War Robots WR Data and Analysis

Club Dredd- A Statistical Perspective

Am I killing the thing I love? I try and find some data to take an unflinching look...

When I was young, my mother used to take us on Summer vacations to Cape Cod, Massachusetts. There’s a cottage up there that belonged to my grandparents when they were still alive, and that our extended family all used for vacations afterwards. The beaches, the flea markets, the bustle of Provincetown, all wonderful and magical childhood memories, but not every memory I have of those trips is a good one.

It wasn’t uncommon to find things left behind from other guests at the cottage, usually inflatables or half-filled bottles of suntan lotion. But one year, I came across a paperback book that one of my family must have been reading. It was Peter Benchley’s Jaws, and although I was quite young, I was a voracious reader and devoured whatever I could get my hands on.

And so began a lifelong fear of being devoured in return.

Impressions are a powerful thing, often able to dispel reason and evidence like light on shadow. The average number of shark attacks per year in the US is sixteen, with an average of one death every other year. To say that they’re rare is actually an exaggeration of the word rare. They’re far rarer than that. But if you think I was taking a warm comfort in the overwhelming power of odds on my side each time I’ve entered the brine after that, well, you’ve probably lived a blessed childhood fearing nothing, you lucky devil you.

It doesn’t take long kicking around the War Robots community to run into the notion of the “clubber.” Although the precise definition varies somewhat, in general terms a clubber is someone who takes advantage of the imbalance of the game that’s available at the lower tier of play. To some people, a clubber is someone who has a hangar full of MagGeps, ready to blast new players into oblivion, whether for Gold farming or just the bullying visceral thrill of blasting bots into oblivion.

To others, even one Gepard will do it, a robot pushed to the point of brokenness by a Pixonix eager for short-term cash over long-term player retention. In this notion, veteran players are not unlike rich businessmen on safari in Africa, paying a nice lump of money to go and shoot defenseless animals.

Still others, taking an even more uncompromising position, claim it doesn’t really matter what bots are in the player’s hangar- a veteran’s experience is enough to tilt the game into unfair territory.

As someone who loves both Bronze Tier play and piloting a Gepard, as well as someone who cares about a game’s ecology and long-term health, I admit it’s something of a struggle. Where should I draw the line. I don’t like the notion that War Robots is a game of linear progression, and that some sense of fair play compels me to run up the gameplay ladder to the snipe-and-campfest of Gold Tier play. I love the thrill of Cossack dogfights, the speed and agility of the Bronze battlefield, and the process of developing expertise on the full spectrum of the game. Bronze and Silver shouldn’t be flyover country, but rather separate and distinct gameplay experience. Why should I race past Destriers just to stare at Furies for the next year or more?

By the same token, though, where does that leave me in regards to clubbing? Is the enjoyment I take in Bronze Tier in part due to the fact that I’m a superior player there? Am I afraid I’ll be “found out” (to myself) if I go up to Silver? Although I feel like I’m still learning the game, where is the divide between “better” and “unfairly better” when you take new players into account?

While much of this may be rendered null when next month brings in a much-touted matchmaking change, these are still questions worth exploring. To hear many players talk, clubbers are ruining the game, driving away legions of new players, never to return. Anecdotal evidence is a fuzzy thing, however, often prone to exaggeration like a game of ”Telephone.’

And on a more selfish note, I want to continue to improve at this game. Any athlete will tell you that the most reliable way to get better is to challenge yourself by facing skilled opposition. Playing on some sort of “easy mode” doesn’t offer the kind of test that hones one’s edge. This, too, was nebulous to define. While I don’t have any access to Pixonic’s data servers, was there anything I could do as a player to get a more fact-based lay of the land?

As it turns out, there was. I decided I was going to start collecting some data on my opposition, to see how I measured up to them. To begin with, here’s a refreshed view of my standard hangar as of this writing.

JayTable.png

After many matches, I’d take an extra few minutes and jot down information about the other eleven players in the pool with me. I’d take note of their player statistics, how many and what bots they ran, and their highest level weapons (the latter two stats being the principal drivers behind hangar points). Although my sample size is not comprehensive, what I found genuinely surprised me. I collected data on 106 players (in theory it should be divisible by eleven, but I closed out of a window prematurely- whoops), and here are the conclusions.

Average Player Level: 13.24

This is an interesting metric, in part because of how little it actually tells us. Player levels in War Robots have very little to do with playing skill, and even as a metric for “playing time” it leaves some ambiguity. One of the early “promo offers” new players get is a $5 package that includes a heap of Gold, a fair amount of Silver, and three days of Premium.

I bought that package, and played quite a bit during those three days, capitalizing on the +25% experience buff (I had yet to learn how empty player levels are). Did that make me a good player? No, it made me a new player who played a lot of games and still had little idea what he was doing. That other player, the one who didn’t buy the special offer? Point for point she’ll be a better pilot, because she had to play more to get to the same level.

Still, it’s probably fair to say that an average level of 13.24 shows a fair degree of inexperience in the player pool. I did, however, find one Level 30 player in my population. Running all MagGeps. What were the chances?

Average Trophies: 31.3  –  Median Trophies: 16.5

The trophy count is a record of performance over the last 10 days, and players are awarded one trophy for every 100,000 they do in a single match (rounded up). A sharp discrepancy between mean (“average”) and median here means that you have some outliers that are pulling up the average. That Level 30 pilot, for instance, had 465. A mere six players had trophy counts of 100 or more. By way of comparison, I myself typically hover around the 300 mark (though I did not count myself in the population pool).

The median is more illustrative here of what you’d see most of the time in Bronze. Let’s take a moment to unpack that number.

First, we’ll multiply it by 100,000, which is a fair shorthand for damage. Yes, some get a trophy for 51,000 points of damage, but others will get one for 149,000. This should more or less come out in the wash, or even be slightly high (since 51,000 points of damage is an easier feat than 149,000). Next, we’ll divide that figure by 10, to represent the length of time these trophies are tracked by the War Robots system.

For those keeping score at home, that means our median War Robots Bronze pilot is inflicting 165,000 points per day on the opposition- only in matches where they inflict a minimum of 50,000 points. Again by way of comparison, my approximate 300 trophies means I’m inflicting three million points of damage a day.

Vive la difference, non?

Of course, this metric is also subject to corner-casing, as a veteran player might return after a two-week break and have to start rebuilding the trophy case from zero.

Percent of Players with Three Hangar Slots: 94.34%

I played a game last night (not included in my population, it was right before bed) where one poor sonuvabitch was rocking a single hangar slot. Although I’d like to think it was some hardcore gamer simulating a permadeath environment, in all likelihood it was the newest of glistening newbies who somehow got lost in his Destrier along the way.

The matchmaker has a sadistic sense of humor.

Still, it really takes very little up upgrade your hangar. It costs an absolute pittance- 5000 Silver– to unlock your second slot, a mere 100 Gold for your third. It’s not always apparent to the novice player, but there are advantages to more hangar slots that goes beyond the idea of “lives” in a video game.

First, it’s an increase in force projection- and a fairly substantial one. Napkin math tells us that a player with three hangar slots has a bonus of 50% over a player with two, but knowing you have another bot in waiting can also give you the liberty to take chances and play a little riskier when you need to. I’ve spent a lot of time flanking in my Cossack after an early beacon grab, but with my Orkan Cossack as backup I’ve now been emboldened to really get in amongst the enemy, slipping behind lines and really wreaking havoc. It doesn’t always pay off, but it’s nice to have the ability to try without costing myself (or my team) too much on that line of play.

Second, it lets you broaden your threat diversity. I’ve seen some players- good ones- run hangars full of clones. Clearly its a playstyle they enjoy and want to keep the party going. On the other hand, by having a variety of tools in the box, you have the ability to pick the one best suited to the situation you’re in. For instance, after losing my opening Cossack, I asses the battlefield and see whether the MagGep or the Zeus Schutze will do more to keep the momentum going. The former offers DPS second to none at this level, but the longer range of the Zeus is wonderful for boards like Springfield, Canyon, or sometimes even Shenzhen.

In short, I may joke on the Wiki Forum that “fifth slots are only for pilots expecting to die four times,” but the power upgrade of each hangar slot is considerable. Whenever I’m on my fourth bot in a battle we’ve won, I often wonder if we would have won if I didn’t have that advantage. Seeing this data really brings into focus just how much of an advantage I have. Only the MagGep level 30 pilot and two other players out of the 106 in my surveyed population had four hangar slots.

An equal amount had two.

Average Robot Level: 2.10

I shit you not. This was one of the real surprises. Conventional wisdom says you can get your bots up to level 4, with weapons up to level 6, and remain safely in Bronze Tier. The degree to which players failed to do this was astonishing. It’s not that leveling them up to level 4 costs all that much or is time-consuming, but many players either don’t bother to do it, or don’t know to do it.

Here’s how the bot levels shook out across the entire population.

robotsatbronze
Fancy!

Average Highest Weapon Level: 3.86

I could have gone whole hog on this and noted all of the weapons each bot used, but since the “hangar points” metric for the matchmaker only takes the highest weapon level into account, that’s all I felt I needed to do. Good news here- players at this level at least seem to take to heart the common guidance that weapon levels should be two levels higher than the bots. Huzzah for progress!

Percentage of Clubbers in Population: 0.94%

Yeah, that’s one clubber. And by clubber, I used the most liberal hardware-based definition here. Only one player (the aforementioned MagGep guy) qualified as someone who was deliberately taking full advantage of the player imbalance to harvest Gold or get his puppy-kicking rocks off. One.

In fact, let’s look at one more bit of data, which is the bot types that I encountered during my data collection.

Table.png
The Stalker was the real surprise

This was of particular interest to me. I hear a lot about clubbers anecdotally, how they ruin the game, how the number of new players joining is dwindling (which I suspect is impossible to actually validate unless your paycheck says Pixonix on it), how War Robots is floundering. And for the avoidance of doubt, I am not presenting these conclusions either as conclusive, or to invalidate the experiences of others.

An apparently-recent player posted in the Wiki Forum just today about how he ejects when he sees clubbers in Red or Blue, because they suck all of the fun out of the game for him. I’m not going to tell him that based on my small sample size, his problem really isn’t a problem. Either MagGeps or the matchmaking system are broken, and that’s a genuine concern.

But I have to wonder if it isn’t a little bit like the shark issue I lead this piece with. Shark attacks aren’t terrifying because they’re common; they’re terrifying because they completely fucking suck. These days when I train my three Magnums on some unfortunate soul in DEAD_CITY’s center beacon, there’s a small part of me imagining how shitty that has to feel for the poor enemy pilot whose will to live is rapidly decreasing in direct proportion to their life bar.

Sixteen shark attacks a year, and I’ve never felt comfortable swimming in the oceam for my entire life. It’s not so inconceivable to think that while clubbers are more rare than they would appear given the amount of conversation they generate, an encounter with a five-slot hangar filled with MagGeps or PinGeps is a surefire way to get some new player to decide to leave the water and never look back.

As I mentioned earlier, it may well be that the upcoming matchmaker change next month will completely upset the clubber’s apple-cart, and as someone for whom an argument for “clubber” designation might fairly be made, I surely hope it results in better pairings- and stiffer competition.

I love Bronze Tier. I don’t want to leave it, at least not yet. And even if someday I do, I’ll keep some Bronze-ready bots in storage to come have fun every now and again. But in part as a result of this soul-searching, I’m looking for ways to make it a more level playing field for my opponents. More often than not these days, the MagGep never sees the pitch. Sure I’ll bring it out if things are looking grim, or if the Reds have a clubber that needs countering, but I’ve actually considered swapping it out for a Destrier, but then this wouldn’t be much of a “Gepard Diary” if I did. I love my Gep. I paid a good $5 for it!

But am I a “clubber?” I wouldn’t have thought so. I’ve been laboring under the definition that clubbers are the asshats with the full-MagGep hangars, but on most every metric I stack well above the competition at this level. On the other hand, people who dedicate themselves to better their gameplay, who study maps and read the Wiki and practice, practice, practice… they don’t deserve that label either.

So is it my fault, for sticking around despite across-the-board performance advantages? All-Gepard players for giving rise to the idea of “clubbing” in the first place? Pixonic’s fault for having a matchmaking system that sticks veterans and novices in the same pool? Pixonic’s again for not allowing Bronze Tier to live as its own wonderful thing, but simply as a bridge to Silver and Gold?

And given all that, how the hell do you explain this monstrosity, from last night?

Sinned.png
Forgive me Father, for I have sinned…

Let’s see what next month brings us.

10 comments on “Club Dredd- A Statistical Perspective

  1. What requirements do i have to stay in bronze tier

    Like

    • Hey Oblivion! That’s a great question, and one I’m not entirely certain about myself. As I’m coming to learn, Ters aren’t discretely defined. I’ll be posting a new round of stats later today (most likely), which basically show how my Bronze experience has changed from one week to the next. Although there seems to be a lot of conventional wisdom out there- which mainly seems to state to be in Bronze you should keep your bots at Level 2-4 and weapons from 4-6, there are some exceptions to this. Some Gold/WSP bots add a “premium” to the matchmaker, so if you have one of those then you might need to shade it a bit lower.

      My current hangar is:
      4 Schutze 6
      4 Destrier 5,6
      4 Golem 6,6,6
      4 Cossack 6
      2 Gepard 6,6,6

      This still has me in Bronze Tier.

      The golden rule is, if you see the “snow map” (Yamantau), then you’ve left Bronze and are in Low Silver. As long as you don’t see that one, you’re running Bronze.

      Hope this helps! Like many players, I long for a “source of truth” myself when it comes to this, but I can understand why Pix doesn’t want to reveal it, too.

      Like

  2. Oh and break jt down as simple aa possible pls

    Like

  3. Pingback: Doom Blade versus Counterspell – Gepard Diary

  4. Great read. I’m glad there’s another one out there who’s also into improving skills and gathering data. High five! Although, to be honest, at just 11 days of play, I’m still a noob. Level 24 but a noob nonetheless.

    I’d like to point out though that it may be that you’ve gathered data at the wrong time of day, as the clubbers are very common when I do play. I don’t have the data now, but I’ll try and count. I’ll send you the numbers when I have em.

    Also, I don’t know when this “update to the matchmaking system” is. Has it been implemented as of Jan 15? I have all lower lvl bots and weapons than you and yet I still get to see Yamantau once in a while. 😦

    Like

    • Case in point: I spent my first two minutes and a half in WR today getting pummelled by not one clubber, but TWO. I managed to kill one maggep and one aphid gep (using two separate bots) before the match was over at around the 2 minute mark or so.
      So nah, as much as I like the shark intro thing, it’s not really a once-a-year attack for me.

      Like

    • Hey Kenn, welcome to War Robots! Sounds like you’re enjoying yourself, getting past the clubbers to hit L24 in less that two weeks. Well done!

      I HAVE heard from a few folks that clubbers are more prevalent at some times than others. I’ve tried to space out when I data capture to see if I can account for it, but thus far it’s eluded me. Now that I’ve moved to Silver, though, I’ll see it even less. In fact, I love seeing MagGeps now that I’m running a Boa hangar, because crushing a clubber feels great after being subjected to them for so long.

      I’d love to see what numbers you come up with. What I’ve recorded after every match are trophies, player level, hangar slots, and the level of each bot and level of its highest weapon.

      As for the MM system, it should be right around the corner. Some folks think it’s even trickling in a little already. Definitely check out the War Robots forum if you haven’t already, lots of great info and insight there!

      Like

  5. Pingback: By the Numbers: The New Matchmaker – Gepard Diary

  6. Pingback: To Each According to Their Need – Gepard Diary

  7. Pingback: What Does it Take to Win in War Robots? – Mech*Spectrum

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: