Information about the larger structure of the grid, transmission direction, system redundancies/tolerances, failure cascades is a lot harder to come by. You could build up a picture by accumulating OSINT resources, mapping gaps, making physical observations etc., but synthesizing that into any sort of working model of the grid itself would require some pretty advanced engineering expertise and a lot of institutional knowledge. Anyone with sufficient smarts, training, and know-how is a lot more likely to be invested in the integrity of the grid than its sabotage.
Also, in purely economic terms it's not efficient for an antagonist to max out resources on trying to land a decisive masterstroke. It takes a lot of time, money, and moving parts to pull off such an attack. An ongoing rash of cheap micro-attacks, even poorly correlated, drives up costs in ratchet fashion, without any threat being sufficiently expensive to bring about massive change (eg burying major parts of the transmission infrastructure, switching to a different architecture, radically different security posture). The strategic goal is to raise the cost of 'normal' above the market's willingness to bear, and then exploit social/political failure modes.
Let's not get this specific here but yes it is the radiators.
We're not going to link to the article, but the one we saw went into great detail as to why shooting that one specific spot was both hard to repair and replace in a way that just shorting something out wouldn't. (Smaller target for latter... do we really need to spell these things out with this level of... specifics? O_o)
"If the windings within the transformer itself were damaged, it would probably require replacement of the equipment. Transformers of this scale are rarely manufactured without an order, which means we don’t have a lot of spares sitting around, and the lead time can be months or years to get a new one delivered, let alone installed."