What have they done better than anyone else? Mech graphics... the mechs look gorgeous, the maps are pretty darn good giant shrubs not withstanding, and they are creative in their variety.
Where has circumstance bit them? That cryengine bug that was corrupting the UI sounded nasty, and having to implement HSR from the ground up apparently cost them months. Odd that they didn't do due dilligence on the engine and know this was going to be a huge issue, or maybe they did and there just are no engines available that come with competent net code out of the box.
Where have they shot themselves in the foot?
A few places - let me give a counter example

That is a screenshot from a game that was developed over ~1, 1.5 years back in 2000, 2001. The playable beta existed between summer and December 2001. I loved that game

1) It had community warfare as soon as I joined up. You picked a house and you fought for that house. Holding more territories caused your passive salary to go up. Each world had 100% health and every won battle on that world got 1% transferred to your faction from the faction you defeated. If no one would drop against you, every 60 seconds you would take 0.1% passively, so you could wear down the planet even if the enemy chose to avoid fighting you.
2) Balancing was done by the players.... every combat drop had the two teams line up where they could see each other and only 'ready up' when both teams were satisfied that the drop was fair. No morass of ELO, tonnage, 3 mans, 4 mans, etc. Just players on each side looking over each other's teams and not dropping unless tonnage was matched within 5-10 tons.
3) Organic progression. MWO seriously lacks any feeling of progression, while still feeling very grindy. Grind can't be eliminated but it can at least feel like you are accomplishing something. Here though your first mech can be an atlas using your cadet bonus. I appreciate that they are trying to do something new here by making every mech have a place on the same battlefield but that is trying to square the circle. In MPBT3025 the solution was simple yet elegant. They made it so border worlds could only supply light mechs, so combat there was 4 lights against 4 lights. Every single pilot in the game was a deadly light pilot since you spent >50% of all games on border worlds just the way they flowed back and forth. Assaults on the other hand took months to earn the money for, but they could only defend regional capitols - 1-3 territories per house. They were a rare sight but ******** they owned the battlefield.
4) Electronic warfare - The best games are those that you can learn the rules to in an hour, but that take a lifetime to master. (e.g. chess). Games where the rules are convoluted and opaque and re-re-re-revised every few weeks are very hard to get into. Electronic warfare in general and ECM specifically are pieces of equipment where even having months of experience with and against it I still don't think I fully grasp all the different ways it interacts with artemis, tag, NARC (bahaha... cough), BAP, other ECM, target range module, info gathering module, UAV.... you get the idea.
5) Role warfare. Well the game is launched and there are still no role specific skill trees. Not only that but some of the skills in the placeholder skill trees are apparently non-functional. Pinpoint does nothing I believe, and I recall fast fire along with the AC2 causing ghost heat headaches. Additionally the rewards system seems heavily heavily tilted towards mechs that dish out damage. Spotting, scouting and base capping all seem underwhelming in terms of their rewards. Granted it is hard to properly balance these but role warfare being viable requires that they be balanced. Currently we have a U curve of mediocrity where mediums are not fast enough to shine in conquest, nor are they powerful or tough enough to shine in assault.
6) Customization - I love customization. Absolutely adore it. However from a game play point of view it's a nightmare. It means that weapons have to be balanced as though they are going to be boated and abused in the worst ways possible, and tiny changes in balance can and will cause massive swings in player behaviour since everyone can switch weapons at the drop of a hat. Also in a real time game where different weapons have different targeting characteristics boating is almost inevitable. Take the example of the AC/20, the medium laser, and the SRM6. On tabletop these all have the exact same range profile, and therefore a mech armed with one of each will find it exactly as hard to hit a given target with all 3. In real time though they are a nightmare to use together effectively since they all travel at different speeds, and cool down at different rates. It is generally easier to hit a target with copies of the same weapon than to aim at 3 different places and for 3 different durations to make effective use of 3 different weapons systems.
7) Community warfare is the worst offender in terms of proposed complexity. Lone wolves, loyalists, mercenaries, clans, lord knows who else, vying for contracts (or not.... can clanners take contracts? Can loyalists? Can lone wolves). Garrisons, dropships, bases, planetary production bonuses, black market..... Even having watched the video from the launch day a few times I am having trouble getting my head around whether or not wolf's dragoons are endgame content or not, or if there's any point at all to playing as a house loyalist.
It reminds me of a quote from Bob Colwell, the former chief chip architect at intel who was behind the P6 (pentium pro, grandfather of most of their current processors) "If you choose everything as your targets, you won't hit any of them."
Given that they don't have the human resources to have started CW until now, and the apparent enormity of the task, I'm substantially concerned that CW will suffer further significant delays, be chock full of convoluted mechanics, and be another example of shooting for an "AAA" gaming system that will turn out as another C, C- as the game itself has to date.
I really want these guys to succeed, and I know I couldn't do any better, so let's get those canards out of the way, but having watched the pace and style of development for >18 months, I really do think the quality of the final product would be better if they focused on a tighter, more achievable and less interconnected series of goals.
1) Lobbies....
2) Territory control
3) membership in the great houses... no merc units, no lone wolves, just the houses.
Then go from there.