Thanks for your kind words and suggestions, everyone! There's too much here for me to address individually, but I'll try to hit the most salient points.
The biggest deal-breaker seems to be the shop, followed by the population growth suggestions that I made (artifacts, I admit, of not being able to tell whether the designers want city sprawl or few cities). On top of that, CdrRogdan made some points I want to address.
The shop: The problem with the shop is two-fold. It's not that outfitting your champions is bad or even that being able to buy or sell at any point in time is bad (though I, personally, would prefer that buying and selling only be possible in one's own territory, I can understand why others would disagree). The first problem, as we all pretty much agree, is that you buy and sell at massively inflated gildar prices at the shop. The second problem is that unlike the rest of the game, it has no connection to any of your other resources. In the current game, gildar can somehow take the place of all other resources, but only for the special people, and that bugs me. In real life, I can't hammer gold into a [functional] sword; I have to use iron, and that has to come from my stocks.
While I said "kill the merchant," an admittedly extreme phrase, I didn't mean that it should be impossible to outfit your champions or buy stuff for them. I meant that the moneybags go away ("the merchant" versus, say, "the blacksmith") and all purchases point back to your resources, as if you were buying (or selling; Malsqueak made an excellent point I'd been thinking about, specifically that you could break down items found in your travel into their components and add those to your stores) that item for any of your rank-and-file troops. If we don't make it require city production, that breaks immersion slightly, but it's much, much less so than the current "resources from the ether" paradigm.
For that matter, I could even see still allowing you to buy items, so long as the buying prices stay where they are and the selling prices get way, way lowered. Such an event would occur on the basis that your merchants might have underground connections, and obviously not all of your metal/crystal/whatever is flowing to your kingdom's coffers. It would be akin to rush-buying, except you're black-market buying instead. So long as doing this would be an enormous financial burden to your kingdom (20 turns of net tax receipts for a horse... or I could just use one of my horses), which requires fixing sale prices, and you could only do it every once in a while, it's not a deal-breaker.
Lugging it all back to town and even using production to build the items are areas where I tend towards verisimilitude, but they're not critical. The critical thing is to get resources into the buy/sell screen, and that means getting the gildar [mostly] out. I don't care if they keep the merchant screen and just repurpose the buy and sell options; I just care that they eliminate the magical leprechaun with the unending coffers of gold and replace him with something more reasonable.
Food: I was well aware that this was the weakest part of my post, and I actually deliberately threw in a grab-bag here because of the problems I described in the "City Growth" section above it. I tried to address both the possibility that the developers wanted few cities and that they wanted many of them. Actually, my favorite suggestion is elimination or reduction of upkeep costs for stationed units based on the food level of that city. This would make it wise to create the most powerful, expensive unit you can and station it in your city-- you're not paying for it, and now the city has some real defense. What a difference from the deserted, undefended cities of .86. Aside from that note, there have been good suggestions all around, and I'll let those stand on their own merits.
CdrRogdan, no worries about bubble bursting; you haven't managed it yet 
And of course, I know you're not being an ass. I hope I've addressed (if not answered) your objections about food and the shop above, but I want to look at minimum damage (what I call "ping") and initiative.
Addressing Lethality
First off apparently there is no such thing as 'ping' damage. Although initially I thought it worked this way as well =/. Damage cannot be reduced below 1 because it is a multiplier of sorts. Additionally, the biggest issue isn't with low damage always hitting for minimum as this can be solved with more hp, but that higher damage attacks become impossible to negate without ludicrous ammounts of defense.
I'm afraid I don't follow your first two sentences. What I call "ping" is really just the minimum 1 damage. I'm proposing lowering that to 0 (so with 9 clubmen, I might hit for 0x9 = 0 damage) and then allowing the whole unit to get a consolation prize of 1 damage if they end up hitting for 0. Regardless of how the effect is applied, the result would be the same.
While it's possible to solve the issue with higher HP, that requires me to have higher HP to bring to bear, which means either experience (which I implicitly disavowed in my post by referencing unblooded troops) or higher production costs, and I'm already spending a ton of production on armor. Shouldn't that armor actually protect me from 6x my number of cavemen, especially if I'm paying 12x the upkeep or more? I don't want to have to dump even more production and upkeep into giving myself magic items and HP-raising traits. HP at production time should be a choice when facing primitive troops, not a requirement. "Or" rather than "and". In other words, the designers need to ask themselves, "All else being equal-- HP, civ traits, and the like-- what is the minimum number of the most primitive troops that should be able to overwhelm a single inexperienced soldier of the highest-tech troop in the game?" My understanding is that the answer to this, currently, is "six". If this number doesn't fit the designers' vision, they need to go back to the drawing board.
Agreed about the attack/defense imbalance; no arguments there.
Addressing Initiative
Your notion of making spells use a percentage of initiative would require a different method of calculating the system, and would likely possibly illiminate the ability for other casters to counter the spell. The problem with insta-win spells right more about impulsive and haste not working correctly and less about high initiative. The problem with weapons using initiative modifiers isn't just that it changes unit move speeds, but that it makes calculating the appropriate weapon to use with other items and initiative modifiers an algebraic adventure. Keeping weapons with initiative modifiers rather than give them unique functions or abilities is not the way to go here. Nor is using some strange multiplier of initiative for using a 'wait' command.
Sorry, I assumed that spells that are cast at the end of the initiative phase would still be cast at the end of the phase, allowing time to counterspell, but I didn't make that explicit. I still can't understand the rationale behind the extraneous "extra turns" function; they already have a method for delaying later action, and it's called "initiative".
We agree that the problem with weapon initiative is that it throws entire systems into disarray. In any case, I did approach the post with the idea that they still wanted weapons and items to modify initiative, and I didn't really see a cogent argument in your post against it. Which is fine; it may simply not be your cup of tea. I do think that options of varying initiative values add a desperately-needed bit of variation to the battles, but that's just me.
Lastly, an initiative multiplier on "wait" isn't really strange. The Heroes of Might and Magic game that first implemented continuous initiative (5, I think it was?) had it; the "wait" command made your turn come around more quickly than attacks, defense, or spellcasting, which makes sense; you aren't actually acting, so presumably, you're, well, waiting for a better moment.
