by Andrew “Hamfist” Gross
One aspect of this tournament stood out to me so much that I feel like I can focus on it to the exclusion of all else; namely, the implications of the improved survival movement algorithm.
For those who haven’t been play testers for the last several months: until quite recently—less than a month ago—survival mode gave you the same armor and regeneration benefits that it does now, but the algorithm it used for deciding where to run didn’t work nearly as well. Usually, your agent ran to a corner and just shuffled around in it, as their attacker picked them apart. Every now and then a third agent would wander by and draw the attacker away; but even when that happened, the original attacker was often replaced by some other agent. Players tended to either set their survival threshold very high, so their agent would flee the area when they were still quite healthy (basically, as an instruction that says, “Avoid Combat”); or else they would set it quite low, as a last-ditch “Hail Mary” to help their agent’s chances of getting lucky and surviving if things got really bad. It rarely made sense to enter survival mode at anything other than one of these two extremes.
However, Chaos Agents is currently still pre-Alpha, which means that adjustments and new features are being added on a regular basis. When they improved the survival movement algorithm for Agents, such that they now flee in favorable directions, run out of corners almost immediately, etc., it changed the game in multiple ways besides just making it less irritating to find yourself with low health and in survival mode.
If there’s one aspect of my play during this tournament that I take pride in, it’s recognizing the utility of the improved survival mode, and using it often. Concretely, this meant that on more than one occasion, I would be committed to Hamfist attacking for the rest of the game, and Hamfist would have just burned an opposing Agent down to a sliver of health, while he himself would be at 50% health or even a bit higher; yet I would not continue in attack mode, and instead would switch to the survive, setting my survival threshold at 80% or so.
This ran the risk of some other Agent “stealing” my kill; but Hamfist himself in Attack mode is stealing other Agents’ kills, too. If Hamfist continues pursuit of a fleeing Agent who’s at maximum range, he’s only going to get two or three shots in for the whole turn, and is at the mercy of his target for where he goes on the map– winding up in combat with other agents is a common outcome. Meanwhile, he’s not getting the bonus life regeneration that Survive mode brings. With Hamfist’s own Life on level 3 or 4, he can go from 50% to 80% in a single turn while in Survive mode. So by being patient and not having Hamfist pursue an enemy just because I’d “earned” the kill by him doing a bunch of damage, in compensation I got Hamfist’s health high enough so that he could duke it out with a fully healthy opposing agent that might come along— and this exact scenario happened more than once. Not only that, but because the survival movement algorithm is so much better, that Agent that Hamfist didn’t pursue wound up living every time, or maybe all but one time, so he was still able to collect the kill later.
Using this tactic, Hamfist racked up 16 KOs in the two preliminary rounds and the finals, winning Brawlmaster all three matches and Survivalist twice (fortunately the Finals was one of those times). Players who have regularly competed against me when I’m using Hamfist would probably tell you that he deals out plenty of damage and can be difficult to deal with once he gets rolling; but they’d also tell you that he’s nowhere near invincible, and that he gets killed without winning any victory conditions plenty of times. While I’m sure that luck played an important role in obtaining these tournament results, I am also quite sure that without employing this tactic—which only became an option once the survival movement algorithm was greatly improved, and which I only realized the potential of within the last ten days or so—Hamfist would have been KO’d before securing a victory condition in one, two, or even all three rounds.
I don’t think I’m doing an adequate job in conveying just how important the additional options that the improved survival movement algorithm gives attacking agents are, and how much more often I really needed to think about what to do, rather than feel like my course was set and I just needed to go along for the ride. It’s as if the game used to have Collect and Attack, and then this third option that was really just about trying to avoid combat or giving some hope in a desperate situation; but now it has three legitimate modes that you need to consider each turn. (Yes, collecting is a legitimate option, even when you’re not vying for Shardmaster: in some situations, you can look at where shards are on the map, and select “collect” as a way of moving your Agent to that area.) The survival option has become a valuable weapon in an attacker’s arsenal, when before it was an option that only an agent on the losing end of combat would consider.
Oh, and a shout out to Popularium’s engineers, who kept working past the start of Gen Con and through the weekend in order to get obstacle avoidance in place for the finals; and to their Q&A lead, who made the decision to use it with only the testing they could squeeze into a single day. In this case at least, fortune truly favored the bold, and it was a delight to watch Hamfist running around obstacles—even when he was being pursued and that meant taking an extra shot for the turn. What a treat to have such a neat feature debut in the Finals!