I posted the wrong link - sorry about that. It seemed intuitively obvious to me that gerrymandering caused a lot of vote distortion but as I've been reading more deeply into political science research, I've seen that what's intuitively obvious ain't necessarily the truth.
In the article I meant to link, political scientists actually ran simulated elections based on gerrymandered and non-gerrymandered districts (taking actual vote totals and distributing them differently). There was very little effect, going all the way back to the 1950s. The biggest effect they could generate was 7 elections being tipped - and that was making some very generous assumptions on the pro-tipping side.
The big difference maker is incumbency - as a bunch of Republicans got swept into power during the 2010 Tea Party elections, those Republicans had the incumbency advantage in 2012 (post-gerrymandering). Surprise, surprise - they had approximately the same "incumbency effect" as Democrats - 5 to 7 points.
This "proposed" electoral college change was largely spit-balling - someone managed to get Reince Preibus to answer a question about it. He gave a non-commital "that sounds neat" type of answer. But the Republican governors all shot it down pretty fast - most of them are in hostile or 50/50 territory and have no desire to rock the boat.
Meanwhile, the Democratic-leaning National Popular Vote has actually passed in states controlling about 25% of the vote. This is an even more blatant attempt to rig the system (it's the law in - lo and behold - a bunch of solidly Democratic states) but you don't seem to be worried about that one.
Frankly, I don't like either proposal (I'm a libertarianish voter that spreads my vote all over - my last ballot had Democrats, Greens, Libertarians and one Republican on it). I believe federalism is a good thing and would rather it be strengthened than weakened even more.