Figuring out percentages

I am not totally understanding how percentages are figured. It seems as if I have a run of “no thank yous”(even if most of the ratings are “no thank yous” for everyone) that my percentages go down like a stone.Yet…when I have a good string of like its or love its…they continue to go down…or rebound only marginally. Why is that? Thanks!

4 Likes

Holly, I thought I was just imagining this. Same here. I am fairly certain that the scores are relative to other creative’s scores somehow. So if others are doing better than you are, then your score goes down even if you are getting a string of loves and likes it can mean that so are others. That’s how I have understood the scores and yet I still can’t seem to wrap my head around how it really works. @Grant, I know we ask this all the time but can you explain this to us again so it sticks in our heads? I am experiencing the same thing as Holly.

3 Likes

I wonder if it has to do with WHEN the percentages are updated? For example, say you’re having a string of no thank yous for most of a given day, then the next day you’re doing great but your score takes a dive. Maybe this is because it’s reflecting the previous day’s averages, or the previous weeks…etc.

@Grant when can we expect to see the effects of a positive or negative impact on ratings at any given time?

3 Likes

Percentile Ranking is based on the number of like it and love it rating (high ratings) you’ve received within the last 180 days, relative to the number of entries you submit, in relation to your fellow Creatives. (Wow…that’s a mouth full.)

The first thing to keep in mind is that a string of like it and love it ratings in your Activity Feed, does not necessarily directly correlation to the changes to your Percentile Score.

Here are three of many situations that explain this:

(1) High Ratings Relative to Number of Entries - You’re particularly fast and inspired and submit double your usual entries. You typically submit 50 entries a day and receive 10 high ratings, but for two days, you’ve averaged 150 submissions and receive 30 high ratings. In this situation, while your number of high ratings increased, your percent of high ratings is the same (20%).

(2) High Ratings in Relation to Your Fellow Creatives - There are one or two active contests where CHs are rating using a non-standard curve. They’ve given 75 love it ratings and 150 like it ratings. You receive an influx of high ratings, but so has everyone else. So, again, your number of high ratings increases, but everyone has been graced with this influx, so your high ratings in relation to your fellow creatives does not change.

(3) High Ratings Recently Received - You’ve had a good day with several high ratings. One of your better days, but certainly not your best. Your Percentile Score is based on your last 180 days of participation, so each day your high-rating-over-number-of-entries is calculated and considered; but, each day, an old high-rating-over-number-of-entries is dropped off. Therefore, it is possible to add a good day but lose a better day and see a small drop.

Also note that like it and love it ratings are indeed weighted differently, so this is another factor.

To summarize, your Percentile Score is based on 180 days of participation, number of like its, number of love it, number of total entries, performance of your peers, rating strategy of individual CHs, and more. Therefore, you should not expect to see a day-to-day shift that is related to the number of high ratings you received within the last 24 hours. There’s simply much more too it.

Finally, as I’ve said before, your Percentile Score is a productive measure that we find helpful for the platform’s success. It is not an objective measure of your skill or worth–and we love you all equally :slight_smile:

3 Likes

[quote=“Grant, post:4, topic:2185”]
…and we love you all equally[/quote]
Awww, thanks! :sparkling_heart:

1 Like

Yay! Thank you so much Grant! Can you put this answer somewhere at the TOP of the forum so we can go back and read it? I DID understand most of this to be the case but I did not remember the 180 days thingy and I did know about “relativity” but not really the great way you explained in in #2.

This also helps to explain why there might be drops when you are inactive sometimes, right?

1 Like

Hey Grant, this brought up a thing I wanted to ask if SH could do… is there a way you could “archive” our rating stream (the one under the bell) so we could go back in further in history? Sometimes I wish I could do that…go back as far as it goes… because it “can” sometimes tell you why you were doing so well “back then” but not as well now.

@Commulinks - I’ve added this to the FAQs

And, yes, inactivity can drop your percentile score.

1 Like

@Commulinks - Can you tell me a little more about what you’re looking for.

Thank you for asking, Grant. Under the activity feed (the bell in our account), we can only view activity back so far. I’d love to be able to see the activity feed as far back as I want to go.

The biggest problem with the rating system is that the ‘No, Thank You’ rating can, and usually does, apply to every name except for the winning entry. If one can simply follow the brief, then it should be rated as on the right track, but it usually isn’t. 99% of the time, you can follow the brief and the CH will still rate it as a No, Thank You.

So instead of having a normal distribution curve if we were to plot the ratings on a graph, better known as the bell curve, we have a downhill ski slope. Which would give most economists a head ache.

5 Likes