Upgrade Wars - Thoughts, ideas and criticisms needed!

I like the concept. It stands out to me. Giant robots battling is a theme that has a pretty broad appeal. Definitely a game that if I saw it on the shelf I would pick up and look at the back of the box before going to bgg to watch or read some reviews.

The mechanics are simple in a good way. The game is accessible. It does not involve too much math and is free from result tables. Gamers understand the mechanic of discarding cards to pay for other cards. There are a lot of good elements here and the game is promising.

Several places where I think you could improve the game:

Not sure why thematically you went with "printers" instead of factories. Factories are intuitive with the theme.

Is the 4 player game a team game? The layout suggests that some players can move from their printer zone directly to their neighbor's printer zone. That fine if they are on the same team, but bad if they are competing against each other.

You offer too many options for set up in the set-up part of your rules. The first time I read a game's rules I want to know how to play my first game. I'm not ready for options like that. I recommend listing only the random set-up options here. Then at the end of your rules, provide "advanced" or "alternate" rules for set-up where you describe the drafting rules.

How much risk is there that the program pool will not have all of the needed elements? What if we randomly select only upgrade cards?

Related, what is the risk that one player's personal pool is clearly better than another player's pool?

Mechanically speaking, I don't think you need a "build goo" phase at the beginning of each player's turn. If you want to build or upgrade a bot during the print phase then just discard the cards then.

Your rules on multiple war bots attacking at once are, I think, highly favorable to the attacker. Let's imagine I have two war bots with an attack value of 4 and I want to attack a warbot with a defense of 2. With your rules my attack is at 4+4+1D6 against a defense of 2+1D6. Attackers worst result is 9 and defenders best result is 8. No matter what attackers will win and inflict 8 damage (attacker always inflicts full attack value). If instead each attack was separate, then it would be 4+1D6 vs 2+1D6 and then 4+1D6 vs 2+1D6. Now one of those attacks will probably succeed and deal 4 damage, but both succeeding is less likely. I'm not sure if I am making sense on this point.

One of the major lessons to take from Sentinels of the Multiverse is that damage trackers should be in the box. If these are PnP rules, then that's okay, but you should not ask players to provide their own way to track damage in a game they paid for.

I'm not clear on what the construction bots do. You use them as goo throughout the example game, but any card can be used as goo, so I'm not sure what else a player could do with construction bots. Or are construction bots the only thing that can be goo? Are they the only way to pay for programs? If so that needs to be much clearer.

I hope my feedback was helpful. Good luck with your game!

The Goo icon from the cards should really be used for the card backs as well: the large G makes no sense when you have an icon for Goo already.

On the matter of Printers and flavour: I really like the Printers! But they need to show through in consistent conceptualization. Right now there is too much of a 'chunky robot' feel to the cards: things like modular platforms and missile system upgrades give a feel of Lego-ing parts together. The elements that I really like are the things like Nanoswarms, Goo, reactive armour, synthetic muscles: the more sleek, cyborganic robot. The programming upgrades are somewhere in the middle. These kind of naming upgrades to flavour can be made before any art upgrade, and should be, as they will directly inform the art. For example, "Missile Systems" could be "Embedded Missiles", "Projectile Tubes", or "Plasma Darts": they still give the ranged weapon message, but have a much more fluid, oozing, plastic feel.