The theory of decoupling engines from agility is based off of MW4.
Simply put, regardless of your top speed, the engine tonnage spent will not change your agility. Agility would be defined by other factors, relative to chassis (and variants).
For instance, a Gargoyle has the greatest natural agility (it's fixed by design) compared to the rest of the 80 tonners.
An Awesome has relatively crappy agility relative to the other mechs (particularly ones capped with a 300 engine). Boosting the agility of the slower Awesomes to like having a "325 engine" or "350 engine" or so would be a nice agility buff to the chassis.
What this can also mean is that something like the Dragon Slayer (Victor hero) could have an equivalent to an 350 engine "agility-wise" despite having the stock 320 engine or even the 400 max XL engine that it could equip.
That's how you're supposed to look at it.
Ideally, something most agile "may" get adjusted down (except for Lights, because they actually do need the added agility) but something very non-agile like a Highlander or even the Vindicator that has really bad engine caps (outside of the 1AA) can be closer to par to a more agility mech (like a Cyclops and a Phoenix Hawk respectively). Something like a Mauler that would naturally be pathetically not agile (at best running a STD 325 engine, due to the vast amounts of tonnage invested in dakka) would at least be able not feel like a total brick driving it.
It's not that complicated, but people are trying to make something out of nothing and thinking it would just be straight nerfs (or somehow a sub-250 engine would somehow become more popular when that's genuinely not the case - top speed is actually important when you're increase speed by like 10kph or more)... where it's actually a lot more complicated than that.
Edited by Deathlike, 29 April 2017 - 05:09 PM.