There is more to improving things than EFFORT. Don't forget UNDERSTANDING.
Every bullet attribute contributes to a sum. With this, it would very rarely be appropriate to separate bullets based on any single attribute(weight, meplat diameters, etc), or even generalized grouping of some -but not all(like base to ogive comparisons).
If a bullet weighs a molecule more, while it's meplat is a smidgen larger, or it's diameter is a c-hair smaller at same length bearing, it may be perfect w/resp to this or that result standard. So separating here based only on weight could be a bad action.
On the flipside, some bullets weighing exactly the same could be separated, because that consistency in weight is failing to counter other variables..
You cannot truthfully declare that normal bullet weight variance makes a difference one way or another, without qualifying this with all other bullet attributes and performance standards in play. All measured, and not assumed.
Ask yourself when considering the separations of primers, brass, powder, or bullets,, what are the chances that unqualified actions would be correct?