Longer answer: discrete categorization of life is going to be messy, but that's our fault; humans are obsessed with discrete classification and attempt to apply such classifications to systems which aren't so simple. Clearly monotremes are similar to other mammals in many respects but very different in others. In trying to determine whether or not platypuses are mammals, you're really question the nature of the manmade discrete category of 'mammal', not the nature of platypuses themselves. Platypuses remain the same no matter how we choose to classify them.
Are peaches and cherries the same species? What definition would you like to use?
Cabbage and broccoli? Yes.
West and east African elephants? No, it turns out.
Quaggas and zebras? Yes, it turns out.
But all surviving monotremes live in and around the water, in places where they mostly compete with marsupials. Marsupials have an advantage on dry land. But if a marsupial mother with babies in her pouch swam, she'd drown them. So monotremes have an advantage in the water.
Placentals, of course, usually outcompete both. With occasional exceptions, such as the success of opossums in North America.
> all surviving monotremes live in and around the water
Wait, what? The surviving monotremes are the platypus and several species of echidna. The echidnas are woodland animals without any particular aquatic adaptations.