Re: Paralyze vs Stellar Cascades
 

Nope. I did extensive testing on this at one point. You can see it clearly in the debug log if you really want to get into it.

In a little more detail: When the AI casts a spell, any spell, it tries it on a bunch of targets and sees what gets the best result. It will try targeting enemy units with buffs, friendly units with artillery, whatever. It figures out which of those gets the best results, then goes ahead and actually casts it, rerolling all the random parts.
So the Dai Onis wolves are valid targets for the bless casting, even though they're not sacred, as even any enemies within range would be. Since there's only one actual blessable target around, the test casts are either successful or not. If a test cast on one of the wolves catches the Dai Oni in the spread, then that case may get used and it may not catch the Dai Oni in the AoE in the real cast.

If you look closely, pause at the right moment, when the Dai Oni doesn't get blessed, you will always see a wolf getting hit by the bless effect.
Try sending a higher precision self-blesser out with a couple companions, you'll see he misses himself too.
I first noticed it with Jotun thugs, who don't come with chaff, but I was raiding Pan, who has flying PD. They'd fly up round 1, I'd miss the bless and get killed.
Nothing to do with Precision. Mod in an AoE 1 version of bless and the problem goes away.

Re: Paralyze vs Stellar Cascades
 

Wow. That's fascinating.

Re: Paralyze vs Stellar Cascades
 

I thought the targetting system was well known, but guess it's a case of very little is ever really new, as it's often just knowledge that has been forgotten. thejeff has it right though, which you can see from reading the log file (btw it's impossible to fully understand how some of the mechanics work without being able to read the log file correctly. As trying without that ability often just leads to the spread of mis-information, or classic cases of "well what I saw was", and then basing a load of misconceptions from that faulty observation).

But one thing, Bless can miss if there's only one target (such as a lone thug). It's very rare, and I mean insanely rare. But it can happen, and one of the old-guard Russian guys, can't remember who (someone like ano, WingedDog, Psycho, Dimaz, Kuritza. Or perhaps vfb) posted a save file of it a few years ago (sure I have it bookmarked on another machine). I don't think the spell targetting has been patched since, so it could in theory still happen in v3.27. But then I'm certain some things have been patched (either intentionally or unintentionally) without a corresponding patch note, so maybe it doesn't happen anymore. Either way though (for the benefit of newer players) Bless missing a lone thug is not something you should ever plan to script for. ie. never script Blessx2 with a lone thug on the basis of the first one might miss, as that will just give him a free pass to cast something himself, and likely ruin your day as a result.

Although this Bless missing scenario is why there was a call at one point a few years back for CBM to make a self-bless spell to avoid it happening to lone thugs. Think it wasn't considered worth doing though as it would have taken up a spell slot to do so, and at a time when spare spell slots were at a premium (since it was before Illwinter patched in a load of extra spell slots).

Re: Paralyze vs Stellar Cascades
 

I'm not saying that's not possible, but I don't recall it and I've never seen it.
For a long time it wasn't understood exactly why self-blessing sometimes didn't work.
I remember the suggestions for a self-bless spell, but those were in response to this problem since it can't always be avoided: Dai Onis and other summoning creatures, flying enemies, etc.

It wouldn't be a bad idea, now that there are spell slots to spare.