Politico framed the issue as one of Sanders being in the political trap of being too successful to want to concede months ahead of the convention, and yet not successful enough to win. But the failure to concede or even change strategy has ramifications that extend well beyond his own candidacy for the presidency: "Choosing to back off too soon would anger or disappoint Sanders' millions of loyal supporters, his team worries. But deciding to continue fighting could risk damaging the likely Democratic nominee ahead of the general election, though that’s not a concern that weighs heavily on their thinking."
Such a stubbornness to continue to the convention evokes memories of the 2000 election, with Nader's 4% of the vote ultimately serving as a spoiler in key states, most notably Florida given the historic closeness of the race and its resolution in the Supreme Court. Of course, one can make a number of arguments on the Nader front: the race didn't need to be that narrow if the Democratic Party had a stronger candidate, and we don't know with absolute certainty how (or if) Nader voters would have otherwise voted. But the point is that in a very close election, even just a couple percent of the voting population can make enough of a difference as to swing an election outcome from the Democratic to the Republican Party. And while Sanders is not threatening to run as an independent candidate, his willingness to remain on the attack with respect to Clinton provides enough ammunition for the Republicans to use in the general election, and arguments for Democrats and independents not to trust her leadership, that he risks empowering the interests that he purports to so oppose.So while this is not a wholly analogous case -- we are still talking about the Democratic nomination and not the general -- there is reason to question his motives (policy as he suggests, or ego as his actions imply) and to discuss the potential for his campaign strategy as promoting a spoiler outcome moving forward.