Angels do often invest with each other, if only because they don't want to put in enough money individually to fill out a round. And there are a lot of angel fu

Angels do often invest with each other, if only because they don't want to put in enough money individually to fill out a round. And there are a lot of angel funds where that co-investing process is more formalized (shared due diligence funds, etc.). I don't think, however, that co-investing by itself really increases the odds. Maybe if I co-invested with someone who picks better or gets better deal flow :). But without increasing the actual # of deals, I don't see how you get closer to those big hits. As for ideas changing, the argument is not to bet against that fact, which would be silly, but instead to bet that the second or third idea will be similar to the bigness potential of the first. That is, if the founder can't come up with an initial idea that doesn't seem like it can realistically result in any kind of decent exit, the probability that they will random walk to one in the future drops. It's of course not zero, but I'm just saying you can increase the odds.