Being "unreliable" is much more than exploding transformers. One means to measure this is with capacity factors.
What is the capacity factor of a typical windmill? I see that with the best technology we have today wind power has just reached 50% capacity factors. Under ideal conditions wind could approach 66% capacity factors. Solar does much worse with 25%, but I'll give it 33% to account for future improvements and perhaps some negative bias in reporting.
Then there are the power sources that burn stuff and/or boil water. This we have oil, coal, natural gas, biomass, and nuclear. These have capacity factors that top out around 95%, average above 80%, and on the low end are about 75%. I'll be nice and give them a 66% capacity factor to again account for any bias.
Hydroelectric is probably king in this. Capacity factors that exceed all others. Given an additional pumped hydro capacity it can account for lack of capacity from other sources. With multiple generators on a single site there might be a single generator that is down but the facility is still producing power at or near rated capacity barring exploding transformers. As awesome as hydro is I can confidently give it a capacity factor that is so close to 100% that it will be 100% for this thought experiment.
So then how much does this electricity cost? Hydro and the burn/boil group all cost nearly the same. Hydro wins on this up to 33% because of capacity factor. Hydro also is reliant on geography, we've dammed up all the rivers worth a dam in this world, there is no growth in this.
Wind costs about the same as the burn/boil group by rated capacity (a bit more really but I'll be nice again) but with its production capability reliant on the weather and geography it takes two to three times as much of them to get the same power out. This is part on capacity factor and part on the fact that the wind blows when it wants, not when we need it.
Solar is the worst. With a capacity factor of even 33% we'd need three times the amount of capacity from a 100% source like hydro, and some means to transfer or store the energy to when and where it's needed. Ignoring capacity factor solar power costs three times what we get from burning and boiling. With capacity factors included it goes up almost three times again. If I stop being nice and start looking at reality the cost of solar is near to or above ten times what it would take if we burned and boiled our power into existence.
You can mix and match all of this if you like, hiding much of the costs of renewable in the noise that is real life capacity factors, but it works out that solar is always a loss. Wind is either a small loss to a break even. Hydro is always a win but then we run into issues of geography and "environmentalists" that would rather see people starve than have a few common bait fish killed in a hydro turbine.
These environmentalists will likely complain about any energy source that emits carbon. On the basis of carbon output per joule produced nuclear and hydro rule, nothing beats those two.
What good is a nuclear power station when it's transformer blows up? It's good in that we can easily afford to build redundant nuclear power generation so that loss of any one will not affect the grid. It's good in that with a capacity factor well above 66% that by building four power plants for every three we need that we can be well assured that the power will never go out.
(Don't bother trying to correct my simplified computations. I am well aware that my accuracy sucks. Point is that it should be close enough to the real world to show that nuclear and hydro should rule the world.)