Statistics dialog "Inflicted" and "Taken" unclear

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:

Re: Statistics dialog "Inflicted" and "Taken" unclear

Postby ElderofZion » October 19th, 2017, 2:47 am

Ok, so for a human that is impossible, so it is impossible to prove.

@Inky I already said that its not just overall campaign stats (though these come from the individual scenarios) even though like I said loading did affect them and later in the campaign the diffference in luck was pretty significant though the sample size was bigger than earlier when the luck was more even. I also said that after the loading in the particular scenarios i do get luck, how many times do I need to repeat myself.

If merely loading a save in the campaign makes the luck stats "biased" and the only way to get rid of that is restarting the whole campaign then it just proves my point.

I think Pentarctagon has clarified everything, no need to argue about luck any longer.
@Elder but we have asked for the disclosure of all the cheese and we treat it seriously. Everyone on this site can get banned, everyone.
@Elder ale prosiliśmy o wyjawianie wszystkich serów i traktujemy to poważnie. Każdy na tej stronie może zostać zbanowany, każdy.

- Sagez
User avatar
ElderofZion
 
Posts: 218
Joined: July 11th, 2015, 2:13 pm

Re: Statistics dialog "Inflicted" and "Taken" unclear

Postby Tad_Carlucci » October 19th, 2017, 8:33 am

Let's say your luck going into the last scenario is even but you've managed to make it. But the dies roll against you and your crushed for the final scene. So you reload that scene and manage to play through with even luck and win.

You luck is even, it's balanced, right? No.

You're biasing the long-term averages to 'good' luck by discarding scenario with 'bad' luck. On average, over hundreds, thousands, or millions of campaigns, luck should tend to neutral (0). But you discarded all negatives so, instead, you'll average to a positive value. Instead of rerolling each combat and ending with an obvious, hugely positive luck, you did it with scores that and neutral or only slightly positive.

So, if it takes millions of samples to statistically prove, how can we be positive it's fair without any? Simple. We use a known-good implementation of a random generator and show each use does not introduce bias.

Similarly, if you want to shows bias we'll believe, you will need to show an error in the RNG logic or implementation or our use of it.
I forked real life and now I'm getting merge conflicts.
Tad_Carlucci
Developer
 
Posts: 204
Joined: April 24th, 2016, 4:18 pm

Re: Statistics dialog "Inflicted" and "Taken" unclear

Postby ElderofZion » October 19th, 2017, 1:12 pm

Lets say you read my posts and understand them. But you haven't and you keep up this strawman. So I do not understand what you are trying to achieve.
@Elder but we have asked for the disclosure of all the cheese and we treat it seriously. Everyone on this site can get banned, everyone.
@Elder ale prosiliśmy o wyjawianie wszystkich serów i traktujemy to poważnie. Każdy na tej stronie może zostać zbanowany, każdy.

- Sagez
User avatar
ElderofZion
 
Posts: 218
Joined: July 11th, 2015, 2:13 pm

Re: Statistics dialog "Inflicted" and "Taken" unclear

Postby Ravana » October 19th, 2017, 1:34 pm

RNG discussions have already been done.

This topic is about making statistics UI easier to understand.
User avatar
Ravana
Moderator
 
Posts: 1460
Joined: January 29th, 2012, 12:49 am
Location: Estonia

Re: Statistics dialog "Inflicted" and "Taken" unclear

Postby Tad_Carlucci » October 19th, 2017, 2:58 pm

Other than the floating text suggested some time ago, I don't see how to clarify it in the space provided. So I'd suggest a wiki page reference on the screen and describing things there.

Damage inflicted is the running total based upon the actual combat results. The expected value is the maximum possible, adjusted for terrain, etc. Damage taken is the same values, but from the opponents sides. The toolhelp suggestion is a floater to explain the percentage is the ratio of actual to expected. One set of values resets at the beginning of each turn, while the other only resets at the very start of a campaign.

We commonly call these percentages luck because the main factor is randomness. Being controlled by randomness we shoul not expect them to be exactly 100% or that inflicted vs taken should show a relationship. We should, in fact expect surprisingly large variations although, given the factors, this should be exceedingly rare.

Loading, whether to restart a scenario, or rerun a combat will skew these values (generally in the player's favor). Save/loading for combat results will cause a huge swing.

So, while we call it luck, these values are more appropriately used as a gage for cheating. We just have to remember that, while extremely unlikely, a false-positive is possible.
I forked real life and now I'm getting merge conflicts.
Tad_Carlucci
Developer
 
Posts: 204
Joined: April 24th, 2016, 4:18 pm

Re: Statistics dialog "Inflicted" and "Taken" unclear

Postby revolting_peasant » October 21st, 2017, 4:14 pm

Tad_Carlucci wrote:Other than the floating text suggested some time ago

That's a super-easy default option which, in case nothing else is done, I really request be added.

Tad_Carlucci wrote:I don't see how to clarify it in the space provided.

Instead of:
Code: Select all
5739 / 5629.8 +1%
you could just say:
Code: Select all
1% over expected
and have a tooltip list the number. Or possibly
Code: Select all
5739 (101% of expected)
, or
Code: Select all
5739 (1% over expected)

Of course, if you're willing to use more space - which you should be, as the previous dialog is wider - then there are two good options in my opinion:
  • Add an extra row as part of the column header, so that we have:
    Code: Select all
    Overall           This Turn
    Actual/Expected   Actual/Expected
    -------------------------------------------
    X / Y (+1%)       etc.
    etc.              etc.
  • Go for the wider
    Code: Select all
    Overall              This Turn
    -------------------------------------------
    X / Y (expected+1%)  etc.
    etc.                 etc.
User avatar
revolting_peasant
 
Posts: 56
Joined: May 29th, 2012, 5:45 pm

Previous

Return to Ideas

Who is online

Users browsing this forum: No registered users and 5 guests