Jump to content

Armchair Analysis: Tukayyid


31 replies to this topic

#1 DoctorZuber

    Member

  • PipPipPip
  • Bad Company
  • 75 posts

Posted 30 April 2015 - 02:57 PM

It is done. Clan wins with (36/63) territories owned giving them 57.14% control of Tukayyid. Historically, this would mean that Clan now owns Terra (assuming IS can be counted on to honor their agreement) and there would be no 15 year truce. (which historically was broken in under 7 years.) Practically, it means I get free cockpit clutter that says clan won the event.

But who cares about any of that stuff. Let's take a moment to look at the numbers and see just how this event turned out in the real world.

Totaling up the numbers shows that Clan players earned 85566 points where IS earned 78315 points. I am making an assumption that this is the same 1 point we earn for finishing a CW battle with a score of 80 or more. Follow along and I will demonstrate why I think this is the case.

Start with the big number of 85566 for clan. The event was 7 days, so (85566/7) this is an average of 12224 a day. There are 12 players per battle, so this would mean (12224/12) 1018 per day. There are also 63 battlefields so this number reduces down to (1018/63) 16.2 per player per day. IS meanwhile earned 14.8 per player per day.

Now this isn't really how many points the average player earned. Without statistics to show how many actually participated I can only speculate on that number. This is instead how many were earned per side in the same span of time.

Now as I said, I am assuming that these points are one and the same with the 1 point we earn for any battle with an 80+ battle score. So the theoretical max here should be 48. a score of 16 would imply that the average player is putting roughly 8 hours into CW each day. Actually, this number should be slightly higher since you also need to take into account the occasional game where a player fails to make their 80. So that leaves us with maybe 8 to 10 hours invested on average. With an event like CW I expect to see a much higher turnout of hardcore players so that seems pretty plausible.

Since all of the numbers seem to fall within expected ranges, I will continue with the assumption that the posted faction results are in fact the same 1 point per 80 we earn individually.

Assuming max attendance this number should be near triple this. Que times are really irrelevant from the broad analysis here. Theoretically, each day 63 battles are fought, every 30 minutes, for 24 hours for seven full days. Theoretically, assuming nobody ever fails to get their 80 and the que never runs empty the total for each side should be 254,016 points. The fact that the numbers are barely a third of this can only be explained by the fact that the ques were not always full. So the conclusion that must be drawn here is that over the course of the entire event roughly 2/3 of the potential battlefields were unused.

What is also interesting is that IS earned notably less points overall. The same number of battles were fought by both sides. The only possible explanation for the difference in points is players who failed to get their 80. Apparently, IS had more pilots get drubbed out without earning their points than Clan did this event.

And that is pretty much all I can derive from the data available to me. Hopefully, someone finds the analysis at least a little bit interesting. Peace.

EDIT: With the additional posted information from PGI, here is some more information.

First, I am no longer certain what exactly the "points" are Clan vs IS. They do not line up with the posted numbers for how many points players actually earned, nor do they seem to match with anything else either. So Who knows.

However, the posted points does bring a shocking reality. I took the list from 1 to 50 of how many players earned each number of points. In total, 15671 of the 17224 players who attended, earned 1 or more points. This means right out of the gate, 1553 players (10%) gave up before earning even one point. Adding this to the 4000 who earned 3 points, almost a third of the players who participated gave up after only a few games.

Now the next number I'm looking at is the 334558 games played. With 17224 players this means the average player played 19.4 games. Comparing this against the points earned, remembering that 1553 earned zero, we find that the average player also earned only 7 points for their trouble. That is a scary number of games where people failed to get their 80.

Last I want to look at the population numbers. Now I get that people love to speculate and wildly exaggerate when it comes to population numbers. I hear 5 to 1 and 10 to 1 pretty often, and even admit that I may have used such numbers myself. Clearly they are a lie, at least on the whole of it. It's still very possible that there are some wilder imbalances on specific time zones, but on the face of it CW attendance for tukayyid was 10346 IS to 7312 Clan. Which while significant, isn't an overwhelming population gap.

What I do find significant, but lack the numbers to verify is how much that experience changed from day 1 to day 7. On day 1, I saw an IS que hit 60+ where the clan que never reached above 12. That seems like a pretty significant difference. Towards the end of the event however, I saw both ques dwindle down to zero at least briefly with 1 to 10 being common on the IS que.

And in that, there were only 380 trades from IS to Clan. 52 actually traded back from Clan to IS leaving a net change of only 328. In the face of these numbers, that's only a few percent change at best. Not terribly significant. The only other possibility to explain such a dramatic difference in the experience from day 1 to day 7, is people just giving up. I suspect a lot more IS players threw their hands up in disgust and left than Clan did. I honestly can't blame them. I'd be pretty mad about the ques too if I was in their shoes.

Edited by DoctorZuber, 02 May 2015 - 11:29 AM.


#2 Ted Wayz

    Member

  • PipPipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 2,925 posts
  • LocationTea with Romano

Posted 30 April 2015 - 03:10 PM

You are assuming that everyone was trying to get their 80 every day of the event. I think that is incorrect.

Given the shorter queue times the clan was able to get their full rewards well before the event ended. After that point they switched their play styles to focus on winning the event. This in turn dragged down IS scores at the end of the event putting salt in the wound of those struggling through the queues to get base rushed in 5 minutes and have no one reach 80 points.

You would need to see scores over time to really judge this event and that data is not available.

#3 Fate 6

    Member

  • PipPipPipPipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 2,466 posts

Posted 30 April 2015 - 03:18 PM

View PostTed Wayz, on 30 April 2015 - 03:10 PM, said:

You are assuming that everyone was trying to get their 80 every day of the event. I think that is incorrect.

Given the shorter queue times the clan was able to get their full rewards well before the event ended. After that point they switched their play styles to focus on winning the event. This in turn dragged down IS scores at the end of the event putting salt in the wound of those struggling through the queues to get base rushed in 5 minutes and have no one reach 80 points.

You would need to see scores over time to really judge this event and that data is not available.

IS scores would actually be higher than Clan if what you're suggesting is remotely true

#4 Kiiyor

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Big Daddy
  • Big Daddy
  • 5,565 posts
  • LocationSCIENCE.

Posted 30 April 2015 - 03:21 PM

Hmmm, very nice science.

I'd say that there's quite a few extra factors involved with the actual numbers of battles that would have taken place - some shorter than 30 minutes, some much longer - if you think about the amount of time spent staring at the EOM screen, then back in the lobby, then getting everyone ready (if you're a 12 man) or clicking ready if you're a PUGGER in between watching DareDevil or whatever. Also, there was never a time in the event that I saw even wait times for the pre match queues, as there were always 5-15 or so more 'Sphere groups waiting for a Clan opponent.

All minor things though, and i'd agree with the 'Sphere numbers pointing towards more of their players getting annihilated. It's where the newer guys are. Sure, there are trial mechs for the Clans, but trial mechs SUCK (except maybe the StormCrow) and building a Clan dropdeck is hideously expensive when compared to the 'Sphere.

Throughout the event, I saw some truly poor dropdecks for the 'Sphere, while the vast majority of Clan decks I encountered were all slight variations of the usual MadCat/StormCrow decks of DOOOOM. Unless you build a very tight 'Sphere deck that compliments those everyone else brings, you're in for pain. This, IMHO, is one of the bigger aspects of Clan dominance in CW - their most popular mechs all synergize incredibly well, whereas the 'Sphere... doesn't.

#5 DoctorZuber

    Member

  • PipPipPip
  • Bad Company
  • 75 posts

Posted 30 April 2015 - 03:26 PM

View PostKiiyor, on 30 April 2015 - 03:21 PM, said:

Throughout the event, I saw some truly poor dropdecks for the 'Sphere, while the vast majority of Clan decks I encountered were all slight variations of the usual MadCat/StormCrow decks of DOOOOM. Unless you build a very tight 'Sphere deck that compliments those everyone else brings, you're in for pain. This, IMHO, is one of the bigger aspects of Clan dominance in CW - their most popular mechs all synergize incredibly well, whereas the 'Sphere... doesn't.


That is an excellent point. The shorter list of mechs available for clan does limit the potential for useless team compositions. This will likely change as more clan mechs become available.

#6 Vassago Rain

    Member

  • PipPipPipPipPipPipPipPipPipPipPip
  • Bridesmaid
  • Bridesmaid
  • 14,396 posts
  • LocationExodus fleet, HMS Kong Circumflex accent

Posted 30 April 2015 - 03:29 PM

The IS advantage of having an estimated 596868236 times more players than the clans was completely nullified by CW's 'two to tango' system, easily guaranteeing a clan win from the outset. It didn't help that the wait times quickly grew past 40 minutes, to the point where PGI offered free faction transfers to those who were interested.

I give the event 5 out of 5 captain obvious stars, and a beta 2 ™ medal of excellence.

Edited by Vassago Rain, 30 April 2015 - 03:31 PM.


#7 DoctorZuber

    Member

  • PipPipPip
  • Bad Company
  • 75 posts

Posted 30 April 2015 - 03:30 PM

View PostTed Wayz, on 30 April 2015 - 03:10 PM, said:

You are assuming that everyone was trying to get their 80 every day of the event. I think that is incorrect.


I think that in this case "trying" is irrelevant.

Failing to get your 80 actually requires a certain amount of effort, or occasionally, a really nasty curb stomping. Judging by my own modest performance, and general observations of final scoreboards overall I expect less than 10%, possibly even less than 5% failed to get their 80 irregardless of whatever they may have been "trying" to do.

#8 DoctorZuber

    Member

  • PipPipPip
  • Bad Company
  • 75 posts

Posted 30 April 2015 - 03:34 PM

View PostVassago Rain, on 30 April 2015 - 03:29 PM, said:

The IS advantage of having an estimated 596868236 times more players than the clans was completely nullified by CW's 'two to tango' system,


I agree, that this did a lot to even up the event over all despite how irritating the long Que times must have been for the excess IS players.

It would be both interesting and most disheartening to see statistics for how long the Que times got and just how often players gave up in disgust to go do something else.

#9 Ted Wayz

    Member

  • PipPipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 2,925 posts
  • LocationTea with Romano

Posted 30 April 2015 - 03:34 PM

Sorry Fate, and I will back up my statement with information instead of a one line dismissal.

The clans ratio of coordinated groups was much higher than the pugliscious IS. This is a fact due to concentration and makes sense. IS will have more pugs because their mechs are cheaper and more accessible to new or casual players. This gave clans an advantage coming out the gate. They were playing against less coordinated groups and able to score more points. At least the OPs data shows that part.

Once they attained their rewards then the clans could focus in on winning. Half the matches would still require they kill the opposition, which they were superior at, but for the other half when they were attacking no one made 80 points and all scores were deflated in the match. Sure IS might have a slightly higher average in those matches but the 40 to 30 does not have the same effect as 220 to 150. It would bring down the overall average but not give the IS advantage in points.

Should I break it down further or are we getting on the same page?

#10 Vassago Rain

    Member

  • PipPipPipPipPipPipPipPipPipPipPip
  • Bridesmaid
  • Bridesmaid
  • 14,396 posts
  • LocationExodus fleet, HMS Kong Circumflex accent

Posted 30 April 2015 - 03:38 PM

View PostTed Wayz, on 30 April 2015 - 03:34 PM, said:

Sorry Fate, and I will back up my statement with information instead of a one line dismissal.

The clans ratio of coordinated groups was much higher than the pugliscious IS. This is a fact due to concentration and makes sense. IS will have more pugs because their mechs are cheaper and more accessible to new or casual players. This gave clans an advantage coming out the gate. They were playing against less coordinated groups and able to score more points. At least the OPs data shows that part.

Once they attained their rewards then the clans could focus in on winning. Half the matches would still require they kill the opposition, which they were superior at, but for the other half when they were attacking no one made 80 points and all scores were deflated in the match. Sure IS might have a slightly higher average in those matches but the 40 to 30 does not have the same effect as 220 to 150. It would bring down the overall average but not give the IS advantage in points.

Should I break it down further or are we getting on the same page?


It really doesn't matter how many comp or trash players there are out there. What matters is that one side has better stuff, and the other has a huge population advantage.

The clan tech advantage was played to its fullest, while the IS population advantage had zero beneficial effects. In fact, it probably made a lot of players not bother with the event due to excessively inflated wait times. I'm one of them.

CW is quite the joke.

#11 Skull Leader2

    Member

  • PipPipPip
  • Guardian
  • Guardian
  • 78 posts

Posted 30 April 2015 - 03:47 PM

Thanks for putting the time and math in to OP.

I think this event showed me that the IS mechs and teams CAN stand toe to toe with Clan but it requires more coordination and good piloting. I think the clan tech (more weapons/ton, longer range usually) helped reduce the advantage of good IS piloting. I would also go out on a limb and say based on what I saw, a lot of people new to CW were running builds built for PUG matches or one round combat. CW requires more planning and different setups for many mechs. For example, LRM builds were nullified a lot by the fact clanners would drop with 3-4 hellbringers with ECM. That doesn't happen often in pug matches. I would love to spitball and theorycraft in the CW section about how differently people outfit their mechs for CW vs. standard matches.

#12 DoctorZuber

    Member

  • PipPipPip
  • Bad Company
  • 75 posts

Posted 30 April 2015 - 03:49 PM

View PostVassago Rain, on 30 April 2015 - 03:38 PM, said:

It really doesn't matter how many comp or trash players there are out there. What matters is that one side has better stuff, and the other has a huge population advantage.

The clan tech advantage was played to its fullest, while the IS population advantage had zero beneficial effects. In fact, it probably made a lot of players not bother with the event due to excessively inflated wait times. I'm one of them.

CW is quite the joke.


Yes, the population advantage, at least of the face of it, is irrelevant in CW. However a commonly held assumption is that the larger number of players at least potentially means more opportunities to form larger outfits to form, which implies that they will also be more skilled because they are outfits who practice together. Whether or not that is true, I couldn't say. I have no data to confirm or deny that theory.

As to your implied argument of OP mechs (clan) vs better skilled players(IS), I can also only speculate. I don't have any data to verify your theory that IS players are more skilled. Nor do I have any data that could confirm or deny your theory that clan mechs are OP.

There are however a lot of players who disagree with that theory and think that IS mechs on average are actually stronger than Clan Mechs.

#13 Alistair Winter

    Member

  • PipPipPipPipPipPipPipPipPipPipPip
  • Storm
  • Storm
  • 10,823 posts
  • LocationBergen, Norway, FRR

Posted 30 April 2015 - 03:53 PM

View PostVassago Rain, on 30 April 2015 - 03:29 PM, said:

The IS advantage of having an estimated 596868236 times more players than the clans was completely nullified by CW's 'two to tango' system, easily guaranteeing a clan win from the outset. It didn't help that the wait times quickly grew past 40 minutes, to the point where PGI offered free faction transfers to those who were interested.

I give the event 5 out of 5 captain obvious stars, and a beta 2 ™ medal of excellence.

Hmm... now I need to make more medals.

http://mwomercs.com/...d-beta-testing/

#14 Telmasa

    Member

  • PipPipPipPipPipPipPipPip
  • Bad Company
  • Bad Company
  • 1,548 posts

Posted 30 April 2015 - 05:33 PM

I think alot of this has to consider the "new player" population disparity between IS and clans, too - I think the amount of experienced players (unit or no unit) are about the same on both sides, but the pool of newer players (or simply casual ones) is probably hundreds of times larger on the IS side.

Which probably could explain a number of things about this event.

#15 CrushLibs

    Member

  • PipPipPipPipPipPipPip
  • The 1 Percent
  • The 1 Percent
  • 546 posts

Posted 30 April 2015 - 05:53 PM

This event made the organized vs pug disparity very obvious.

12 man IS would mop the floor against 12 man clan about 90% of the time but Pug vs clan was directly opposite.

Pilot skill was a large part also. IMO

#16 EgoSlayer

    Member

  • PipPipPipPipPipPipPipPip
  • Wrath
  • Wrath
  • 1,909 posts
  • Location[REDACTED]

Posted 30 April 2015 - 08:00 PM

View PostDoctorZuber, on 30 April 2015 - 02:57 PM, said:

It is done. Clan wins with (36/63) territories owned giving them 57.14% control of Tukayyid. Historically, this would mean that Clan now owns Terra (assuming IS can be counted on to honor their agreement) and there would be no 15 year truce. (which historically was broken in under 7 years.) Practically, it means I get free cockpit clutter that says clan won the event.

But who cares about any of that stuff. Let's take a moment to look at the numbers and see just how this event turned out in the real world.

Totaling up the numbers shows that Clan players earned 85566 points where IS earned 78315 points. I am making an assumption that this is the same 1 point we earn for finishing a CW battle with a score of 80 or more. Follow along and I will demonstrate why I think this is the case.
<snip>


The Tournament page clearly states that the leader board is for victories; http://mwomercs.com/tournaments

Quote

  • Faction Leaderboards are Ranked by Total Faction Player Victories.
  • Unit Leaderboards are Ranked by Total Unit Player Victories.


It has nothing to do with qualifying matches. A win with less than an 80 match score is a victory, and would count for the leaderboard, just not for the personal challenge. If anything it's a number of matches counter, which for the record works out to 52.2% Clan (85,566) wins vs 47.8% Inner Sphere (78,315) wins.

Edited by EgoSlayer, 30 April 2015 - 08:03 PM.


#17 Davegt27

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 7,097 posts
  • LocationCO

Posted 30 April 2015 - 10:35 PM



#18 Elizander

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 7,540 posts
  • LocationPhilippines

Posted 30 April 2015 - 11:25 PM

View PostVassago Rain, on 30 April 2015 - 03:38 PM, said:

The clan tech advantage was played to its fullest, while the IS population advantage had zero beneficial effects. In fact, it probably made a lot of players not bother with the event due to excessively inflated wait times. I'm one of them.


I read the forums and the reported wait times were too silly for me as well so I just did some normal solo queue drops the past week.

#19 Paigan

    Member

  • PipPipPipPipPipPipPipPipPip
  • The Blazing
  • The Blazing
  • 2,789 posts

Posted 30 April 2015 - 11:40 PM

View PostDoctorZuber, on 30 April 2015 - 02:57 PM, said:

[...]
Start with the big number of 85566 for clan. The event was 7 days, so (85566/7) this is an average of 12224 a day. There are 12 players per battle, so this would mean (12224/12) 1018 per day. There are also 63 battlefields so this number reduces down to (1018/63) 16.2 per player per day. IS meanwhile earned 14.8 per player per day.
[...]


I am still having trouble getting my head around the math in this paragraph.
I don't think this is correct.

~16 points per player per day would mean in average, every player would have to grind 8 hours MWO non-stop.
If you count in sub-80 matches even more.

I can guarantee that the majority of players did NOT grind MWO for 9+ hours nonstop a day, that would mean a few would have to play 50, 80, 100 hours per day to compensate.

This can't be right.


Maybe the score does not use the 80-match score points at all.
Also, not all 63 zones were fully occupied all the time.


To get to a per-player number, you would have to first factor out the amount of games.
ASSUMING every zone was CONSTANTLY played on (which is not true), the number of games per day would roughly be 63 * 24 * 2 (2 games per hour, again big assumption, 24 hours, 63 zones) = 3024.
Simply dividing by 63 is not enough, imho.

Also, not all players are online all the time, they are scattered accross time zones, resulting in something like several actual players together making up one "statistical player".


There are far to many variables (average playtime per day, etc.) to get anything meaningful out of this.
You can't just throw in some prominent numbers like 12 and 63 and get a proper result.

Edited by Paigan, 30 April 2015 - 11:56 PM.


#20 DoctorZuber

    Member

  • PipPipPip
  • Bad Company
  • 75 posts

Posted 02 May 2015 - 10:11 AM

View PostPaigan, on 30 April 2015 - 11:40 PM, said:

To get to a per-player number, you would have to first factor out the amount of games.

Also, not all players are online all the time, they are scattered accross time zones, resulting in something like several actual players together making up one "statistical player".

That is correct, and I indicated as much in the very next paragraph. Without the actual attendance numbers to compare against the only number I could create is a pseudo theoretical average player.

The only conclusion I drew from this was the overall attendance. Which was about 1/3 of the max potential attendance, give or take. The theoretical (48 x 63) 3024 matches per day is a rough estimate at best since it fails to take into account prep time, and matches that end in less than 30 minutes. We can at best estimate how many matches are possible, not give an exact number. Prep time is actually more or less a known constant, but match length is variable, so we can at best estimate the total matches in a day. However, most matches do end at 30 minutes, and the prep time of under 5 minutes partially offsets this discrepancy anyhow. So I think this estimate is at least reasonably accurate.

View PostPaigan, on 30 April 2015 - 11:40 PM, said:

There are far to many variables (average playtime per day, etc.) to get anything meaningful out of this.

I was quite careful what conclusions I drew from this because I know the data available was incomplete.

I estimated how many matches were played as opposed to how many were potentially possible. Roughly 1/3 of the potential matches were played overall.

I also indicated that more IS players failed to "get their 80" which is fairly well supported since the number of matches is known to be equal on both sides, so there's really no other way to explain the discrepancy in total score between the two sides.

Also, since PGI has posted additional information, and since it is a quiet day at work, I am likely to sit down and see if I can add to this a bit.

Edited by DoctorZuber, 02 May 2015 - 10:12 AM.






1 user(s) are reading this topic

0 members, 1 guests, 0 anonymous users