War Robots WR Data and Analysis

By the Numbers: A League of my Own

Jay takes some sample data from his first ten League matches. What's changed- and what hasn't?

As I wrote Tuesday, the League system has dropped for those of us in iOS after the Android folk got to work out some of the bugs. Only fair, I suppose, since we were the ones who had to endure the new matchmaker rollout first.

I played my five placement games, which found me assigned to the Diamond 3 division. I then played ten games, recording the statistics of the players I was grouped with each time (as per usual). The results were not what I was expecting.

Here are my own numbers at time of writing:

Level: 30

Cups: 224 (-134)

Victories: 681 (+97)

Now, let’s take a look at the numbers!

header

Average Player Level: 29.34 (-0.58)

No surprises here. The dip is marginal. I’m a level 30 player up against high-level competition, don’t stop the presses.

Average Victories: 841.56 (+100.95) 

That’s a noticeable jump. I use victories as a proxy for experience (whereas cups I use as a proxy for activity), so this tells me that I’m in a pool of players that’s even more experienced than before. That’s not entirely unreasonable to expect- after all, my victories has gone up 94 in the same period of time.

Average Trophies: 430.62 (+46.34)

Now this is interesting, in that both “experience” and “activity” have both climbed at an approximately equal pace. Not only am I facing players who have been playing longer, but I’m also up against those who play more frequently- at least within the last ten days (as cups measures).

Is the matchmaker savvier in its picks? Are there factors in the League assignment that aren’t immediately apparent? Did the matchmaker rollout crater some of the ranks of the more casual players, who simply gave up and did something else? It’s impossible to say, but this raised an eyebrow.

It certainly lets me feel a bit like an underdog, since I trail in both categories.

Player Divisions

Here’s a new one! What sort of league opposition is the matchmaker sending my way? Turns out, this raises some very interesting questions as well.

divs

Most of my opposition comes from Diamond 3, which is my own division and makes perfect sense. A further 26.61% come from players who haven’t played enough matches yet to qualify for placement. Okay.

But compare the difference between one division lower (Gold 1, 31.19%) and one higher (Diamond 2, 1.83%). That’s staggering. I fought nearly as many Gold 1 players as I did Diamond 3 players, which means that the matchmaker is quite happy to take players a division lower than me and put us together in matches. It would stand to reason, then, that the Diamond 2 matchmaker would be just as greedy about “reaching down” to fill its matches. It wouldn’t be unreasonable to expect that I might occasionally wind up as fodder for those slightly above me, just as the Gold 3 and Gold 2 players in particular were.

But it never happened.

Granted, my sample size isn’t colossal, but given the disparity between both outcomes, it seems that something more might be at work here. One possible explanation is player population. We might well have an “equal opportunity matchmaker,” but I’m not getting picked for up-division matches because there are much less of them. After all, the League system uses static criteria for divisional assignment, not relative ones (it’s not “grading on a curve”). Could we have that many fewer Diamond 2 and above players, compared to Diamond 3 and below? How steep are the sides of the pyramid?

header

Regular readers will recall that I used to track what proportion of players had three hangar slots versus others, but as of the previous article that statistic has become obsolete. I’m no longer in territory where such a narrow look offers useful insight. Going forward, I’ll simply be looking at overall hangar slot representation.

Here’s what my peers are rocking:

slotsgraph

Not a huge change, though it does run contrary to expectation. If anything, I’d expect the number of players with a full five hangars to increase, not decrease. Hangar slots are a very loose correlate to player dedication and competency, in that we naturally expect the best players to fill out their hangar. “Slots before bots,” and all that. But I’ve seen some very competent players rocking a three-slot roster, and there are good players out there that are relatively new and still squirreling away every Gold piece they get for that grindy fifth slot.

Average Robot Level: 7.64 (-0.08)

Average Highest Weapon Level: 8.18 (+0.07)

These are razor-thin changes, but fascinating ones all the same. The last stats post was on 09 February. This means that my level of competition- with regards to bots and weapons- has stayed on a plateau, even as I’ve spent all that time upgrading.

There’s a slightly concerning interpretation here, in that while my gear has gotten better I’ve actually flatlined in terms of player development, or am running a suboptimal hangar. I’m not going to worry about it too much until my next stats post, because the introduction of the League system is something of an x-factor here. It could simply be that the matchmaker is taking my division into account when finding matches, so I’m not truly comparing apples to apples.

But if I was assigned to this division because of offensive output, then that sort of closes the circle. I have two controversial builds on my active roster, the Pin Stalker and Hydra Doc. I had both bots on the roster from the 9th, but I had just changed from the steady Tulumbas to the more experimental Hydras.

Food for thought.

What Bots are People Running

If that curious bit of consistency wasn’t enough for you, have a look at this:

woche

When the only things you can really highlight is that the rare appearances of a Butch or Jesse have gotten rarer still, there’s just not a lot to cover. Across the board, only minor variations from the previous sample. This indicates that the metagame is a fairly settled affair at my level of play. “TT Furies are ruining the game” is a common grouse I hear, but I’m a pretty far cry from that world at present. It will be interesting to see as I continue to progress when the numbers will start to change.

header

It’s been a curious update. We’re seeing a high degree of consistency in what I’m up against on the field, but more substantial changes to the people who are piloting them. If I’m worried about skill stagnation, perhaps I can take some heart that I’m not alone.

I’m excited about the League implementation. I’ve had very enjoyable matches in the wake of it, and I hope that continues to be the case.

1 comment on “By the Numbers: A League of my Own

  1. Pingback: Battle Stories: Strange Brews – Gepard Diary

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 )

w

Connecting to %s

%d bloggers like this: