To be honest, your post was just OK, but it lead me to your button generator (http://www.performable.com/buttons/), which is hands down the best online button generator I've come across.
Having the "Yes, send me product updates (big improvements coming) & marketing tips!" checkbox enabled by default gives off a bad smell. It is the little brother of e-commerce sites that sneak stuff into your cart when you aren't looking.
Let subscriptions be an active decision instead of an accident.
Those 3 chart types are the only data visualizations I recommend using. I should have left out the pie chart, since I don't use those.
I wasted many years of my life at Compete obsessed about "data visualizations" only to learn that no one except me and my statistical team cared about them. Please don't waste as much time as I did playing with exotic visualizations.
Ohh I am sorry I thought you were emphasising the data the charts were measuring rather than the actual types of chart themselves!
If you only mean that we should use only line, bar or pie charts, then I would say that scatter graphs at time have a good use of showing outliers, such as some guy who spends say ten minutes on the site, or visually able to show for example how many people spend ten minutes and how many 3. Communicate in other words much more information than a line, bar or pie chart can at times. I mean, people did not invent these other charts for no reason :)
I was referring to the accuracy of the test results on abtests.com in general not the site in the link. Many of the A/B examples had multiple changes to the site but only noted one as making a difference.
In your case adding a headline to the form was overlooked.
You were using form submissions as a metric from what I can tell. So changes specifically to the signup form itself (such as the header) is likely to have as much of an impact as the video.
Did you track how many people watched the video then signed up as a metric? That may be more telling.