...But for a simple game, as D&Dnext is meant to be (at least at its core), then these other options are less likely to be available. In which case, I think that the melee/magic gap for wizards probably should be kept closer, at all levels, to what it is at first level, and at first level should probably be brought closer to AD&D than to 3E/4e. The fact that a fighter is a skilled combatant who only improves with levels can be reflected in other ways - like manoeuvres, for example, or doing close bursts with weapons - rather than in sheer bonuses to hit.
I don't have a strong preference on what the exact gap should be for D&D Next. I'd probably have a preference for how I want it to be at my table, but ... Whatever the gap, it should be relatively consistent across levels in the sheer bonuses. Whereas, I'd expect the other, more interesting stuff to diverge steadily as a character went higher and higher in a chosen niche.
Then, let's say for sake of illustration only that the gap is set at 8 for attack bonus. That is, unless you just totally go off the reservation (not just a 10 in Str but a nasty negative, or usiing a weapon that you aren't proficient in, and thus get some kind of penalty, or whatever) then you'll be within 8 points of some character fairly dedicated to a weapon, like the typical fighter with his longsword and shield. (There might be some problems with ability mods in this part, but then I'm assuming a new system here that can work those out.)
This is not including conditional buffs on the fighter from the cleric or anything like that. It's what the fighter uses when he is smashing level-equivalent or slightly lower creatures. If the fighter is after something tougher than that, then the wizard should be using magic or finding himself at a severe disadvantage if for some reason that isn't an option--though not as severe as, say, a 5th level 1E wizard out of spells.
Magic items and choice of weapons might also cause more of a divergence, but that's on the wizard. If he has a a better Dex than Str, goes for a short bow, and then tries to have a magic shortbow handed down from the other characters when they get something better, he can keep in the range. If he makes no efforts, he probably can't.
Namely, the wizard has bothered to do the kind of minimal but serious effort that you'd expect out of a guy who mainly uses magic for tough stuff, but runs around with adventurers all the time. Then say the same kind of calculations can be done for damage output with these weapon (using normal attacks) and so forth.
My conclusion is that the wizard should have some minimal magic available--if he bothers--to temporarily buff his own weapon usage into maybe
as much as half the difference (but certainly no more)--i.e. 4 points of attack bonus in this example. This doesn't make him as good as the fighter. It probably makes him dish it out about as good as a cleric, and take hits like the rogue. This magic should not, typically, give him anything else--such as fighter like powers.
Because the point of such magic should
not be to turn the wizard into a fighter and fight the tough battles that way, but to give him a viable way to turn into a minor but decent support fighter in mop up or other oddball situations.
The cleric would have similar magic, albeit more limited, because he has less of a gap to cross. An alternate form of a paladin might have similar magic that makes him conditionally better than the fighter in certain narrow situations (assuming he was conditionally worse in other such situations by default, which might not be a good assumption).
And if that is the range, it should be the same at 1st level as it as 10, 20, or 30 or whatever the max is. Again, assuming that the wizard makes that minimal effort to keep up. Also, the wizard should more or less be stuck with one weapon that has this "decent support" characteristic, unless he puts considerably more effort into it. Exactly what weapon-- crossbow, bow, staff, sword--should be up to the wizard.
This makes such support magic flavorful, but not overpowered, and serves the explicit purpose of keeping the wizard player engaged during those relatively tedious times for him or her when using powerful magic is not a good answer--and presumably items are either not available or also overkill.
Finally, setting the range explicitly this way means that hybrid casters (however built), such as a fighter/wizard multiclass or "bladesinger" type can follow more of the cleric or paladin style, but using some of this same magic. That is, if a wizard can get up 4 points with this conditional and not entirely reliable magic, then a fighter/wizard starts about the middle of the range, uses the same magic almost assuredly, and can just match the fighter on the raw numbers for a time. But not, however, the sheer magic power of the wizard in pure wizard mode, or sheer martial prowess of the fighter in pure fighter mode.
It is indeed starting at the +0/+1 BAB and similar numbers in each edition, then divergent scaling, that has set the foundation for busted buff magic in the first place. 4E went a little too far in normalizing this, IMHO, and thus didn't leave enough room for such buff magic to really fit in this manner. (Of course, given that 4E was designed to avoid mop up and minor fights for the most part, this wouldn't matter as much. Or wouldn't if the design had been fully realized.)
