And from their numbers, it doesn't look like they're using a reasonable estimate for Chernobyl. There've been some ridiculous estimates out there from both sides, ranging from "only the couple dozen who died directly" to "millions".
One can look at the approximately 10% higher mortality rate within the exclusion zone to get a rough sense of the consequences, but without knowing demographics, it's hard to draw conclusions from that. Probably the best (peer-reviewed) analysis I've seen compared doses with the US military's mortality data from exposure to the nuclear bombings in Japan. You get a figure of about 4000 extra deaths with moderate confidence and 5000 with low confidence (the error bars can be in either direction). So very rough ballpark of 9k deaths, plus the first responders and the like.
Yes, even when you include things like that, nuclear's death toll is lower than coal, no question. But it's not as low as they make it out to be. Their bias is obvious.
The deadliest nuclear accident, Chernobyl, was caused by defense department testing.
Yep, nuclear disasters can happen from both man and nature. That's hardly a comfort. Will "defense department testing" cause the next major nuclear accident? Very unlikely. But there almost certainly will be a "next major nuclear accident" - we just don't know what form it will take. It's a "known unknown".
Whereas Fukushima was all user error?
No. But if you're going to include "dam-induced casualties from storms", then you should include "people spared from storms by dams" also, it's only fair. And thus hydro's death count would be strongly negative.
Perhaps it is possible to offset renewables in such a way that they can provide 90% of our power needs, but no one has ever done it.
Speak for yourself. I live in Iceland where over 99,9% of the grid is renewable (primarily hydro). 99,9% renewable baseload at that. There's even serious preliminary work looking into building the world's longest submarine power cable to export power to the UK.
Again, not saying that hydro is my preference - I've stated my preference above. Just pointing out that your claim is wrong.
(Concerning the power cable: I'd support if A) they'd only be adding geo plants and wind to meet the extra power need, and B) the government would tax the power sales to the point where the cable makes just barely enough profit to economically justify its existence... but I'm sure that A) they'd probably just dam up the highlands some more - who gives a rat's arse that we have some of the world's most abundant and cheap geo and wind power that could easily compete on the European market, hydro gives a tiny bit more profit margin!; and B) the government would hardly push back at all on power export royalties because, hey, JOBS! Jobs damming up the highlands!)