I'm not making any declarations or anything, just consider this as a bit of a 'hiccup' in fitting an accelerating universe into my still-solidifying understanding-model of said universe.
What gives me pause is the relation between 'explosion mechanics' and gravity.
Even in a high-gravity environment such as ours, at the 'epiforce' of an explosion, where the outward expansion of whatever combustible has either just reached its 'maximum combustion' (where the most fissile material is 'lit' at one point in time) or overcome whatever contained it, it will project any contained or proximate material at the highest speed, but from there after, the energy will drop and projectiles will be ejected at a slower speed, and so on and so on until the explosive engine's combustible is exhausted. Those 'epiforce' projectiles will, of course, travel the farthest from the explosion epicentre.
Now take the same model and transpose it into a 'no-gravity' environment.
Again, the 'epiforce' projectiles will attain the highest velocities, and those projected after, slower ones (et cetera et cetera)... but this time, there is nothing to slow those projectiles down (well, there is, but I'll get to that in a second). So the 'outermost' projectiles will be travelling at a much faster velocity than the later 'inner' ones, and this relative 'difference', as the velocity of all projectiles is constant (without considering other later factors), will grow over time. Already we have a model where, from the perspective of the innermost projectiles (pretend that they are standing still), the outer projectiles are accelerating.
That's fine enough on its own as a 'small' model, but the universe is hardly that, and there's the enormous gravitational forces that it contains to factor in.
It would be the slower 'projectiles' closer to the 'big bang' explosion epicentre that would be the first to succumb to mutual gravitational attraction and form stars (then planets). There's also the density of the ejected matter to consider at each point in the explosion (but I don't have either the math nor engineering/mechanics knowledge for that), but I would think that, even if the 'rate of explosion' was constant (which it is most likely not), the faster-velocity material on the outer rim would also be more disperse (over a wider circumference), thus slower (and less likely) to accumulate into larger masses.
So, one way or another, toward the epicentre of the (former) explosion, we would have a 'core' that would be increasingly denser and have a higher gravitational mass, and, logically, a centre of gravity as a whole.
Now factor this onto those outward-travelling 'projectiles'. The universe's gravitational pull on these, I would assume, would follow the gravitational constant (yet one increasingly polarised on the projectiles as they travel further away); the math here, again, is complicated (for me), as one would have to factor in velocity, the gravitational force relative to it, and its gradual diminishing over time (as the projectile grows more distant). Yet, all the same, in all cases, we would have a model where the projectiles towards the explosion epicentre would slow each other down much more quickly than those towards the outer rim. So, here, the centre of the universe is slowing at a much faster rate than the outer rim, which may give the illusion that the universe's expansion is accelerating when, in fact, it isn't.