Recent War Day Win %

[MatthaisUK]

Firstly, thanks for your great site.

There's an inconsistency with the "last 10" & "last 20" win percentage stats when accounting for double attack war days. My assumption was (and I feel most would think the same) that this was that statistics counted the last 10 battles, but it appears that instead it counts the last 10 entire war days, which can amount to more than 10 battles - 12 in my case, shown in this screenshot:

In that case my last 10 war day battles resulted in 6 wins (3 × 1/1, 1 × 2/2, 1 × 1/2) & 4 defeats (3 × 0/1 & 1 × 1/2), giving me a win percentage of 60% for my last 10 war day battles, but the "Last 10" statistic displayed 50%, which left me quite perplexed.

It's only after my next win was recorded and my "Last 10" stat creeped up to 58% despite my record still being 6W & 4L (60%) from my last 10 battles, that I appreciated what was happening:

In my opinion, either:

• The statistic should calculate based upon the last 10 war battles only, although I appreciate this may be awkward when the 10th battle split falls in the middle of a double attack war day with a WL or 1/2 outcome.

• The statistic should be clearly labelled as "Last X War Days (X battles)" - e.g. "Last 10 War Days (12 Battles)" in the case of my two screenshots. This would also help draw attention to the fact that some war days are 2/2 as they are coloured the same shade of green as 1/1 outcomes and could be overlooked.

Either of these options would remove the current ambiguity regarding what the statistic is referring to.

[smlbiobot]

It's last 10 war days. We are not going to change that. Clans use this info to evaluate a candidate's recent performances. If you compare multiple players and use battles instead of war days, then they won't align. The last 10 and 20 is here so you know that they are performing in the most recent wars.

[smlbiobot]

Regarding labeling, these must be viewable on desktop as well as mobile. When you have very long labels that you suggested, they won't fit on a 320px wide screen. Our site is responsive and aims to be usable for users on mobile or desktop. There will be trade offs.

[MatthaisUK]

Clans use this info to evaluate a candidate's recent performances. If you compare multiple players and use battles instead of war days, then they won't align.

Clan's will primarily use your Clan War analytics page to evaluate recent performances and this aligns player data by taking into account skipped wars and does a great job of additional battles ("count" column & two medal icons as opposed to an easily missed "2" and shared highlighting with single war wins or loses - only 1/2 stands out)

The player profile information we're talking about won't align anyway as it take no account of skipped wars - you have to spot the gaps between dates. If you wanted all players profile data to align, doing Win Rate per season (7 wars) would make far more sense instead IMO.

Regarding labeling, these must be viewable on desktop as well as mobile. When you have very long labels that you suggested, they won't fit on a 320px wide screen. Our site is responsive and aims to be usable for users on mobile or desktop. There will be trade offs.

I appreciate the challenges of responsive design. However, clearly you've worked around displaying information at limited resolutions before. If labels aren't suitable, then tabling the data is an obvious alternative:

I've included Played/Completion % there as an example of another stat, which would play to your priority for clan's to evaluate the player, which you could include with this format (could include collection day stats too).

Statistics are only useful with context and currently you're presenting these win rate stats in a way where they can easy be misconstrued.

[ijm8710]

While I have no problem with the current labeling, Matt did lead me to a request I’ve been meaning to make for some time.

This would also help draw attention to the fact that some war days are 2/2 as they are coloured the same shade of green as 1/1 outcomes and could be overlooked.

Sometimes I’m left tallying my own %. Since we lack summary totals, are you open to the possibility of adding more distinction for boxes that include more than one battle as currently they blur going 2/2 and 1/1 and 0/1 and 0/2.

Perhaps a red outline or bold outline around those specific boxes to reflect multiple battles took place?

[smlbiobot]

1. Skipped wars

This is not possible to determine as there is no way to determine this from the api. The clan war history right now is implemented in a very clever way that reads from data compiled from clans to determine how a player had performed historically. In reality, there is no way to actually fetch historic performance.

2. Table

I can see what I can do and include those columns. Honestly? Our clan family is the one who uses these summary figures and I didn't expect other people to use them. I am fine with displaying battle count for as long as the definition of last 10 stays as battle days.

[ijm8710]

I had voiced ~2 months ago that like mark, there are def other users who would love summary figures.

In ticket #315, had tried to shoot for-

Summary statistics to show ‘lifetime war stats’:

• wars won
• total wars
• missed final battles
• total cards and/or avg. cards collected
• average collections played

[smlbiobot]

@ijm8710 Please stop going into other tickets and create unrelated issues + branches + references. Your comments made these issues very hard to follow. Please don’t hijack other people’s issues to create your own issues.

[ijm8710]

Sml seriously what is up with your demeaning attitude dude. This is the second time ever I’ve ever commented on another thread in your repo. That is how github works, to collaborate. But this is par for the course for you to basically shoot everything down and say how it won’t work, can’t work and you’re too freaking busy. Is my comment not directly related to the conversation itself. You were literally taking about summary statistics which is something I ticketed in the past and chimed in on; is that really hijacking? I’ve been extremely accepting of your critique on twitter in being concise but if you offer a product and want to provide a free environment open to users, it would be much appreciated if you weren’t being, quite frankly, a complete jerk. Sorry for blowing up dude but it’s really a lot. Seriously, how is my last comment unrelated to the topic at hand?

[smlbiobot]

I’m asking you to please stop. Moderators on this Github has advised that we just block you, but I have been relatively nice here. I’ve already commented elsewhere that we have our own priorities when it comes to these issues.

[ijm8710]

I’m not asking for leniency and I am not asking for problems and as I’ve repeatedly stated I’m not forcing for attention to any one ticket. I’m only commenting on a ticket that has my interest. Again, the OP remarked on summary statistics which you elaborated on and I elaborated on further. What was off topic about that?

[MatthaisUK]

I can see what I can do and include those columns. Honestly? Our clan family is the one who uses these summary figures and I didn't expect other people to use them. I am fine with displaying battle count for as long as the definition of last 10 stays as battle days.

That sounds like the best solution to me. Users may not be using them in clan the same way as yourselves (as I say, I'd expect them to use the Clan War Analytics page instead), but I believe many will be checking this section on their own and other player's profiles and potentially getting an inaccurate impression unless they actually check the war results below the statistic as it stands.

Thanks for your time.

[6Sean]

Summarizing the thoughts here:

Target: Player Profile Page

• [ ] Consider adding a table to achieve the following:

TABLE: I can see what I can do and include those columns. Honestly? Our clan family is the one who uses these summary figures and I didn't expect other people to use them. I am fine with displaying battle count for as long as the definition of last 10 stays as battle days.