... but it might be worth forgoing more of the subtle strategy for embracing the more luck-based gameplay.

Let's bring up an example: trading card games like Magic the Gathering, Yu-Gi-Oh!, Hearthstone, etc. have similar concepts to this, and work pretty well. Yet, it's pretty obvious from how many quitters and people who have no idea how to play that this ... really doesn't, except for a very certain niche.

So, what's one main difference?

In the TCGs, when you lose, it doesn't take fifty years for you to officially lose. Oh, wait, there's ways to play from behind? Yeah maybe when you're mildly behind, but when you're constantly having units pour into the mid lane and your opponent just gets bigger and bigger, it just gets completely unpleasant to play and usually you don't even have the illusion of countering their army because you didn't roll a unit that can actually turn the tides with the resources you have.

Overall, it's just miserable to be on the losing side. And it's ... really not much better to be on the winning side, because it feels pretty hollow and unrewarding and the game just won't end.

So something I think will help? Take out the lives of the god, remove the ability to upgrade his health and regeneration ( or remove upgrading him, period ), decrease his survivability substially in the first place, and perhaps give every unit a blanket movement speed buff so they get to the portal and god quicker.

Yes, I'm actually suggesting making it so that you lose faster. The thing about losing faster is that it lets you retry a lot faster, and allows you to play with a new loadout faster, and play against a different loadout faster. Reducing the iteration time between periods of play really feels like it would do the concept of random unit loadouts a lot better that the current, overly drawn-out length the current version has. Perhaps income, unit cost, or something should also get a major rework so the late game options are actually appropriate for the new late game, or instead of removing lives the game resets back to its starting state upon a god losing one, except with higher base resource numbers so there's still a sense of going from a tiny army to a larger one full of elite units. The latter approach would be akin to how TCGs handle multi-round matches, where losing badly at first isn't too big of a deal because your opponent still has to beat you two more times before they actually win.

These are pretty big changes, and with them a bunch more changes would have to come, too, I'm aware, but the current vision of this is honestly not very fun and often miserable to play. It's also really hard to jump into, and doing so usually results in a really long time before you have any time to consider what went wrong. Granted, this concept isn't very newbie friendly in the first place, but it still feels like it would be a lot better if it reduced the barrier of entry.