38 replies to this topic

[Bishop Steiner]

[Novakaine]

7000 out of 972,322 is a poor showing.
Where's my Fozzie Bear Holo!

[Moku]

Player base is definitely small with only 8k participation. When you hope there's 1-3k online and you see the same groups and players a couple times a night. New players coming in is probably much less than the ones going out just from my recruiting observations.

[hammerreborn]

Novakaine, on 27 February 2013 - 09:03 AM, said:

7000 out of 972,322 is a poor showing.
Where's my Fozzie Bear Holo!

Where is that second number coming from?

[xDeityx]

Syllogy, on 27 February 2013 - 08:32 AM, said:

Since 8000+ people participated in the tournament this last weekend, and that's probably only representing a portion of the active player-base, and an even smaller portion of the total player-base, I'd say no. The player-base isn't too small.

ELO, because of it's nature, will take a while to take root in the live environment, be tweaked, and actually be made functional as a matchmaker.

League of Legends has orders of magnitude more players than MWO and is only trying to match 10 players in a match as opposed to 16 and eventually 24. Despite this, LoL still takes hundreds of games to calibrate a player's Elo accurately and regularly has over 1:00 queue times depending on which game mode you queue for and what time of day you are playing. LoL is also criticized for having too much variance of Elo ratings in a given game...you might have a 1500 playing with a 1300 despite the fact that they are on totally different levels.

I'm not saying that PGI shouldn't use some sort of rating system, whether it be MS TrueSkill, Elo, Glicko/Glicko-2, Harkness or any other rating system adopted from Chess or elsewhere...but let's not paint it as better than it really is: a hackjob adaptation of a 1v1 system because we can't come up with anything better.

[Broceratops]

the reasons only 8k people out of 100k+ opted in was

1) they announced it half a day before it started, and only on the website and a single thread in a section of the forum few people go to
2) it started on a friday afternoon / morning if you were in NA
3) it ended in the wee hours of monday, or if you're in europe, when you were supposed to be at work already
4) it was primarily a huge grind, with skill being a distant distant factor, and with 2+3 above if you weren't able to get off work early or not work monday, it would pretty much eliminate you from the top spots. even then, you had to play a minimum 40 hours over a 60 hour time frame to be top 10. if you wanted to win, 50 hours was a necessity.

so i don't think they expected a lot of participation anyhow. they were just testing the system.

[Novakaine]

Hell I made it up.

[xDeityx]

Broceratops, on 27 February 2013 - 09:14 AM, said:

the reasons only 8k people out of 100k+ opted in was

only on the website and a single thread in a section of the forum few people go to

They also sent an email. Your points stand though, the email was sent after the start of the tourney.

[Tatula]

Even though not everyone opt-in for the tourney, PGI was still collecting data on the non-tourney players.

[Mawai]

Overall, I have found pretty good experiences with the matchmaker. Although there have been a few matches with unequal tonnage ... generally skill and team work have been just as significant if not much more important factors.

In particular, the tonnage of the mech does not necessarily reflect its effectiveness. Looking at the post match statistics in terms of damage done, kills and assists ... lights and mediums have often contributed more than assaults and heavies depending on the loadout and pilot skill.

In your 5 raven example ... I'd probably favour the 5xRavens to win (if they were mostly 3L variants) ... with coordination they would be able to pick off anyone who separated even a little bit from the group. As long as they didn't charge right in and worked together a bit their odds of winning are pretty good.

[Mawai]

Moku, on 27 February 2013 - 09:04 AM, said:

Player base is definitely small with only 8k participation. When you hope there's 1-3k online and you see the same groups and players a couple times a night. New players coming in is probably much less than the ones going out just from my recruiting observations.

I'd disagree. In order to play the tournament you had to explicitly opt-in. This meant going to a website and pressing a button. Most people don't even read the forums. A lot of people don't even bother with the advertising on the mechlab screen and a lot of people could probably care less about a tournament. So ... my guess is that the 8000 represents a relatively small fraction of the active player base.

[KingCobra]

I would guess around 20000-30000 active players a month which is really not a lot for a IP like this. I'm not going back to Google past MechWarrior IP's and give you the figures but they were in the 100000-300000 range a month player base. MWO is not those games it was a new idea that really has not lived up to its MechWarrior IP rep so far. The matchmaking system relies on input by the players stats to match up the closest players with same set stats that's all it does not even take into account mech tonnage or type of mechs being dropped onto maps as far as I can tell. Its a bold grand experiment but its costing them time and player base when past MechWarrior IP'S did a much better job and there 20 years old? The overall mistakes PGI is making are ok if they can integrate what worked well from past MechWarrior PC game IP's turn the game around and start gaining back a bigger player base and overall profits. That is really the bottom line for PGI/IGP money they might make a few bucks now real quick but lose out on millions down the road with there unstable concepts.

[DaZur]

Mawai, on 27 February 2013 - 09:48 AM, said:

I'd disagree. In order to play the tournament you had to explicitly opt-in. This meant going to a website and pressing a button. Most people don't even read the forums. A lot of people don't even bother with the advertising on the mechlab screen and a lot of people could probably care less about a tournament. So ... my guess is that the 8000 represents a relatively small fraction of the active player base.

But..but...but... It's so much easier to theory-craft the premise of a feeble player base when one elects to use snap-shots numbers like the opt-in totals!

[Tonosama]

The player base is going to get a lot smaller when they split the community with regional servers. Not sure how that will impact ELO matchmaking.

[Bilbo]

KingCobra, on 27 February 2013 - 09:59 AM, said:

I would guess around 20000-30000 active players a month which is really not a lot for a IP like this. I'm not going back to Google past MechWarrior IP's and give you the figures but they were in the 100000-300000 range a month player base. MWO is not those games it was a new idea that really has not lived up to its MechWarrior IP rep so far. The matchmaking system relies on input by the players stats to match up the closest players with same set stats that's all it does not even take into account mech tonnage or type of mechs being dropped onto maps as far as I can tell. Its a bold grand experiment but its costing them time and player base when past MechWarrior IP'S did a much better job and there 20 years old? The overall mistakes PGI is making are ok if they can integrate what worked well from past MechWarrior PC game IP's turn the game around and start gaining back a bigger player base and overall profits. That is really the bottom line for PGI/IGP money they might make a few bucks now real quick but lose out on millions down the road with there unstable concepts.

There were 100,000 playing this weekend.

[KingCobra]

Bilbo, on 27 February 2013 - 10:05 AM, said:

There were 100,000 playing this weekend.

I think your joking right? hahaha or maybe you were thinking games played? 100,000 divided by 8000 players=125 games each give or take a few games by each player? but lets put it to the test shall we? HELLO PGI ARE YOU READING THIS TOPIC? can you please post the player stat on player numbers/games played/time online/ETC for this past weekend starting FEB 22-FEB 25? and we would see who is correct?

[Bilbo]

No need for a test. I can read:

Bryan Ekman, on 26 February 2013 - 11:30 AM, said:

Thank You to everyone who participated in our first ever MechWarrior Online player challenge!

We saw some excellent BattleMech combat through the course of the weekend, with over 100,000 players joining the 8,026 challenge participants in some epic battles.

We are happy to announce the official winners:



Players will receive a cockpit banner on March 5th during downtime.

We've collected lots data, heard your feedback and we look forward to March and new player challenges! Until then, enjoy some really cool challenge statistics!

[Pihoqahiak]

M3atloaf, on 27 February 2013 - 08:43 AM, said:

There were only 8k people involved in the tournament but there were another 100k that were online over the weekend joining them who didn't opt-in. 108k players isn't a small player base.

Where are you getting your numbers from?

Edit: Ugh, of course it was at the end of the thread so I posted this before I got to it. I find it hard to believe that there were over 100,000 players this weekend, but hopefully it's an accurate number.

Edited by Pihoqahiak, 27 February 2013 - 10:27 AM.

[hammerreborn]

Those are actually some really cool numbers. Apparently all the people who opted in fared better than the average player, with a greater K/D and W/L ratio. And captures are really low, as I thought, even though it seems like someone posts a new thread everytime they get capped. And it looks like most people ended the match with 1 kill and 3 assists, which seems really low on the assist side.

Also Bilbo where are you finding that?> I've been searching for the it and have absolutely no idea where they hid it in the giant web of forums.

Edited by hammerreborn, 27 February 2013 - 10:44 AM.