Visual Representation of the feedback after a major PoE patch

Just to throw some fuel on the fire here, steam charts only show who's actually logged into the games steam client, it doesn't specifically say this is how many are playing league.

You also have standard players (which is no different to other league comparisons) but you now have to also factor in royale, a fairly popular mode, there might be quite a few people who are purely playing royale that are included in those chart numbers.

"

Thehealyhealy wrote:

Just to throw some fuel on the fire here, steam charts only show who's actually logged into the games steam client, it doesn't specifically say this is how many are playing league.

You also have standard players (which is no different to other league comparisons) but you now have to also factor in royale, a fairly popular mode, there might be quite a few people who are purely playing royale that are included in those chart numbers.

The interesting statistics to derive from steamcharts isn't how many players actually are playing the leageu in absolute numbers.
The figure we're discussing is how many have left the league in relative numbers.

I'm not sure, but I think someone mentioned that GGG had said that 50% of their playerbase played PoE through the steam client.

Now, I'm no statsman, but I do understand representation and statistical signifiance, somewhat. A 50% samplesize for a game where the client is evenly dispersed amongst different segments of the world is most likely representative of the overall trend.

Meaning, if Steamcharts show 50% losses in playernumbers, it's most likely safe to assume that PoE is losing around 50% of their players from the stand alone PoE client aswell.

I'm pretty sure Royale players and people playing Standard also are counted into the Steamchart figures. You still play those modes through the PoE client.

You appear to have really misunderstood my post.

My post was stating that anyone logging into poe client is counted in those steam chart numbers, not just league players. My post was suggesting that Royale may be inflating the numbers of people still playing as they are counted in those charts. So when people are comparing drop off relative to other leagues Royale is a factor that should be considered.

Since Royale wasn't in previous leagues the comparisons of drop off players might actually be more severe than the current steam chart numbers suggest as previously people who did not enjoy the league and didn't play standard would just quit the game and move on to other games, now with Royale mode some of these players may still be playing poe by purely playing Royale.

For example, todays peak was 54k, let's create a total random number guess and say 5% of that 54k were purely playing Royale, they weren't playing the league at all. If Royale didn't exist then that 54k would actually be closer to 51k.

"

Deadandlivin wrote:

"

Thehealyhealy wrote:

Just to throw some fuel on the fire here, steam charts only show who's actually logged into the games steam client, it doesn't specifically say this is how many are playing league.

You also have standard players (which is no different to other league comparisons) but you now have to also factor in royale, a fairly popular mode, there might be quite a few people who are purely playing royale that are included in those chart numbers.

The interesting statistics to derive from steamcharts isn't how many players actually are playing the leageu in absolute numbers.
The figure we're discussing is how many have left the league in relative numbers.

I'm not sure, but I think someone mentioned that GGG had said that 50% of their playerbase played PoE through the steam client.

Now, I'm no statsman, but I do understand representation and statistical signifiance, somewhat. A 50% samplesize for a game where the client is evenly dispersed amongst different segments of the world is most likely representative of the overall trend.

Meaning, if Steamcharts show 50% losses in playernumbers, it's most likely safe to assume that PoE is losing around 50% of their players from the stand alone PoE client aswell.

I'm pretty sure Royale players and people playing Standard also are counted into the Steamchart figures. You still play those modes through the PoE client.

the last time GGG released the numbers it was 65% steam 35% their own client

"

Thehealyhealy wrote:

You appear to have really misunderstood my post.

My post was stating that anyone logging into poe client is counted in those steam chart numbers, not just league players. My post was suggesting that Royale may be inflating the numbers of people still playing as they are counted in those charts. So when people are comparing drop off relative to other leagues Royale is a factor that should be considered.

Since Royale wasn't in previous leagues the comparisons of drop off players might actually be more severe than the current steam chart numbers suggest as previously people who did not enjoy the league and didn't play standard would just quit the game and move on to other games, now with Royale mode some of these players may still be playing poe by purely playing Royale.

For example, todays peak was 54k, let's create a total random number guess and say 5% of that 54k were purely playing Royale, they weren't playing the league at all. If Royale didn't exist then that 54k would actually be closer to 51k.

Ah, sorry, I thought you meant the opposite :)
I've touched on that before.

A very interesting split in my opinion is to see how many are playing softcore trade vs how many are playing SSF or any type of hardcore league right now.

My guess is that SSFHC and SSFSC players like the new league much more than people in trade league.
Of the 54k who logged in today, I would assume that SSF has lost far fewer players than trade leagues.
This are only guesses however.

"

Deadandlivin wrote:

You might be correct, I'm not good at reviewing stats. But I don't understand how you're getting those numbers.
I'll take Ritual as an example:

Ritual Day 1 peak:


Ritual Day 35 peak:


If I'm not misstaken, you just count the losses by dividing the Day 35 peak by the day 1 peak.

90539/157091 = 0,576 -> 57.6% left playing -> 42.4% had quit the league.

I don't understand what numbers you are using to find that Ritual lost 46% of it's population by day 20 when my calculations that 42% had left after 35 days.

All good mate and your calculation method is correct but I think the difference is in the data source. I started out using steam charts but after another player pointed out to me this site: https://steamdb.info/app/238960/graphs/ I've been using that instead. The data is almost identical (slight variations likely attributable to point in time extraction of data) but the big benefit of the latter site is you can extract the data into excel (see the down arrow on the top right corner of the graph.

This allows me to look at each individual day for each league and run a formula that automatically calculates the percentage decrease. One issue I've noted with using Steam Charts is that as you look at larger periods of time, it condenses the data points so you can't look at individual days any longer.

For example, launch day for Ritual was 157,103 on SteamDB (SDB) and 157,091 on Steam Charts (SC), so close as to be negligible. However, the daily data on SDB shows that the league launched on 15 Jan 2021 which is also confirmed on the wiki: https://pathofexile.fandom.com/wiki/League It's pretty easy to tell just by eyeballing the data as you have a sudden massive player spike on launch day. SC however shows you the correct data but because it's truncating the dates it tells you the league launched on 28 Dec 20.

That date is actually day 193 of Heist and had a player peak of 14,613. Totally not your fault as SC doesn't explain on their site how they manage these types of situations, and it's pretty strange how they display the correct peak number but don't correlate the date correctly. That's just poor data representation skills on their behalf.

The 1 Feb 21 date data you had from SC indicates 90,539 and SDB indicates 90,952 so again near identical but as we've discovered that the start date was wrong, 1 Feb 21 is actually day 18 of the league and not day 35.

So that's the solution to our conundrum and is certainly not your fault. You were simply relying on the data source to display the correct data and you weren't to know that the site was flawed.

It's actually quite interesting to download the full league history of PoE and graph the data. There's three observable phases in the development of the game:

1. Start through to Legacy. Leagues generally consisted of a reasonably simple mechanic and while this piqued interest and resulted in a player spike above the baseline, it wasn't huge and player decline from league launch is a very gradual rate.

2. Harbinger was the first massive player spike league with a near 50% player increase from the previous league. It's also the start of a steeper daily decline in player numbers. While more players are playing the game, it seems they are doing so for a short burst then returning for the next league. The player base is changing.

3. Synthesis can be attributed as the first of the logarithmic player decline leagues characterised by significant player decline in the early portion of the league. A combination of power players completing their goals rapidly and other players not engaging with and continuing the league.

More broadly, from Betrayal through to Heist, launch day populations are relatively consistent around 110k with the notable exception of Blight at 75k. Ritual and Ultimatum jumped to around 155k and then Expedition dropped to 116k. So excluding the outliers of Ritual and Ultimatum, Expedition was actually on par for initial performance. There's very little evidence of major external influences such as summer holidays, pandemic or anything else.

Having said that though, the 9 day player drop is worst for Expedition followed by Heist, Synthesis and then Blight. Expedition is an absolute stand out though with roughly a 10% greater decline than even the worst previously performing league. And that includes the possible Royale boost that Thehealyhealy mentioned previously.

As has been said before though, while it's interesting to review the stats and draw our own conclusions, GGG have far more data at their disposal and only they are able to determine what is working or not according to their internal plans.

Honestly it's difficult to tell how big a hit the game is taking from comparison between this league and pretty much any league that's not a summer league.

Comparing it to Ritual/Maven isn't the best because you're comparing a small league against a larger (+atlas expansion) league. Adding more shit to the atlas brings in more players who want to push their atlas. This is especially true for standard players, whose atlas is ez unlock to 164/164, and who padded the Ritual numbers when they came back to pick up their passives.

The better comparison would be with Legion and Harvest, and even then Harvest had its issues with "captive audience" effect due to the pandemic.

If you truly believe the current league numbers are 100% due to the current load of nerfs, you have to wait and see what happens when the dust has settled. First two weeks of league a LOT of shit gets broken and patched, unplayable game modes and mechanics earn GGG so much scorn and vitriol they actually go and fix them, etc.

Blight, Metamorph, and Heist were on track to shed loads of players due to invulnerable monster glut crashes, organ spam, Tane causing you to click him during another transaction and crash your game, the stash blockade scammers in Rogue Harbour, etc. People screamed bloody murder, devs got smacked upside the head and told to fix their broken crap, and miraculously these leagues went on to do well. Synthesis was an absolute joke for weeks, a neverending chain of instance losses and c*ckblocks on the memory nexus planner. This too got fixed even though it took most of a month, and the league still did well compared to spring 2018.

Give them time to sort their shit. If they double down on stripping away diversity tools and QOL in the UX/UI, then next league will show it too. If they come up with innovative replacements for the things they took out of the game, and these replacements aren't too heavily gated behind streamer RNG, then ppl will be back.

Thing is, those leagues you mentioned had to combat broken league mechanics, I myself did play synthesis, even though I lost quite a few maps which was annoying for sure, it was one of the best integrations of seamless optional but rewarding league content.

Doing the content inside the map worked flawlessly, puzzling them in wasn't time consuming, you easily could discard bad pieces. Once you were done with your current setup you could just run it all at once.

I think synthesis NAILED integrating league content into your mapping flow.

But enough of synthesis.

Expedition suffers from a polarizing league mechanic, you either love or hate it, I do not see a lot of middle ground there. (I for one hate it as it kills the flow on top of the flow killing core changes)

And on top of that the over the top change to the core gameplay experience.

Maybe I wouldn't hate the core gameplay change if they just nerfed the support gem multipliers. Step by step wokring into this state, finding issues in the new gameplay experiences along the way and adressing them.

But changing that crapload all at once, made me have a terrible gameplay experience this league.

And no I won't see any change for the better until next league, when I will most likely give it another shot. If it still feels like shit, there is always 3.17 right?

Chris already stated 23% of players quit right at the patch notes. On launch, the peak of its performance, it started 23% lower. Chris obviously has stats across both platforms. Those numbers line up with steam charts. It really boggles my mind the denial.

Are players leaving: yes
Are many players unhappy with the changes: yes
Will it kill poe: No
Will this continued direction kill PoE: Personally yes, but I am unsure.

Can only repeat the same boring slog so many times GGG and if PoE 2.0 is PoE but slower and....slower. Well why would I play? Far better games out there for tactical, slow combat. If I wanted a slow, methodical and unforgiving game, I'd play Dark Souls or GTFO. Not an "ACTION" RPG :)

"

Johny_Snow wrote:

Do note that Harvest was one of the best leagues ever

ehhm...no? Its was maybe best just for people who live to craft. For people who dislike crafting - this league was pure sadness. With 99999 clunky stuff and awful implementation.