I don't have a strong feeling about the statistics here. But I guess a $1,000 mistake on 1 in 5 of all returns does a little unlikely. Bigger mistakes on fewe

I don't have a strong feeling about the statistics here. But I guess a $1,000 mistake on 1 in 5 of all returns does a little unlikely. Bigger mistakes on fewer returns also seems unlikely since the higher the bracket, the less likely you're relying on any default or software. But I take the point that H&R Block and Intuit could simply start from the filled out returns provided by the gov't. Harder to understand their objections given that.