Huh… weird. I don’t understand how it’s even possible for him to have been wrong about that, since IIRC the original claim was that they had a formula for how much damage a spell of a given level should do, and then for a few iconic damage spells they wanted to feel extra powerful, they went a few dice above what that formula suggested. Did he just… hallucinate doing that?
Because he's human?
This is why 6E will be designed by gnome artificers.
IIRC the original claim was that they wanted players to preferably pick the iconic spells, so they tuned the formula so that the other spell would not steal the scene from the icons. But I don't think they were ever clear on what was the intended power level.
But, I mean, this isn’t just an “I made a math error” kind of mistake. This is claiming you did a thing, and then 10 years later being like “actually I did some math and apparently I didn’t do the thing I said.” Like, brother, are you saying you lied about it 10 years ago? I don’t know, maybe the podcast explains it better, I’ll listen to it when I have some time, but right now I’m just very confused what his counter-claim even is.
The judgement in these posts is strong. People make mistakes all the time.
I do remember him saying, back when 5e was still in playtesting, that he specifically wanted big, splashy damage spells to be the most effective, because they feel the most fun to use, and that his intent was to design the math such that the character optimizers would be able to pull out their abacuses and show that, yes, fireball is the best spell. That was a pretty general claim that I could see meaning they designed those spells to be at the peak of what their math allowed while others were more average. But I could have sworn he later made the more specific claim that how they ended up achieving that goal was by figuring out how much damage a spell of a given level should do, and then giving those spells one or two more damage dice above what they should do for their level.
I'll say this, it is clear his new work is soundly math based, if that matters to you (or others).
Code for A.I.
