think of it as a "linear" pie chart. A full pie is 100%, so the graph shows a breakdown of the total out of 100% for each point in time.
Using a linear (1 dimensional) graph is easier than using multiple pies which requires 2 dimensions and a lot of space :P
Comment has been collapsed.
Your detailing of the issue could be presented a bit more clearly, it's a bit confusing as it is now.
That said, once I figured out what you meant, you do seem to be correct (about there being an issue, not seemingly in regards to the nature of the issue), at least by consideration of how the graphs are displaying on my end- if we consider the graphs as being in the standard format that they at first glance appear intended to be interpreted as, then all the graphs can only be assumed to be displaying in an absolutely bonkers manner at the moment.
Hovering over "Per Type", for example, states, for example,
49% Public
3% Invite
39.4% Group
8.6% Wishlist
However, looking at the chart itself, instead gives the following:
100% Public
50-something% Invite
40-something% Group
Seemingly an accurate 8.6% Wishlist
(Proper calculations intentionally not made.)
Neither by absolute (ie, totalling to 100%) nor relative (ie, the highest individual representing a value of 100%, with others relating their own percentages to that number, as per 8oz = 100%, 4oz = 50%) considerations does any of the visualization aspects of the graphic seem to make any sense at present (again, that is if we assume the charts to be standard format).
Gift Feedback's the absolute weirdest, as it's currently claiming a 100% rate of Not Received in its visualization. :'P
We can can thus conclude that the charts must actually be stacked charts, despite any lack of indication towards such in either the descriptions or visualizations. Thus, the issue ends up being that the charts just aren't presented or described capably relative to their intended purpose, and not that there's anything glitching in how their calculations are being presented.
Comment has been collapsed.
I concede the miscategorization of my post, more importantly I thank you for seeing the thing that I saw and turning into words that word good. Sometimes my brain is made of bad meat. Also, dang, you have been a zealous advocate, I'm gonna close the thread cause you done said it all, but Im the sort that attracts contradiction, so otherwise the debate will not stop.
Comment has been collapsed.
As Beauregarde noted, that's clearly not what it's supposed to be representing, based on the color matching and the labeling. It's also batshit irrational if viewed as a standard chart, since it arbitrarily picks which ones to combine, making the visualization functionally useless for standard comparisons.
Presumably, of course, these are intended to be stacked charts. Stacked charts work by filling in the sections so that you can compare the depth of the sections. More info / example. That's not something the current charts are doing. Since they're intended to provide a comparison of the various numbers relative to themselves, they're also not typically calculated against percentages, though in this case we're contrasting percentages to begin with, so it's clear why that formatting is being applied.
In addition to that atypical presentation and the lack of necessary color filling, nothing in the descriptions suggest it's supposed to be read as a stacked chart, either.
So, considering it as it's presented, Beauregarde's interpretation is entirely reasonable.
Comment has been collapsed.
Maybe it's been fixed by now but I genuinely don't see any problems with any of the graphs. It all looks right to me.
Comment has been collapsed.
The issue is one of poor presentation, so it's a bit hard to tell if there's a bug or not without clarification on whether there's a discrepency in how we're viewing the graphs. The graphs are supposed to be colored in, but aren't, and they don't give any indication that they're supposed to be read as though they were colored in. Assumping that's not a weird per-browser quirk or similar, it'd be a design issue instead of a bug. Else, well, it's something for cg to fix.
Comment has been collapsed.
They are properly displaying as such. There are colored areas that don't overlap and that give 100% in total.
Maybe they can be made less transparent, but cg probably was lazy and made transparency same for non-stacked area chart and stacked area charts the same.
Comment has been collapsed.
24 Comments - Last post 7 minutes ago by WaxWorm
19 Comments - Last post 2 hours ago by DanteOP
161 Comments - Last post 2 hours ago by wigglenose
13 Comments - Last post 4 hours ago by VicViperV
1,961 Comments - Last post 9 hours ago by Gamy7
1,042 Comments - Last post 10 hours ago by sensualshakti
769 Comments - Last post 10 hours ago by OwieczkaDollyv21
22 Comments - Last post 3 minutes ago by LittleBibo1
56 Comments - Last post 18 minutes ago by blueflame32
1,052 Comments - Last post 24 minutes ago by Bubles
452 Comments - Last post 30 minutes ago by ucho
6,344 Comments - Last post 33 minutes ago by Tigerci
797 Comments - Last post 39 minutes ago by DarthLonginus
175 Comments - Last post 49 minutes ago by wigglenose
https://www.steamgifts.com/stats/community/giveaways
How come the "per type", "by CV", and "Gift feedback" line graphs have each replaced 1 set of data points with straight 100%'s? I mean, if there were a superfluous line I wouldn't comment, but theres actual data being overwritten and omitted on a medium that purports to provide all the data it describes in an "at a glance" visual perusal. But Im sometimes less high-functioning than I might think, so perhaps I just sound wackadoo, in which case, my apologies
Comment has been collapsed.