So what if we gave normal MM games tonnage? Nowhere near as high as 240 or so, but around 100-150. And this would be different if you were going to drop in clan OR IS mechs. So say you want to pilot clan mechs, you could have, say, 110 tons to work with. Do you want to take two 50-55 ton mechs? Go ahead. Do you want to take one Direwolf? Ok, but once you are dead you are dead! Now an IS player would have 140 tons to work with, 30 tons more. That means they could take a Mauler and a Hunchback/Centurion etc. That's quite the combo, something a player dropping with two Stormcrows would be on equal footing with. Please note that these tonnages are just examples, and could be anything. But I think it should definitely be significantly less that CW tonnages for the sake of keeping the games lighter and faster - that's part of the difference between CW and MM.
Overall, I think this would benefit the game a couple ways:
Balance, obviously. Going this way means you can approach balance a little more from that lore perspective. We will never get zellbriggen, but this way we can show the tonnage/numerical superiority of the IS vs clans. A IS assualt and medium could engage 2xclan mediums. Would we get rid of quirks? Probably not completely. Some very, very mild quirks here and there. Say an Awesome gets mild heat reduction quirks, or the Hunchback gets a few structure points to the hunch. Nothing game changing, just quirks for fun, really, and to showcase the lore of a mech. Provide some context for it.
Match Momentum. Probably around 60-75% of the games I play are terrible one-sided stomps. I'm talking about the victor only losing 2-3 mechs. This is because with each lost mech the amount of firepower a team can retaliate with drops dramatically. With a single respawn or two, if you use a lot of light mechs with your tonnage, that means your team is punished less for an early loss. I have the bad habit that once I see my team is down by 2 mechs and the enemy hasn't lost one that I basically give up. I know that is wrong, but my prediction is right more than 75% of the time. If there were respawns there could be a greater chance for a comeback. Also, I think it would spread combat out over greater distances, rather than two blobs crashing into each other. Battles become more frantic, as reinforcements arrive etc. Speaking of reinforcements...
Role Warfare. Lighter mechs, after respawning, will reach the battle faster. In CW you don't see a lot of light mechs too often, but with 110-140 tons I think you could see more, because a light mech's speed becomes more valuable the more long and dynamic the engagement is. In giant blob fights the light mechs can be dispatched by an unlucky/lucky salvo. If the combat devolves and spreads out, having a light mech or two in your stable helps you and your team out a lot. That's wonderful if the enemy is respawning in a Mauler, Battlemaster, Zeus, but your Firestarter just arrived to support a 2v2 engagement in the middle of the map - who is that fat assault mech helping right now? This also applies to medium mechs. In the lore, medium mechs are your bread and butter, so I think we could see the % of weight class shift towards the lighter side. The burden of skill on light mech usage is lowered because, wow, you have two more respawns at 110-140 tons! That dude in his assault/medium combo only has one respawn! Work with your team and you are golden. Respawns also increases the value of scouting. Find out where the enemy is coming from. Find out where there reinforcements are coming from. Find out if they are trickling in one at a time or coming in waves. And if your scout mech gets wrecked, well, you have a respawn. Maybe a respawn in a much heavier, chunkier mech. Game on.
Fun. It's more fun to have an extra life, I think. Round a corner into an enemy lance? Game over. At least this way you have a second chance. This can also make the game more approachable to new people. Speaking of which...
Player Retention. My playing group will get new people trying out the game, and our recommendations at the moment typically revolve around "get a stable of Hunchbacks or Thunderbolts". Why? Because they are solid, simple in build, and fairly cheap. The grind in this game is steep. It revolves around the all-mighty C-bill. It's why I don't mind, because some games lock tiers behind ranks, or have nasty increasing repair costs. In MWO, grinding c-bills sucks, but c-bills are everything. You can play anything, as long as your money talks. MWO is brutally honest in what you have to grind for, but thankfully it is one thing (mech mastery aside, but I never consciously work for that - it just happens with my good mechs). For new players, allowing a tonnage respawn means they can try out more mechs faster, are punished less for learning. Even when it comes to learning maps! Also, light mechs are cheaper! You can fill a stable of light mechs so easily for your 130-140 tons, and then add a medium into the mix, and so on. It adds a better feeling of progression, albeit in an organic way rather than a F2P game where the progression is also a barrier. I bet that if you can get a new player to just elite his first trio of light mechs and really make them his or her own their chance of staying in the game system will increase dramatically. MWO is a confusing game for new players, and something as simple as adding the respawn ability and increasing the gaming value of lighter, cheaper mechs would pay dividends. Literally, in the case of PGI.
I think this is the type of balance MWO needs, and could really flourish. This concept is far from perfect though, and I would love any feedback! I think this would make for a happier, healthier community and I believe in it. I have serious concerns about the future of this game. Balance has been a constant issue and this could finally get MWO ahead of its balance issues instead of playing catch-up.
Sorry this is so long. I spent too much time thinking about this. TLDR: add CW-style tonnage limits to normal matchmaking.