37 replies to this topic

[Nightbird]

Grus, on 02 April 2021 - 08:53 AM, said:

I think the recent uptick was due to the rumor that PGI was going to be under new management.

I don't most of the pop know or cares about such news. I think it's more likely hiring Daeron and rolling out a series of highly rewarding events.

[w0qj]

The steady stream of older players returning is definitely after announcement of hiring Daeron, with hopes of MWO development restarting...

There were very few premium mechs played before Sept 2020, mechs like TBR-Prime(I), BLR-1G(P), HBR-Prime(I), WHK-Prime(I), HBK-IIC(O). I know, because I myself was doing research into Clan Wave 1/2/3 to see which packs to buy. [I wanted that Mad Dog MDD-Prime(I) ].
(Now, if only there's Set-of-6 to enjoy the OmniMech quirks, then I would jump at it! But I digress...)

But after announcement of hiring Daeron in September 2020, there was a lot more older returning players sporting premium mechs like TBR-Prime(I), BLR-1G(P), HBR-Prime(I), WHK-Prime(I), HBK-IIC(O). It was a very noticeable anecdotal observation.

Nightbird, on 02 April 2021 - 09:32 AM, said:

I don't most of the pop know or cares about such news. I think it's more likely hiring Daeron and rolling out a series of highly rewarding events.

w0qj, on 02 April 2021 - 06:54 AM, said:

There's a lot of older players returning, if you take time to play Tier 4/5 you regularly see premium mechs like TBR-Prime(I), BLR-1G(P), HBR-Prime(I), WHK-Prime(I), HBK-IIC(O), etc.

Talked to a few of them, and they invariably said that "they heard that MWO development has restarted" and returned to play!

Here's to hoping that they would continue to play in a few months time

Edited by w0qj, 02 April 2021 - 09:48 AM.

[PocketYoda]

Grus, on 02 April 2021 - 08:53 AM, said:

I think the recent uptick was due to the rumor that PGI was going to be under new management.

It was actually to do with the changes, the game had no updates or changes for two years then they have now people came back until the last patch killed it again.

[Nightbird]

GARION26, on 02 April 2021 - 06:39 AM, said:

Night bird thanks for the really cool data analysis.
I forget if you have done this previously but have you/can you do the following?
1) Show overall retention of the player base ignoring WLD breakdowns
2) Show retention by quartile of the player base (divided into quartiles on whatever measure you pick)

The reason I ask is I suspect the majority of the player base clusters around WLD ratio of 1.0. The people who lose or win at a very high rate are I would suspect a fairly small group of the overall player base. The graphs you've show aren't easy to interpret in terms of which lines most represent the player base as a whole (though clearly the trends are mirrored at all WLD levels.)

One point I've made when you've posted this in the past is people who win a lot or lose a lot both tend to have the lowest retention rate. Despite some believing highly skilled players 'want to farm' others, the folks who win at the highest rate in MWO tend to drop out of the game almost as much as the ones who lose at the highest rate. Some of that may be opposite sides of the experience spectrum "This new game is really hard I give up," vs "I've solved this game and it's no longer a challenge."

Separate question I wonder if there is a way to show % of players with a certain number of months accrued. For example in any given season what percent of the players have >6 months or so lifetime accrued months of play (or you could do lines based on multiple cutpoints <1month, >12 months etc). Not sure how that helps us visualize data but I was thinking about it when looking at the graphs -is any given season's total population a lot of older players returning or newer players.

I created a graph showing cumulative number of games versus retention rate, but keep in mind my data starts at season 17.



Overall, this graph is not very informative, since as expected people who play more games are better retained. It's like saying taller people are heavier on average.

[Dogstar]

One thing I found interesting about the stats in the first post is that both high and low win rate people are the least retained and the most retained are this with middling win rates.

Obviously I would expect players with a low win rate to quit more as there's the implication that low win rate equates to low satisfaction with the game but I'm surprised that the highest win rate players are almost as dissatisfied with the game.

It seems there's both a poor new player experience but alos not enough challenge for really good players? What do you think?

Unless the high win rate leavers are ones who only play a few lucky matches and thus they are not representative?

Edited by Dogstar, 04 April 2021 - 03:58 AM.

[DUMAR]

One of the best things they could do to help maintaining newer players is fix the unintuitive interface.For someone who trying to get in to the game it is confusing. Just look at the quirks listings. I have to go to the store just to see what quirks that chassis has because there a huge confusing list and its hard to tell what are skills and whats mech quirks. The social tab is just not great. They don't have too overhaul it but streamline it and make thinks little more clear for newer players. Some tool tips are just wrong and say things that are not in the game anymore. There just small little things they could polish so things are more clear for newer users.

[Thorqemada]

I feel it would be helpful data if the retention bar for each colour would also show how big of a part of the whole playerbase it is.

A bad retention on a tiny part of the playerbase is not catastophic - but on a major part of a playerbase its bad.

Vice Versa a good retention bar does not show how good or bad it is for the game until you know how big of part of the total playerbase it is...


But it is safe to assume that at this point MWO is ruled by estabished players and your playing experience will strongly relate to the amount of estblished players fighting for vs against you etc.

At some point it will simply become an insourmountable effort for a new player to stand tall against the established playerbase not to speak of catching up in a reasonable amount of time experiencing a reasonable amount of success/fun that keeps them playing especially in a generally antisocial game like MWO in regard to the lack of tools and oportunity for a new player to get socially involved in an intuitive way with the established playerbase.

A 5 minute slaughterfest will not build ties between players...

Edited by Thorqemada, 04 April 2021 - 08:51 AM.

[KodiakGW]

Nightbird, on 01 April 2021 - 10:03 AM, said:

If one season had 100 players and the next season had 105 players, a retention rate of 80% tells us that we lost 20 old players and gained 25 new players.

Or 25 new alt accounts. That seems more plausible.

[Nightbird]

KodiakGW, on 04 April 2021 - 09:17 AM, said:

Or 25 new alt accounts. That seems more plausible.

Any data to back that up?

[KodiakGW]

Nightbird, on 04 April 2021 - 10:24 AM, said:

Any data to back that up?

You know damn well that the only people who can have definitive proof is PGI. And, you know damn well they don't share any information with the community, especially their "metrics" they used to make changes.

But, one can be observant. One can observe the posts from players saying they are using alt accounts. One can observe a streamer running alt accounts, but doing things to make the stream fun and interesting (like purposely giving themselves a handicap so not to ruin other players experience). One can observe the players in the match that have never been on Jarl's until 1-4 months ago (not like PGI is spending lots of money on advertising), while a number of active accounts go idle. One can also observe 99th percentile players who haven't played since November, but still making observations about current status of the game.

[Nightbird]

KodiakGW, on 04 April 2021 - 06:05 PM, said:

But, one can be observant. One can also observe 99th percentile players who haven't played since November, but still making observations about current status of the game.

Observations can be wrong. For example, I play FP so I don't show on the leaderboards often. I'm sure there are people who play alts, but going from a handful of observations to saying 25 of 25 new players are alts is a stretch.

[KodiakGW]

Nightbird, on 04 April 2021 - 06:13 PM, said:

Observations can be wrong. For example, I play FP so I don't show on the leaderboards often. I'm sure there are people who play alts, but going from a handful of observations to saying 25 of 25 new players are alts is a stretch.

Just has a match with this one. Nope, must be a new playa!
https://leaderboard....?u=GoneCommando

Please. Either you are ignorant, or want people to be ignorant.

[Nightbird]

KodiakGW, on 04 April 2021 - 07:22 PM, said:

Just has a match with this one. Nope, must be a new playa!
https://leaderboard....?u=GoneCommando

Please. Either you are ignorant, or want people to be ignorant.

Say that is an alt account, does that mean 100% of all new accounts are alt accounts? If not, then what percent?

[GARION26]

Nightbird, on 04 April 2021 - 07:25 PM, said:

Say that is an alt account, does that mean 100% of all new accounts are alt accounts? If not, then what percent?

Agreed and similarly unless you are positing there is some reason for one month to have more Alt accounts generated then the next it really doesn't make sense to focus on alt accounts as the explanation for changes in MWO population.

You could make up all sorts of odd stories about alt accounts effecting population which no one can easily prove or disprove, and makes discussing the player base size data meaningless. For example maybe after the launch of clans in MWO ALL players had an alt account one for IS and one for clans so 1/2 of all accounts were alts (or maybe people had an account for each IS faction.) And maybe the decline in population since then has been just people no longer using their alt accounts as there is no reason to do so after FP became more flexible, so actually the player base is unchanged over many years. You could make that story up, it doesn't seem realistic at all but if you start making unprovable assertions about Alt accounts being a major component of player base size changes you could really make any story up you like.

Or we could just assume alt accounts are created or fall out of use at a rate that essentially parallels the same rate as non alt accounts and then have meaningful analysis and discussion of the player base like Nightbird is trying to have.

Edited by GARION26, 06 April 2021 - 09:24 AM.

[LordNothing]

Vorpal Puppy, on 01 April 2021 - 04:25 PM, said:

I'm going to make a broad generalization here, and maybe I am not understanding your chart, but what your data tells me is that this game loses approximately 1/3rd of its player base every month. That seems really high, but I am not up on player retention trends in online gaming.

If so, have you tried to compensate in your data analysis for players who play more sporadically, or alternate between multiple accounts? You would just have to open up your comparison from a one-month window to say a 3-month window.

i think a bit of churn is normal. players come back play for a couple weeks and move on, eventually they come back, but they take a break long enough for the stats to indicate them as a departing player. do they then get counted as a new player when they return or are they just added back into the retained player pool? ive had month long stints playing other things or im just too busy to play games or working on some project.

[Nightbird]

LordNothing, on 05 April 2021 - 12:06 PM, said:

i think a bit of churn is normal. players come back play for a couple weeks and move on, eventually they come back, but they take a break long enough for the stats to indicate them as a departing player. do they then get counted as a new player when they return or are they just added back into the retained player pool? ive had month long stints playing other things or im just too busy to play games or working on some project.

Right, the longer a period you consider it acceptable for a player to be absent and still be an "active" player, then lower the calculated churn rate. However, to meaningfully show how happy players are, I think a rate comprised of people willing to put in 10 QP games per month is a good start. They're already not contributing much to the active player count on steam charts (https://steamcharts.com/app/342200) at only ~2 hours played per month. At per 3 months, per 1 year, or per any played time in history, then the churn rate will look better but not be more indicative of the health of the game.

Instead, I think it's more important to note events that happened during upticks and downturns in retention rate. Which actions helped boost retention? Which hurt?

Edited by Nightbird, 05 April 2021 - 01:14 PM.

[KodiakGW]

Further proof: Screen shot from someone over on Outreach that was complaining about team tonnage variance. You can see the unblocked names over on Outreach.

https://i.imgur.com/pq29RxF.jpg

So, lets talk numbers. Last month there were 2707 "new" players, 22735 total. So that means 11.9% of players were "new." In the screen shot there are three players with only 1 season under their belts. It is obvious that the Rifleman was a new player, and the team buddy above him is an alt account so they could teach them how to play. When you see the name, you will know that there are 2 other players with really close names to him/her. The 99%'er I left the name in because it is an obvious alt account who knows about the ongoing NASCAR issue in the game. So 12.5% in that game are "new". Pretty close to the overall.

So I would say the majority of "new" accounts we are seeing are alt accounts. In the case of that game 2/3 of the "new" players were alt accounts. This, with the other observations I mentioned prior, could be turned into quantifiable proof. If someone wanted to spend the time. I would, if they listened to even one of my thoughts. TBR mobility buff in April?

We need to all stop saying "look at all these new accounts" whenever they make a bad change, or even a good change (like new player experience the prior patch), that makes people wanting to create alt accounts. But, this argument is moot... if they follow through with the April changes they say they will be making.

[Nightbird]

Urg... this is a meaningless topic to pursue. To put the problem simply, I'll use the lottery as an example. If you see a picture of 3 people with lottery tickets, and 2 of them won prizes, it does not mean the odds of winning the lottery is 2/3. Around where I am, the odds of winning a prize from a lottery is 1 in 25. The odds of winning the grand prize is 1 in 300 million.

In other words, providing one screenshot of anything isn't meaningful at all. You either know how many alts accounts there are, or you don't.

The graph is useful because you can see that when they hire Daeron, retention rate improved. When they announced the maintenance mode for MWO, retention tanked. When they reset PSR, retention tanked for the first month, then recovered.