If you aren't marking up the gas then the operating costs and profits must (eventually) come from the $15/mo, correct? So does this work similarly to the gym b

If you aren't marking up the gas then the operating costs and profits must (eventually) come from the $15/mo, correct? So does this work similarly to the gym business model (i.e. over-subscribe the service with the assumption that some portion of subscribers won't use it very often and will subsidize the heavy users)? The requirement to plan ahead instead of offering on-demand delivery seems like it might encourage that dynamic ("Crap, I'm nearly out of gas and forgot to put in my Yoshi order. I'll just fill it myself this time").