Skip to content

Calculating unofficial stats

dugman dugman
in General Discussions

I was messing around checking out data on a stat aggregator where people unofficially post their game results and it tracks everything. (I won't post the link because I believe the site includes Steam API as an option and the mods have asked not to include links to sites that use Steam API since it's not officially supported. So I'll just stick to talking about the statistics I'm seeing.) Mind you, even though the site has thousands of games in the data it's still a biased sample since, for instance, clearly only people already pretty invested in the game will take the time to track their stats on sites like these. But it's still interesting to analyze.

  • Average overall kill rate pre-patch from May 27 through June 26 was 49.06%
  • Average overall kill rate from June 27 through July 26 (as of me typing this) was 51.43%. That period includes about 7.5 days worth of post-patch data. If you assume for simplicity the overall games per day was roughly the same in both periods, then that works out to the post-patch 7.5 day kill rate being about 58.86%, which is a 9.80% increase.
  • When I calculated this a few days ago, the estimated kill rate increase was about 16%, so there appears to have been a drop off on the increase. This is partly due to variance in the data, but also probably due to survivors now getting used to the new systems and loadouts they're using. If I had to guess I think the final kill rate is going to end up around 55% or so, that's just a guess though.
  • Doing a similar calculation, it looks like Dead Hard usage has dropped from 44.7% to about 13% post-patch. So still being used but a significant drop
  • Off the Record has skyrocketed, it was showing at only 0.34% usage (almost zero) pre-patch to being about 16% usage this week. That's not "almost every survivor is using it" but it's still a large percentage for a perk.
  • Sprint Burst has also seen a big rise. 8.42% pre-patch to about 21% post-patch. I'm expecting this to be the most popular exhaustion perk at this point.
  • Prove Thyself is also popular. 11.7% pre-patch to about 20% post-patch. I'm thinking this is due to the reaction to the longer gen times and killers experimenting with heavy regression builds currently.
  • Interestingly BBQ and Chili didn't see much of a decrease after the patch. It dropped from 34.6% pre-patch to 31.7% post-patch. I do know a lot of killers liked using BBQ and Chili mainly for the aura and treated the bloodpoint bonus as just a freebie that came with it, so apparently those people are still using this perk.
  • Corrupt Intervention saw a dip from about 22.3% pre-patch to about 14% post-patch. That's not surprising since the perk is still just as useful as before for the killers that have particularly slow early games (e.g. Trapper, Hag, Meyers) but is now less useful for killers that can expect to get a down within the first 30-50 seconds of play.
  • Monstrous Shrine (now Scourge Hook: Monstrous Shrine) stayed about the same, going from 1.1% to about 1.3% . That's still an increase but it's definitely not seeing a ton of use still, maybe because people see it as something to throw on in builds with 3 or 4 Scourge Hooks but not something to run on its own.
  • Iron Will has taken a hit, dropping from 25.3% pre-patch to about 9% post-patch. It's still being used but not nearly as much as before.
  • Thanatophobia is (no surprise) seeing more use, going from about 12% pre-patch to 19.5% post-patch. Some of that might be short term as people experiment with the new flavor of the day build of Legion/Plague with Thana, but I'm sure most of that increase will stick around.
  • Save the Best for Last went from a pretty decent 8.4% pre-patch to a really solid 15% post-patch. It's not the most used killer perk still but it's pretty popular. Not surprising since pre-patch it was already decent and the patch indirectly buffed it to a point that it might be slightly overtuned now.


The post is getting a bit long so I'll cut it here. Anyway as I implied earlier, take these numbers with a big grain of salt since they're based on voluntarily donated game data. Plus I'm having to make weighted estimates based on two week and four week date ranges that include both pre- and post-patch games in them. That said, a lot of this seems to fall in line with what I kind of expected in terms of winners and losers in the perk list and in the increased kill rates people are seeing.

Obviously this data says nothing about stats by MMR since those ratings aren't visible so there's no way to track anything from public data involving, say, kill rates by MMR bracket. It'll be interesting to see if eventually the devs do another one of those linegraphs like they did for perk usage by MMR for their perk update blogs, I'm curious to see how those graphs evolve over the next few months. 🙂

Comments

Categories