Telemetry and Game Balance 8.0.0


#1

A new patch cycle, a new telemetry thread! As always thanks to @MacMan and all the other Developers who provide us with new data. Now that we have @citizenphnix with The Thunderchild, telemetry data should prove to be even more informative than ever before! Feel free to ask questions and discuss recent telemetry here!

Archives:


Telemetry and Game Balance 8.1.0
Average damage outputs?
Telemetry and Game Balance 7.0.1
Telemetry and Game Balance 7.0.1
The Index, Your Primary Source for the Meta of Evolve
Telemetry system
#2

#3

#4

Totally!:grinning:


#5

@MacMan @Insane_521

Do you know how much damage Kala’s Siren Missiles do to the minions?
I felt like I did nothing to the minions in Defend.


#6

Check out this thread:


#7

Thanks Boss!


#8

Hoping to start sharing some new telemetry data tonight!


#9

YUS! If you remember could you tag me when its out? No big deal.

Btw Kala is kick ass. Before I got her elite, I couldn’t stop playing with the teleporters instead of chasing the monster. A very awesome character, and looking forward to that Kelder release as well!


#11

Thanks! We love Kala too. She is a very daring add to the Evolve world, both on the design side and on the fiction side. I am definitely looking forward to what she means long term for the Evolve franchise.


#12

@MacMan, you are the second TRS employee to (recently) make a comment regarding long term plans for evolve. Could you please speak a little to that? I don’t think you guys have given any indications of a T6 being in the works, and as a business you obviously can’t live off of free adaptations.

On that note, sincerely thank you for all the free adaptions released so far and the ones yet to come.


#13

It’s been roughly a year since Evolve released. It’s 8:29PM and I’m still at the office working on Evolve and so are many others. I’ve been working in games for nearly 20 years now and have never experienced anything like it. Usually you ship a game, get a few weeks off and move directly onto a sequel (if it was successful) or something else entirely (if it wasn’t.) You go dark for X number of years again until the next marketing push on the next product.

We love Evolve and are very thankful that we have the opportunity to keep working on it, to keep trying different things and to have the community be part of it all, giving feedback, sharing ideas and experiences. As a studio owner and design director on Evolve, I honestly couldn’t have asked for more. Life is good, and I am excited about the future!!!

The community’s enthusiasm for the adaptations has breathed new life into Evolve. We are indebted to you all!


#14

Now… lets see some Thunderchild data!!! @SQUARENecron

There is a LOT of new stuff in here so I’ll just start exploring and sharing interesting stuff that I find. This data is ONLY from today btw, so I don’t want to pull a lot of solid conclusions from it but we can already start to see trends.

If you followed any of the telemetry threads for very long, you know that one of the things we monitor is Damage output from each character. However, we wondered if that data was really relevant. The theory was, if a character does a ton of damage quickly, maybe they finish their opponent very quickly and as a result, their damage numbers might look low. Or conversely, a character with low damage output might have trouble finishing an opponent, causing the game to go long artificially increasing their damage numbers.

So now we plot average Damage Per Second and compare it to average Damage Per Round. Here is what we find. It actually lines up very nicely!!! The black line represents damage per second of each character each round multiplied by the duration of each specific round. The solid color bars represent pure average damage per round, duration of each round not taken into account. This is good news for us. It means that we were making pretty good assumptions based off of the data we had.

This also proves that we can glean more from the new data. Take a look at the monster damage chart below. Wraith’s average damage per round is very high compared to her damage per second per round. This data is new to me but what I think that means is that she often has long, drawn out games compared to the other monsters and that is what is contributing to her high average damage per round. Her actual DPS/Round is fairly low.

The same logic and data would indicate that Parnell and Lennox finish rounds quickly because they have so much damage potential, when looking at the Hunter chart.

It’s also worth noting that Elder Kraken is primarily played by bot, so ignore that damage output for now.


#15

So what are people playing when they jump into a game of Evolve? Here’s how it breaks down:

As expected, Most people play Hunt, followed by Arena, Defend, Nest and Rescue. No real surprises there.

But when we dig a little deeper, it starts to get interesting.

Most people play Quickplay. That’s good for us. That’s what we expected, and the next largest group plays Ranked. But whoa, there are almost as many players playing “custom games” with their friends as there are playing Ranked. Only a 1% difference. Dang! We may see a stronger internal push for additional features in custom game mode as a result. Also interesting, there are almost as many people playing co-op evacuation as versus evacuation.

Ever wanted to know exactly HOW games were being won in Hunt? Turns out ~9 out of 10 rounds are won by defeating the other team directly. Only 7.2% of Hunt rounds are won through the generator! That is much lower than expected, but we are very happy with that result.


So you guys have been talking the next Micropatch lately
#16

YESSSS DO IT PLEASE THANK YOU.


#17

Meanwhile playing Goliath vs WLM

I kid :stuck_out_tongue:
But seriously this sounds like the data presented only scratches the surface of Thunderchild after reading what it’s capable of. This has me extremely excited to see how everything balance wise plays out with such accurate ways of interpreting the game via data.


#18

THUNDERCHILD DATA RULES!! This is amazing :grin: Thanks for sharing!

For reals, that savage maggie really IS savage. man, that is terrifying :fire: :laughing: :fire:


#19

Time to look at some medic data!

Who does the most healing in Hunt games??? Similar to the damage charts, this shows average healing per round compared to healing per second per round.

I think the big difference with Rogue Val is that she is constantly healing multiple targets. Much higher healing output overall.

Who are the medics healing? This one surprised me.

Of all the people that need healing, I didn’t think Assault would be it. They have a shield. Then again, perhaps the fact that they have a shield means that they tend to be alive for longer which means they can get healed more often. Assaults also tend to be very aggressive and in the monsters face so maybe that is the issue. Or maybe since they don’t have a shield, they get lazy? This is one of those instances where more data creates more questions.

Supports needing the least amount of heals makes sense with that fancy cloak! They can take care of themselves!

Let’s compare medic win rates to see if they have an immediate correlation to the healing output (which would make logical sense.)

Iiiiinteresting. Emet has the highest win rate, though he is ranked #5 in healing output. Val and Caira have low win rates even though they are ranked #2 and #3 in healing output. In Evolve, being a good medic is more than just healing!

Here is a fun one. How many strikes and deaths to Hunters get when accompanied by each medic?

Obviously teams who play with Lazarus carry the fewest strikes but they are expected to die a lot. However, Emet is surprising. This may help explain why even though Emet has a low healing output he has a high win rate. People get strikes and they die a lot when he is on the team, how can they win?!?!?! Respawn Beacon!!! Like I said, Evolve medics are more than healing!


#20

Ok, sorry to double post, but if ANYBODY asks me why Evolve is so different from other shooters with classes, First, I’m showing the monsters, THEN I’m showing this data. The dynamic of “most healing medic isn’t best medic” is revolutionary.

WHY ARE NUMBERS SO COOL!?

EDIT: I actually notice that medic win rate correlates directly with damage done. I wonder how a max damage comp would fare against monsters. Hyde, Bucket, Savage Maggie, and EMET could prove really scary this cycle.


#21

BECAUSE THEY ARE :smile: