Graphing Champion Damage

12
Comment below rating threshold, click here to show it.

Zoopercat

Junior Member

08-29-2012

Howdy! Long time player, first time poster

I'm one of the people behind Ask Mr. Robot, and we just launched a tool that lets you play around with builds and compare results. It's easy - you get to use a drag and drop interface to make your build, then we graph it for you. You can fight our amazingly qualified target dummy, or another champion. Here's a walkthrough on using the builder.

We also created a section that recommends builds, analysis, and gameplay tips for each champion.

Here are some example analysis: take a look at the graphs on the analysis page for Teemo to get answers to some top questions, like "Is pure AD or On-Hit better for Teemo?" and "Should I get some Attack Speed items?". Or check out the analysis for Diana to see a comparison of AP builds, find out if you should get a Wriggles Lantern, and learn how to max your burst damage.

For example gameplay tips, check out Teemo, Ashe or Fiddlesticks. We want to involve the community, so if you have tips, just leave it in the comments on each champion's page.


Beta! 49 Champions are live, more launching each week.

  • This is still in beta, although we vetted each champion that's currently on our site.
  • All 49 champions can be used in the Build Analyzer
  • All 49 champions have builds that are good, however, only 24 analysis done.
  • 22 of the champions have tips in the gameplay section
  • We're adding new champions each week, starting with the ones available in the free rotation.


Hope you enjoy it!


Comment below rating threshold, click here to show it.

Lethadind

Senior Member

08-29-2012

It's a cool utility but it's severely flawed. Try matching a champion up against itself and see the results. Exact same items and everything, and yet somehow one of them comes out ahead by like 220 HP, with different DPS? You've got programming errors in that for some reason.

You should also add an option (if you can) for multiple practice dummies. This factors in teamfights. For instance Diana's passive and Q and W do a lot more damage in a teamfight than when chasing someone. being able to factor in this AoE damage would be cool, though I understand that it might be hard as you focus on "100 - 0" for the practice dummy, still it'd be cool.


Comment below rating threshold, click here to show it.

Zoopercat

Junior Member

08-29-2012

Lethadind, the reason one champion comes out ahead is that one of them attacks first. We always assume your champion (not your opponent) gets the first hit.

We're actually working on a feature for multiple target dummies - great minds think alike, right?!


Comment below rating threshold, click here to show it.

Lethadind

Senior Member

08-29-2012

Quote:
Originally Posted by Zoopercat View Post
Lethadind, the reason one champion comes out ahead is that one of them attacks first. We always assume your champion (not your opponent) gets the first hit.

We're actually working on a feature for multiple target dummies - great minds think alike, right?!
I realized that later.

However, the "vs." thing I *believe* is flawed still. As Diana I never lose to Akali in tradeoffs. Never. Even with similar builds I come out on top. I could not find a single build that even came close to "beating" Akali. As in with both "stock" builds, Akali beats Diana with 1000 health left. It's just not gonna happen. I'm not sure how you can fix it, but I do know it's incorrect, even if the math says it's not.

Yeah I just looked at the "doubt Mr. Robot's logic play-by-play" thing, and apparently Akali starts off using like 3 abilities + her Gunblade before Diana even gets a hit in (You said that the champ you're looking at should get the first hit, Diana's the one I'm looking at). Also, for some reason you start Diana off using her shield (I always start Q + R then W). Doing Q+R to E gives you more damage as your Q comes off cooldown faster for more QR combos. I'm assuming this "testing" thing assumes both champions are right next to each other though, which might be why it starts off with the shield.

Apparently the fight is over in like 2 seconds? Dafuq? I don't think global cooldowns and cast time are accounted for. This changes the fight drastically because Diana's shield orbs only hit Akali twice, not giving her the extra damage nor the extra shield. There's no way the fight ends before all three orbs go off in a normal situation. I realize you can't account for a lot of human interaction but global cooldowns and like an average .3 second wait time for cast-and-click changes the fight enormously, and makes the information bunked.

I'm not sure if you were looking for errors or not. I'm just trying to be helpful with what I noticed, if you didn't want it, sorry. Good luck.


Comment below rating threshold, click here to show it.

Zoopercat

Junior Member

08-29-2012

Thanks for the info. I'll look into that a bit more and make sure there isn't something whacky going on.

For example, when we were alpha testing it, we didn't have delays on Annie's abilities, so she did like 20,000 DPS or something crazy because the fight was over in 0.5 seconds. Anyway, even though we checked everything, there's nothing like thousands of people using it - someone will find something we didn't!

I'll have more info for you tomorrow, and let you know what I find.


Comment below rating threshold, click here to show it.

11b Lead the Way

Member

08-29-2012

What language did you code it in?


Comment below rating threshold, click here to show it.

Cenerae

Senior Member

08-30-2012

Do you factor movement in when taking attack speed into account? The stat always seems much better on paper than it works out in practise because nobody ever does when doing things like this.


Comment below rating threshold, click here to show it.

Zoopercat

Junior Member

08-30-2012

@ Cenerae
We factor in movement speed for champions that have abilities related to it, like Hecarim. Otherwise, we don't factor it in. The tool isn't meant to simulate the game, because it can't - it's PvP - you never know what your opponent will do. It's meant to look at your potential damage to give you the biggest edge you can get in a game.

If you want to look at movement speed, I'd suggest doing 2 different things.
1) Get the same amount of movement speed on 2 different builds, then see what other items combinations get you the most damage.
2) Remove a little movement speed from one of your builds to get a different item. Take a look at the damage gained and see if that's enough for you to justify losing the movement speed.

As for auto-attacking, there's an option to show only auto-attacks. That will help evaluate the damage-to-movement-speed tradeoff.

@11b Lead the Way: We coded it in C#, hosted on the Azure cloud.


Comment below rating threshold, click here to show it.

CordsOfThought

This user has referred a friend to League of Legends, click for more information

Senior Member

08-30-2012

Quote:
Originally Posted by Lethadind View Post

Yeah I just looked at the "doubt Mr. Robot's logic play-by-play" thing, and apparently Akali starts off using like 3 abilities + her Gunblade before Diana even gets a hit in (You said that the champ you're looking at should get the first hit, Diana's the one I'm looking at). Also, for some reason you start Diana off using her shield (I always start Q + R then W). Doing Q+R to E gives you more damage as your Q comes off cooldown faster for more QR combos. I'm assuming this "testing" thing assumes both champions are right next to each other though, which might be why it starts off with the shield.
I think what you have to realize is that it's based on calculations. I do agree though that there should be an option for what skills go off first for both characters, and for potential misses. Without this you'll have people basing it off skill which isn't what this is about. All in all it's still very detailed and appreciate the hard work he put in to it.


Comment below rating threshold, click here to show it.

Cenerae

Senior Member

08-31-2012

Quote:
Originally Posted by Zoopercat View Post
If you want to look at movement speed, I'd suggest doing 2 different things.
1) Get the same amount of movement speed on 2 different builds, then see what other items combinations get you the most damage.
2) Remove a little movement speed from one of your builds to get a different item. Take a look at the damage gained and see if that's enough for you to justify losing the movement speed.
This wasn't what I meant, and you know that given the first part of your response :P

If you can't factor in enemy champions moving away from you, then it's going to make attack speed look much better in general than it really is. Which means it's going to always be innacurate for any real sense of anything. :-/

I know there's only so much you can do with a simulation, but I don't like the idea of using something I know is going to be innacurate before I even punch numbers in.


12