[interface] Percentage Statistics

Brainstorm ideas of possible additions to the game. Read this before posting!

Moderators: Forum Moderators, Developers

Forum rules
Before posting a new idea, you must read the following:
Post Reply
User avatar
Cackfiend
Posts: 436
Joined: January 28th, 2007, 7:36 am
Location: Florida, USA
Contact:

[interface] Percentage Statistics

Post by Cackfiend » January 9th, 2012, 3:45 am

Currently the game has Expected Value (EV) that is based off of the percent to hit and the amount of damage that is expected. I think this statistic throws people off of how often their units are actually hitting/missing.

For example:
2 strong spearman and 1 non strong fencer attack some grunts on grass at day. The damage output is 10-3 10-3 5-4 which is 10 attacks. The spearman hit 5 of 6 swings and the fencer hits 1 of 4 which is a total of 6 of 10 or 60%. The percentage of hits is right on the money of what the game predicted since the grunts were on grass, but since EV is based off of damage the EV is actually +14% Inflicted (55/48) and generally leads to people to start complaining about EV.

Example 2:
2 footpads and a thunderer attack a glider at day. The footpads do 3-2 and the thunderer does 20-1. Both foot pads miss and go 0 of 4. The thunderer hits 1/1. Because the thunderer did so much dmg the EV is actually +4% 20/19.2 but your units only hit 1 of 5 times or 20% on a unit that the game expected you to hit 40% of the time.




Could we have something also that basically just says what your average % to hit has been and how often you have been hitting?

Like... out of 96 swings, at an average of 58% to hit, you've hit 54% of the time

in this example, people would know that theyve been missing slightly more than the game had anticipated.



I personally would REALLY like to have this stat, and it would shut a lot of people up about the RNG/EV :)
"There's no love in fear." - Maynard James Keenan

I'm the guy who's responsible for 40% Gliders in all hexes... I can now die a happy man. =D

tsar
Posts: 1
Joined: April 28th, 2011, 7:42 pm

Re: Percentage Statistics

Post by tsar » January 9th, 2012, 9:51 am

DISCLAIMER: I agree that it's important not to complain about luck - it's a part of the game. I just don't think this will help.

EV is actually a MORE fair way to judge whether rng favors one side in any given match. In fact, both examples you give here show exactly why this is the case. Since some units are heavy damage dealers and others are not, it means that some hits are more important than others. In example 1 it is more important for the strong spearmen to hit than for the weaker fencer to, ditto for the thunderer in example 2, and the EV reflects this difference.
We can make even more extreme examples. If you have a bat and an adept attacking a drake on 30%, then 2/4 total hits could result in either an almost full health drake unit or an almost dead one. Yet only one of these outcomes could be considered "lucky". Again, EV shows this difference and raw percentages do not.
Knowing how often people are hitting/missing is only half the story. Not all hits are created equal.

On the other hand, I think that having more information is always better. :geek: There's certainly no harm in including a little additional statistical data...

User avatar
Cackfiend
Posts: 436
Joined: January 28th, 2007, 7:36 am
Location: Florida, USA
Contact:

Re: Percentage Statistics

Post by Cackfiend » January 9th, 2012, 3:50 pm

tsar wrote:Yet only one of these outcomes could be considered "lucky". Again, EV shows this difference and raw percentages do not.
Knowing how often people are hitting/missing is only half the story. Not all hits are created equal.

On the other hand, I think that having more information is always better. :geek: There's certainly no harm in including a little additional statistical data...
of course you are right, but i dont really care about that :P

i just want to know how often my units are actually landing, not how much damage the game expected
"There's no love in fear." - Maynard James Keenan

I'm the guy who's responsible for 40% Gliders in all hexes... I can now die a happy man. =D

Yogibear
Retired Developer
Posts: 1086
Joined: September 16th, 2005, 5:44 am
Location: Hamburg, Germany

Re: Percentage Statistics

Post by Yogibear » January 9th, 2012, 7:58 pm

Cackfiend wrote:i just want to know how often my units are actually landing, not how much damage the game expected
The question is, why would you want to know that?

Maybe this would be interesting if you suspect that the RNG is flawed. Which i am very sure it is not as i have seen the code implementation and every systematic check on the RNG proved that it works very well.

It could be interesting to know if throughout a single game the number of hits is great enough to satisfyingly even out or if the (statistical) sample is so small that higher deviations are to be expected. But then again, what are we going to do with this information?

Or it could be interesting to get more silent people who like to complain about the RNG, which i understand is what you have in mind. But honestly, i don't think it's going to work. Look at people playing with a perfectly balanced EV. What is happening?

Some (certainly not all) of them are saying: "Ah, yes, of course, the EV is balanced so it looks like the game went fair. But as you know, there are attacks that are more critical then others. See, if i had hit like expected and killed that blocking unit, it would have opened me the way to kill two more low hp units behind it and that certainly would have won me the game."

So people start to blame win or lose on certain key attacks. I don't want to go into the matter of them being right or not. Probably sometimes they are and mostly they are not. But that's not the point.

The point is, they are looking for a cause why they lost. And it's very human to search for a cause that lies outside of yourself :) . I am not even going to blame people for that. They are what they are, simple as that. And because of that i think they will rather look for an RNG excuse instead of analyzing the game and finding a critical mistake that maybe had a much greater influence on why they lost. No matter which stats you provide.

Just for the record: I am not saying that the RNG influence is negligible if you play right. Anyone facing a lucky orc or drake rush (or any other faction for that matter) will have to admit that there is literally nothing you can do against it, no matter how sophisticated you play. One could start to discuss if this is wanted but that's not what this thread is about. And this discussion also is not wanted of course.
Smart persons learn out of their mistakes, wise persons learn out of others mistakes!

User avatar
Rigor
Posts: 941
Joined: September 27th, 2007, 1:40 am

Re: Percentage Statistics

Post by Rigor » January 9th, 2012, 10:50 pm

Yogibear, Cackfiend wrote:
- i just want to know how often my units are actually landing, not how much damage the game expected.

- "Ah, yes, of course, the EV is balanced so it looks like the game went fair. But as you know, there are attacks that are more critical then others. See, if i had hit like expected and killed that blocking unit, it would have opened me the way to kill two more low hp units behind it and that certainly would have won me the game."

- And because of that i think they will rather look for an RNG excuse instead of analyzing the game and finding a critical mistake that maybe had a much greater influence on why they lost. No matter which stats you provide.
yes thats so right!!! SMILIES! Read the Posting Guidelines, point 1d.

maybe what cackfiend wants to express is that he feels that the EV is not really telling you that you lost because of a bad RNG, he needs bad RNG and "situation bad luck" with strikes that went wrong.

you could even spin it a bit further and not include the retaliation strikes in your analysis: only strikes that went wrong WHEN IT WAS YOUR TURN would make the whining about RNG close to perfect. as an extra, i think somebody HAS TO think about creating an addon that makes a screenshot just before an extremely gruesome situation is going to happen and make a mini-icon for the savegame so you know which game it was.
Last edited by shadowm on January 9th, 2012, 11:34 pm, edited 1 time in total.

User avatar
Cackfiend
Posts: 436
Joined: January 28th, 2007, 7:36 am
Location: Florida, USA
Contact:

Re: Percentage Statistics

Post by Cackfiend » January 10th, 2012, 4:56 am

well of course there is EV/RNG complaining and then there is luck complaining


luck is like when you miss 0/4 in order to level your unit that is about to die, or like you said where you fail to kill a unit that would have exposed low hp units


regardless, this statistic would still be awesome to have so I can shut people up (or justify them) when they complain about how often units are hitting/missing
"There's no love in fear." - Maynard James Keenan

I'm the guy who's responsible for 40% Gliders in all hexes... I can now die a happy man. =D

igor
Posts: 11
Joined: September 14th, 2010, 5:46 pm

Re: Percentage Statistics

Post by igor » January 13th, 2012, 10:35 pm

Luck is important part of wesnoth, so to count luck is very useful.

Yes, luck more important in some situations. For example, luck more important when you attack wose with mage than you attack clasher with mage. So, if we count luck in this both attacks (with same luck) like a percentage statistics it will gives a same digits. But any player will say better have luck with wose attack :), and for analyze luck impact in the game for us more important to know which luck was with wose attack. Luck with clasher not so important. Thats why I propose to count a luck not in clean form (like a percents). I propose to count luck like a difference between health damage with normal luck and taking (real) damage in the game. In that case in our example with lucky mage attack at wose this value will be more than with attack at clasher with the same big luck. Indeed, impact of this attack (with wose) more important for the game. Similar, if mage do unluck attack at wose that will be much greater unluck in comparison with attack at clasher with the same unluck.

I have a same propose here (at forum). Today I calculate per turn luck statistic (used my method) in Excel and attach this file to my topic. You may try to change values and see how my method work.

http://forum.wesnoth.org/viewtopic.php?f=12&t=35846

AlexanderK
Posts: 24
Joined: August 4th, 2011, 3:41 am

Re: Percentage Statistics

Post by AlexanderK » January 14th, 2012, 2:27 am

When you play campaigns sometimes you want to miss and when you miss you can consider it as luck, for example when you attack yeti or level 3 enemy troll you want them not to die so fast so you can get some extra xp.

hiro hito
Posts: 201
Joined: November 23rd, 2006, 8:00 am

Re: Percentage Statistics

Post by hiro hito » January 14th, 2012, 10:02 am

I am always amazing that some topics still keep going on!...

Having some relevant stats about luck in the game will be nice.

As far as we all already know, there will always be luck complainers (as me) on server and/or forums.
Having enough proof/argument about those complains will probably make some complainers, or rng defenders, shut up!...
"Of course His Majesty is a pacifist. When I told him that to initiate war was a mistake, he agreed.Thus, gradually, he began to lead toward war."-Emperor Shòwa (Enlightened Peace)'s chief cabinet secretary

User avatar
Cackfiend
Posts: 436
Joined: January 28th, 2007, 7:36 am
Location: Florida, USA
Contact:

Re: Percentage Statistics

Post by Cackfiend » February 14th, 2012, 1:09 am

so any chance any of the devs could code this into 1.11? :)

its always nice to have more options!
"There's no love in fear." - Maynard James Keenan

I'm the guy who's responsible for 40% Gliders in all hexes... I can now die a happy man. =D

Post Reply