EPTAs I type this I'm watching Bisu play an awesome Brood War PvZ. He's controlling two forces at the same time: The main force consisting of Archons, Zealots, Dragoons, and High Templar; the second force consisting of a Shuttle holding two High Templar and an escort of Corsairs.
The Zerg is holding by the skin of his teeth through clever placement of Lurkers and Dark Swarm, along with the occasional plague. Meanwhile in a different corner of the map Scourge and Corsairs are dancing, showcasing both players' mechanical capabilities.
Bisu's forces rally in the center taking map control cutting the Zerg's supply lines and securing his own double mineral base. Both players are on at least 5 bases. Both players have fought more than once. Even now a group of Zealots assaults a lonely Hatchery as the main force takes out another Zerg mining base. Each assault requires preparation, rallying the troops in an attacking formation, then ordering them to attack a certain location.
Each battle takes more than 10 seconds with each player fighting for efficiency. Both armies seem evenly matched, constantly maxed.
Enter SC2. Deathballs seem to be the biggest complaint. Players turtle until their ultimate army. They gather at the center of the map and exchange lasers for 5 seconds until the player who fired more lasers wins. That's the gross oversimplification critics seem to make.
With all the Starbow talk I decided to play a couple of games. My last game was a PvT. I expected the usual Terran mech, with Siege Tanks, Goliaths, and Vulture runby's, not ignoring a map littered with mines of course. I opened up with a Reaver drop, taking out at least 5 SCVs and equalizing the worker count. Another Reaver drop followed in which multiple Terran units were brutally murdered. Then the game proceeded to enter into a 5 vs 5+ base campaign mission in which all I did was try and deflect Vultures, taking out mines and keeping up my unit count as the opponent slowly crept closer with his Tanks, Turrets, and Goliaths. I (over)made some Arbiters to try and Stasis the Tanks when I attack them and my opponent responded with Science Vessels. To my dismay I realised too late that Stasis has a channelling aspect in Starbow and Goliaths tore them to shreds, nullifying the value of my Arbiters. Needless to say I was horribly smashed to bits as I had nowhere near the required strategic mind to think of recalling a ball of units into the enemy base.
Now I think back on the match, comparing BW, Starbow, and SC2. I wanted to come up with my own solution for breaking deathballs. None of that 'make all units faster' business. None of that 'make all units retarded' business either.
Blizzard's solution seemed to be to make fast units that work better on their own, such as the Reaper rework, Oracle, Medivac boost, proposed burrowed roach speed boost, etc. Starbow's solution seemed to be to rely on the innate design of the units borrowed from BW as well as the inefficient worker mining to force players to control different points of the map at the same time. In my experience both solutions have failed.
Before I write why, I first want to reason out why players would prefer a deathball over spreading units out in SC2:
1) It's easier to control.
2) The more units in your ball, the more damage it deals, obviously.
It is better to attack at one place with all your units than to attack at multiple locations with different groups since you are more likely to break the location you are attacking. The only reason to attack another location is to pull the enemy army away from your actual target or to force your opponent's attention between multiple points.
3) Spreading your army creates vulnerable points which your opponent can exploit with his ball, enabling him to take out your army bit by bit rather than a single fight taking everything out. Games where one player decides to split his army have a higher chance of ending in a base race.
4) Harassment specialists take away from the army supply, reducing the overall army strength.
Why do I think that SC2 and Starbow failed to address the deathball?
SC2's solution still leaves the core units forced to move in a ball. Everything else is harassment. Engaging in a fight rarely leaves room for retreat unless special abilities are used.
Starbow's solution emphasises the defender's advantage coupled with a different economy. A defended area requires more opposing units to break. At first glance this isn't a bad solution, it's how it should be. Defender's advantage should be part of a strategy game. But it doesn't solve the deathball. It will only lead to long macro games where each player greedily rushes to his dream composition, defending his bases with the bare minimum as he harasses the opponent with whatever fast units he can make. This will still lead to an eventual climax where both players decide to move out with their army and exchange damage. The players will never take units away from their main force unless they intend to stall for an eventual push.
I'm assuming the hope was for players to contest new expansions. This doesn't solve deathballs on large maps, or turtling. The games will be 30 minutes long with both players sending out their harassment units until they're maxed.
What breaks the deathball?
The Roman army had multiple levels of units. After the Marian reforms, a Roman legion consisted of 10 cohorts. A cohort consisted of 6 centuries. A century consisted of 10 contubernium, the smallest tactical unit. The contubernium consistead of 10 men, one of them the chief of the contubernium. In total a legion consisted of 600 of these units, each unit with its own leader. Each unit able to act on its own.
Now imagine having only one commander and no structure. The first thing one should realize is that units would be a lot more single minded. Another thing would be that it would be impossible for all the troops to be coordinated. More troops does not mean stronger army. It means worse organization, less efficiency, more food wasted, more wages paid.
In SC2 and Starbow, armies do not lose efficiency as they grow, they gain efficiency. 24 void rays can kill more units per void ray than 1 void ray would.
In BW, a bigger army did not necessarily mean a better army. Having 24 mutasis not as much better than 12 as having 12 is than 6. It's more difficult to control 24 mutas, more difficult to retreat in case of a failed attack, and more difficult to advance with good organization. It's easier to micro 12 mutas than it is to micro 24 mutas. It takes more time for an army of ground units to take an offensive position hence it's more efficient to just send small chunks of that army to be more efficient on their own since they are easier to control. In SC2, this is not true.
What does this mean?
My opinion is that limited unit selection is the only thing that can reliably break a deathball. Defender's advantage is automatically granted to the player whose units are already positioned well.
I'm just gonna trail off here since I'm very sleepy but I hope what I said makes sense, and if not, I look forward to the rebuttals.
