Outfit Leaderboard - Ideas Draft

[robocpf1]

I plan to submit this post to the PS2 community in one way or another, maybe when players are allowed to propose roadmap ideas like Higby suggested - or I might try making an illustrated video - but I wanted to get some PSU feedback first. If you don't like the idea of outfit leaderboards in general, telling me in this thread doesn't really help your case, I know some of you don't want them. I do, however, think they will get put in the game, and so I am here to suggest a way to implement them.

---

This post will put forward some ideas about the proposed “Outfit Leaderboard” system on the PS2 roadmap. It’s a long one, folks – but this is a controversial topic and I owe it enough space to properly illustrate my points. If you don’t have much time or interest, skip to the bottom paragraph in [brackets] for a TL;DR summary.

One of the problems with leaderboards is finding this “metric,” as SOE says, by which to measure how successful an outfit is. Kills? K/D ratio? Experience gained? There has to be some notion of an outfit’s “score” in order to determine who is winning. That’s the whole point of a leaderboard. However, Planetside is not a simple deathmatch game – there is much more to it, and so it follows that more needs to be measured.

My answer is not to design one new “super metric” but to use all of the metrics. Let the players decide what is important for them. Allow sorting for any metric the player wants. There can be both “total kills” and “kills per outfit member” – the large outfits might compete for the massive amount of kills – “Hey TR Outfit, that’s a nice 1000 kills you got this week. We got 5000.” - while the smaller outfit can say “Yeah, but our members killed, on average, 300 people each this week. Your outfit killed only 50 per member. Quality over Quantity.” The planned distribution of the API could let us create our own leaderboards, too.

You could show both a “simplified” and “advanced” leaderboard. The simplified version shows the composite metrics, like “most territory captured” and “most kills”. You could then access the “advanced” leaderboard for the breakdown. I’m talking about really in-depth breakdowns of the kind that would show some very cool statistics, which I will get into. Having “all of the metrics” creates a less biased statistics pool that the players control. SOE has been big into letting the players control our experiences – this is one more way to do so.

“Territory captured” is a unique Planetside concept, which is based around control of the many hexes in the game. We can do so much more with this statistic though.

“Territory Captured – No Activity”
“Territory Captured – Enemies Detected”
“Territory Captured – Enemy Squads Detected”
“Territory Captured – Enemy Platoons Detected”

This breakdown would show how contested the territory is that this outfit is capturing. You could clearly see, in comparison to the other outfits, if this outfit is taking hotly contested territory or just ghosting around.
Each outfit is a different size, however, so we could use relative stats, like so:

“Territories capped while outnumbered”
“Territories capped during an even fight”
“Territories capped while outnumbering the enemy.”

I’ve already briefly mentioned a kill stat breakdown – How many kills did this outfit get while in tanks? As infantry? In the air? You can go deeper though - why not have a metric for “amount (or percentage) of kills that had been alive for less than a minute” and “amount (or percentage) of kills that had been alive for more than a minute”? You can tell who has been spawn camping and who is fighting in the trenches.

Another problem is properly adjusting the scores so that outfits with an exceptionally large number of players and outfits with a smaller number of players are both fairly scored. It isn’t fair to small outfits to be pitted against the overwhelming kill stats large outfits can generate, and it isn’t fair for the large outfits’ achievements to be thrown aside in a game where the tagline is “Size always matters”.

So, to be fair to all outfits (as I would expect every outfit to have some inactive members) it is also important to have an automatic process to mark players “inactive” with regards to the leaderboard. Outfit A has 1000 members. This week, only 500 logged in. For the leaderboard listing this week, Outfit A’s scores are adjusted for having 500 members – not 1000. This makes it so outfits do not have to kick members that have simply gone inactive, like military personnel on deployment. The bottom line is, if they aren’t actively contributing to an outfit’s score on the leaderboard, don’t count them as part of the outfit’s membership for that leaderboard listing.

[The nature of the competition is that outfits are already comparing themselves to each other – and no matter what super-metric you design to encompass everything, there are outfits that will be unhappy with it because their playstyle doesn’t mesh with the metric. So forget about a super-metric, give us all of the metrics, and let the players decide who is winning engagements based on that information.]

If you’ve read to the end, I thank you, and I humbly ask for your feedback.

---robocpf1

[Scourge]

The nature of the competition is that outfits are already comparing themselves to each other – and no matter what super-metric you design to encompass everything, there are outfits that will be unhappy with that because their playstyle doesn’t mesh with the metric.

Your post was a fair bit longer then it needed to be, pretty much the entire point that your hitting on is contained in the above statement.

SOE insists that their game is a player determined experience, yet now they want to compare apples and oranges, using the metric of which one tastes most like apples.

In the end, all SOE can do is provide stats, and let people sort and order them however they please.

[robocpf1]

Examples are important. How many posts have you seen where the OP says "Hey guys, your idea isn't going to work" but then doesn't propose any examples? I figured that was pretty important, but I will definitely work on trimming it down. Thanks for the feedback.

[EVILPIG]

Leaderboards, whether for outfits, or players, is still a bad idea for Planetside. Even though I was on the Planetside 1 Leaderboard for the life of the game, I never felt it is a good idea to have them.

[Scourge]

Originally Posted by robocpf1

Examples are important. How many posts have you seen where the OP says "Hey guys, your idea isn't going to work" but then doesn't propose any examples? I figured that was pretty important, but I will definitely work on trimming it down. Thanks for the feedback.

Examples are definitely important, but never underestimate the importance of making your post digestible! Its hard to find the right balance.

On the subject of Outfit activity, as you point out, most people in an Outfit typically aren't active. In my 100 person Outfit in Planetside 1, when I tallied how many people had been on in the last 2 weeks, it was usually under 40. For a larger Outfit like GOTR, this ratio of inactive versus active is going to rise exponentially.

I think the way planetsidestats.net handled it was just fine. When looking at an Outfits stats, it recorded how many people logged in that day. If sorting by weekly, it included everyone that had logged in at some point that week, everyone that month sorting by a month, ect. What this meant is that on the Outfit points leaderboard, it would rank Outfits by which earned the most points, and have in brackets right next to their score with how many per member.

Having to manually set members status just for the sake of stats would be a hassle. For a bigger Outfit, I would think it would be unreasonable. The stats recording how many members logged in during the time frame your sorting by, is enough I think.

[robocpf1]

Draft has been redone. I cut the word count in half, and I think I did it without the post losing any value.

I continue to welcome feedback.

---robocpf1

[Tatwi]

This may not be what you are looking for and perhaps it is too naive, but I think that once the outfit decals and base capture credits go into the game, when combined with outfit tags, youtube, twitch streaming, forum signatures, and so on, I don't think most players will have any trouble at all knowing what outfits are effective and interesting. I don't think leaderboards offer anything more than the basis of arguments and negativity that will divide the community rather than help to build and grow it.

I'm with EP, they aren't needed.

[robocpf1]

From a purely business standpoint, I think it's a moot point as far as whether they will put in Outfit Leaderboards. The masses will want them and if they're going to implement E-sports that's an almost required addition.

I don't think SOE has a choice. Regardless of how many people are vocal about fighting it, it really comes down to the fact that they want their game to be competitive, and competitions have to have winners.

With that in mind, I aimed to propose a leaderboard system that at least gives as much information as possible in order for the widest range of outfits to be able to see the stats they consider important.

[Rolfski]

One important thing to know about statistics is that when something gets measured it can and will be managed. This can and will lead to all sorts of unwanted game play as we see now, making it a strong case for skipping leader boards altogether.

However, given the fact that this game will see e-sports at some point and the nature of teams/players to compare themselves, supplying no leader board at all or just providing the plain low level statistics just doesn't going to cut it.

So yes, we need outfit rankings at some point. The best way to deal with it imo is to use a concept that is popular within the big corporations: The Balanced Scorecard. Balanced Scorecard is nothing more than a fancy word for using very smart meta/composite metrics that break down into smaller stats. My experience with them is that if you make them really smart, they can actually reduce unwanted behavior.

TL: DR Do not use a single super-metric for a leader board but a bunch of them that can be broken down into smaller ones that add up to a total score per outfit.

[robocpf1]

Originally Posted by Rolfski

One important thing to know about statistics is that when something gets measured it can and will be managed. This can and will lead to all sorts of unwanted game play as we see now, making it a strong case for skipping leader boards altogether.

I see this as an opportunity to encourage different types of gameplay. What they do with the leaderboards will cause players to look at the stats and give them more importance. If they only go by K/D, total kills, total territories, or something along those lines, players will go out of their way to farm and ghost cap.

If, on the other hand, the leaderboard shows a large variety of different metrics to measure your success by, players won't feel as forced into the Kills/Farming/Ghost capping that we are already seeing. That's how I look at it, anyway, and I don't think that's a far stretch.

[EVILPIG]

Originally Posted by robocpf1

I see this as an opportunity to encourage different types of gameplay. What they do with the leaderboards will cause players to look at the stats and give them more importance. If they only go by K/D, total kills, total territories, or something along those lines, players will go out of their way to farm and ghost cap.

If, on the other hand, the leaderboard shows a large variety of different metrics to measure your success by, players won't feel as forced into the Kills/Farming/Ghost capping that we are already seeing. That's how I look at it, anyway, and I don't think that's a far stretch.

No matter what the metrics are, they will be chasing the dragon.

[Scourge]

Originally Posted by EVILPIG

No matter what the metrics are, they will be chasing the dragon.

We used to lay "Screamin traps" in PS1 all of the time.

Essentially, the player Screamin was on a quest to complete the Tiny Robotics Support merit. Which meant repairing spitfire turrets. Thus, if you damaged a spitfire turret, and waited ready to shoot at it again, sure enough, Screamin would come along.

And merits didn't even show up on any leaderboards! Its useless to fret about what players are being encouraged or discouraged to do. If people can do it, then they will.

I personally know people who stopped playing PS1 after the Gemini merge broke stat tracking. I used to read planetsidestats.net every day. When that went away, the community lost something. And player behavior in game did NOT change. It was as plasma and MAX spammed as ever.

I have reservations about a outfit leaderboard system. I even have reservations about the stats being tracked right now. Robo here never struck me as much of a leaderboard player, but hes done the best job yet articulating how to make one that wont piss all of us off. That is the least we can hope to do.

[robocpf1]

Redrafted a final time.

Any additional feedback is welcome.

[Hamma]

I'm pretty neutral on the subject. On the one hand, I don't think leaderboards should exist at all. On the other hand.. we all know they are coming and will be here to stay. I think multiple ways of sorting outfits is important and a combination of many metrics is also important - everything from kills to territory and facility captures.