Assists in Team SWAT, and Halo as a whole, have a mysterious nature about them. Nobody knows where they come from. No one knows whether or not they are valuable to the outcome of the game. And, most importantly, there is little to no knowledge available that tells us how to get one. Hell, I don’t think I’ve ever been right predicting my assist totals after a game.
The most viable definition that I could find stated something to this effect: “An assist is defined as a shot or multiple shots that drain more than half of an enemy’s shield, which is then followed by another player, either a teammate or free-for-all opponent, killing the enemy before his shields recharge completely.” I have also found a few anecdotes claiming that the shields cannot recharge at all, and a few forum posters who said that there is a small, static amount of health underneath the shields that must be partially damaged in order for an assist to be recorded. Whatever the method may be, assists are not as common as we perceive. The typical 4-on-4 Team SWAT match will yield a player anywhere from zero to five assists (strictly based on my own experiences) although only in a few rare instances have I seen a player record no assists in a game. Three seems to be a common average for a single player.
So after this deliberation, I ask this question: how significant is an assist in the eventual outcome of the game?
The answer, in short, is not that valuable.
In this match, the losing team actually had three more assists than the winning team.
Looking at some games I had played, a string of 30 consecutive Team SWAT matches, I did some simple tallying (in order to keep it clean, I omitted two ‘ShWATguns’ matches and one ‘SWATball’ match, bringing the total to 33 consecutive games, but only 30 of which I considered useful). I classified each game into one of two types: a game in which the team with fewer assists won, or a game in which the team with the most (or tied) assists won. I could have also eliminated games in which assist totals were tied amongst both teams, but I chose to include it because common wisdom would lead one to believe that an assist is better than no assist, and thus should help chances of winning a game.
>= assists record: 16 wins, 14 losses
< assists record: 14 wins, 16 losses
While a difference of two wins is signifcant enough to warrant the belief that assists are indicators of a greater winning percentage, now I will take away the “assists tie” option, so that any game in which each team had the same amount is eliminated (hence, only 27 matches are now considered):
> greater assists revised record: 13 – 14
< fewer assists revised record: 14 – 13
While the sample size is now smaller than before, at the very least it has clouded the view that assists increase chances of winning. However, when looking at this data, it is pretty easy to tell that, if nothing else, assists had little to no effect on the game’s results. Personally, I am ready to concede my original notion that assists were valuable. I have changed my tune to such a great extent that I would not be afraid to call them totally worthless. I can’t say how many times I’ve been in a firefight with another player, only to be killed and have a teammate suddenly appear and pick up the kill, nulling the spread. I call this “cleaning up”, as it doesn’t directly help gain a lead, but it does help keep pace with the other team. Ideally, I would have killed the enemy and my teammate and I would then proceed together. Or, in one of the rarest Team SWAT scenarios, I would have been able to hold off the enemy long enough for my teammate to come kill him. While this isn’t something you would see in the postgame statistics, it is valuable nonetheless, as it gave our team a net gain of 1 when, for all intensive purposes, we should have “cleaned up” only and had a net gain of 0. So Bungie, what is an assist – what is it really?
-Joe
Posted in Analysis
Tags: 360, advantage, Analysis, assist, assist definition, assist value, bungie, halo 3, headshot, kdr, kill, mvp, online, ranked, ratio, spread, statistics, stats, team swat, xbox 360