I did this as a look into Eldar specifically and shared it over on Warhammer Competetive on reddit, but I feel it paints a picture most of us can appreciate, so I'm sharing it here as well:
I've been seeing a lot of talk about how blasts would impact Eldar, and I wasn't seeing anyone lay out numbers to back anything up, just a lot of conjecture about MSU being the best way to play from a wounds perspective.
So I crunched a bunch of numbers, did all the tested Aspect Warriors against the worst profiles for easy math, and came to a couple of conclusions:
And while it's easy to argue that you could kill X, Y, or Z by applying buffs A, B, or C then we're looking at a larger investment to remove that unit, meaning the points invested only increase and the ROI continues to drop. Additionally any modifiers in the defending unit's favor (-1 to hit, +1 to their saves, re-roll failed saves) also make the unit harder to kill, often with less investment than it'd take to kill said unit (especially since the first two can be gained from terrain).
- While the 6-10 model range can't be shot less than 3 times from Blasts, on average anything that isn't D3 already shoots at least that many shots or more on average (ex: D6 averages 3 shots, 2d3 averages 4 shots). This means the floor is raised on the minimum number of shots we could expect, but it doesn't actually change the average when comparing our options. Making these units basically the same for mathhammer against anything that isn't D3.
- The worst amount of damage from a single shooting attack (I'm counting double shooting Leman Russes as 1 shooting attack since they need to hit the same target both times) was 5.33 from the Wyvern against a unit of 20 Guardians standing in the open with no protection of any kind. The Wyvern makes back 35.53% of it's points in that round of shooting, hardly coming across as a real boogyman it's been claimed to be.
The tl;dr is that I just don't think blast weapons are as devastating as we've been lead to believe they are. They're better than 8th edition when aimed at the right targets, sure, but they're not exactly going to delete whole units a turn from a single volley like some seem to be implying.
And of course here's the data for those who want to see the numbers for themselves: https://docs.google....dit?usp=sharing