anymore. You may still need to walk back to him to correct him near the snap on the long line, though. Just as you’d do if he were wearing a tab.
What this does is it plays a mind game on your dog: He never knows if he’s wearing the tab or the long line. After a month or so AFTER the last time he’s tried to run away, if you’ve done your proofing correctly, you can substitute the long line for the tab. It’s at this point that you’ve extinguished the behavior of running away and you’re just using the tab more as a communication tool.