Keep the scope of the game small. Sticking with the 3025 time period, as has been said many times in this thread, would've done that. Later tech levels and Clans could've been rolled out in later years after the initial game was tested and stable - and successful.
Deliver all five pillars of the game, in a small, manageable style. For example - Information Warfare should've been part of initial 'mech design. The Jenner should've always been a more capable scout than the Hunchback, until the Commando came out, and showed how much better a smaller, more dedicated recon platform could be. There never should've been game 'modes' that weren't part of fighting the Succession Wars. Things like economy, salvage, and Mechwarrior skills should've always been part of a the game.
And by 'small scope' I mean keep everything small until you have it working right. 3025 provides a lot of the limitations you'd need. But, for example, CW could start as simply two factions fighting over a single world in closed beta. Once you manage and perfect the mechanism for conquering one planet, then you can begin scaling that mechanism to cover an entire front. Once you can manage a front, you can scale up until you have whole great houses fighting. *Then* add (and monetize) mercenary units!
Same with infowar. Start out with simple systems like active/passive or detection ranges, and slowly add more nuance to infowar as you grow (again, 3025 simplifies everything). Boost Infowar with map objectives, NPC spotter Boomerangs, etc; get basic infowar working and slowly expand.
*edit* And every drop should've had a role built in for every weight class. Require scouting and reward it. Require skirmishing and reward it. Over time, expand the roles, so that multiple types of scouting/scanning exist (scan multiple map points on some missions, scan every 'mech on others, etc) or skirmish objects, etc.
The game needed some kind of working economy, and not just grinding C-Bills and buying more 'mechs. It needs salvage, and repairs, and improvements, like hiring better techs and a bigger Dropship and other fun things. Again, start very small - just hiring a tech to stand around in your 'mech bay - and build slowly, once initial features are in and balanced.
The applications of 3025/small scope to weapons should be obvious. Tech 1 has SHS and just a handful of balanced systems. Get a reasonable version of TT working, and balanced, with all weight classes and roles working, and then you can experiment with Tech 2.
Map design - every map should've been themed off a famous world or battle site in the Battletech universe. Galtor III. Hesperus II. Quentin. Kittery. Wing. Twycross. Tukayyid. Make maps that Battletech players *recognize*. And make them realistic; so you feel like you are in a real location, riding a 10 meter war machine.
That's not even getting into bad choices of monetization and high costs for 'free to play'. Adding more character to MWO - actual Mechwarriors with portraits, uniforms, skill trees, decals, etc - could've also developed slowly and allowed for true micro-transactions; 1$ decals, 5$ AFFS uniform pack, etc.
Capping the number of 'mechs owned by a single pilot (probably ONE) could've allowed for more micro transactions for more character slots. Character slots could've allowed for a single player to support multiple factions as necessary ... plus spending more $ in micro transactions for more stuff to customize characters ...
Finally, and most important - the question 'Does it feel like Battletech' should've been kept first and foremost in development. MWO, as it stands, doesn't feel much like Battletech at all; yes, the 'mechs look good and have the right names, but the game modes, weapons interaction, heat scale, and 97% of the lore and points of interest of the Battletech universe are missing. If you are going to make a Battletech game, make a Battletech game, not a giant robot shooter with Battletech names.
Edited by Malleus011, 25 April 2016 - 10:53 AM.