This has been said, and better, by Rob Wiblin and others, but I'm restating it here to draw attention to its implications. Many criticisms of effective altruism (even the excellent ones in this Boston Review forum) go like this:
1) X is important (or Y is unimportant).
2) Effective altruism ignores X (or gives undue weight to Y).
3) Therefore, effective altruism is flawed.
If the goal is to criticize the fundamental idea of EA, then I think these criticisms miss the mark. If (1) is really true, then all that implies is that EA should focus on X. If the goal is to criticize the EA community, then these criticisms are more on point, though there's a question of whether a community is defined by what its members do or by the fundamental ideas that unite them.
But similar points have been made by others, so I'd like to focus on two reasons why the framing of these criticisms matters.
1) If they consciously adopted the role of internal critics to EA, those in favor of institutional reform, for instance, could be part of a serious debate within EA over whether focusing on policy advocacy is effective. This would not just be a more precise statement of the criticisms - it could improve EA more than an external criticism. More importantly, it could fill a gap in research on the effectiveness of policy advocacy, something that would be edifying not just for EAs but for anyone who cares about evidence.
2) Framing the debate in this way could lead those who favor a systemic approach to realize their significant ability to impact the world. People who call for focusing on institutions often, it seems, take it as an excuse not to do much to address global problems. If it turns out that we can do more by focusing more directly on institutions than we can by donating to charity, this should call for those who care about policy to follow the EA movement in making a serious personal commitment to change the world.