The new tag stays until a quiz has been rated enough times to be ranked in the weekly rankings. I believe this is 20 ratings. I just checked a few older quizzes still marked "new" and they were all sitting on 19 ratings.

I went and looked and although they have been played a lot, they have not been rated enough yet.

Thank you. I think I may go try a few

The exact criteria for a quiz being "New" are:

- EITHER it has 19 or fewer ratings
- OR it has not yet been through four Monday ranking calculations.

Or to put it the other way round: A quiz will be ranked if it has 20 or more ratings AND has been through a Monday re-rank at least three times before.

The Monday re-rank happens in the morning hours (I think it's around 7am FT time), so assuming it does get the needed ratings, a quiz that goes online on a Monday before that time will be ranked three weeks later, one that goes online on a Monday but after that time will be ranked four weeks later and one that goes online any other day will be ranked on the fourth Monday of its existence.

I did and got double points!

How did you get double points?

I played the quiz and I got a message afterwards along the lines of; As this quiz has been online for some time and has not been played often, we have doubled your points

I think I have a similar problem.

Quiz: Gotta catch 'em all
Category: Pokemon
Rank:114,625
Ratings: 36
Good & Excellent: 19 (53%)
Excellent: 5 (14%),
Good: 14 (39%)

The ratings ain't no bad.

I'm not sure what problem you see. Your quiz has a ranking of 114,625. It has been ranked based on a comparison of the ratings players gave it with the way other quizzes have been rated. It is not shown as New, and will not award the extra points for a quiz that has been online a long time without getting enough ratings to get ranked.

It is almost impossible to guess what a quiz's ranking will be from the feedback you get about Good & Excellent ratings, but none of my quizzes has a Good & Excellent total under 58%; that lowest ranked quiz is at 85,075. (The fact that you don't know how the ratings that were not Good or Excellent are distributed among the three lower ratings means that you cannot make direct comparisons on the basis of the G&E ratings, but they do give you a starting point.)

I noted that quiz to be ranked 114,xxx out of 110,xxx. How can a quiz be ranked lower than the number of quizzes? I know expect that second number is only periodically updated (and any calculations based on the actual total) but I'd think prudence would call for publishing a number that is a few thousand greater than the number expected to be reached by the next update, not lesser. Unless for some reason the cycle's expectations have been exceeded. Are there really 4,000 more quizzes rated than had been planned for in this cycle? Maybe the cycle frequency needs to be shortened?

Edit for the strike through for logical reasons.

My quizzes are right around the same as Looney_Tunes'. I do have one down to 52%, and it's rated about the same as yours, Shivam.

It's also difficult to tell sometimes, as Looney said, based on the percent of goods and excellents.

I have one that is at 60% that is in the low 30,000s. I also have one that is 88% and in the high 50,000s. Very odd; for the most part my percents and ratings go down equally, with some oddballs thrown in there like that.

There are 117,404 quizzes on this site.

You can get the exact number of quizzes by checking your progress on the 5/10/15 Categories badge - it will say [quiz name] - xxx,xxx/yyy,yyy - the number it's out of is the number of total quizzes.

I don't think there are "cycles" of updates. I seem to recall in the past when this was pointed out by a member, Terry posting along the lines of "yeah, that hasn't been updated in a while; I should go do that". I think it just gets updated whenever it gets out-of-date by a meaningful amount.

It would be really hard to predict how many quizzes will go online in any given time period, just because it depends on so many factors: how many quizzes are submitted by users, how long each one takes to edit, editor availability.

Is there really much difference between #110,000 out of 114,000 or #110,000 out of #117,404? I'm not sure it's worth the time to statistically analyze quiz trends to get a more accurate denominator.

Here's another way it could be presented: use the actual number of total quizzes in that statement space. Instead of having a prestated number, link in the actual number and it'll never have to be adjusted in the future by hands on re-writing. Where it says 110,xxx, put in a link to the location called out by Adam.

It would be too hard on the system to constantly revise 117,000 pages with new numbers every time a quiz went online or offline. That's why a denominator is chosen to display instead.

I really think this is making a mountain out of a molehill. It's not strictly necessary to know every quiz's exact ranking out of an exact number of quizzes online. That feature is displayed on the quiz page for a general idea of the quiz's popularity.

I intend no argument. I only suggested ways to improve a number many look to. Now we know there are 7,000 quizzes whose authors look at their rank and see their number is outside the realm of possibilities at FT (the moles from the molehill? ). I think the numbers get recalculated every Monday from the top down. A Monday restatement of the total quizzes would be of more help than being 7,000 out of place. Quiz numbers change as new ones come on line, rankings do not change with quiz additions until later when they are ranked.
And I believe that 7,000 consist of many who have long left FT before standards were raised, sorry paa2, to bolster your point LadyC (not a tactic used by someone out to argue, but by a debater). But still, if a real number can be closely approximated and displayed with no major impact to the system, why not change it? It would only be a once a week event. As new quizzes wait their 4 weeks for ranking and then get it added on Monday, that denominator should be adjusted along with them.

I tried looking up the category of paa2's quiz using the rocking horse Childrens icon but didn't find the Pokemon Category. So I went back through paa2's profile for the quiz and find that the quiz's play results graph is definitely drawn from the imaginary numbers of being outside the realm of other quizzes. The y axis is labeled players and the x axis is labeled score out of 10 with the highest being 10.0000! See:

http://www.funtrivia.com/trivia-quiz/ForChildren/Gotta-catch-em-all-352809.html

I can understand this author's wonderment that something is "bad".

I figure it's a case of simply approximating the amount we have without going into the details. No one wants to know that their quiz is exactly 140,761 out of 140,761. Besides, approximate numbers for quizzes are used in several places around the site (including the homepage. Terry updates them irregularly, but it still gets done when need-be (similar to when the amount of sunnies gets expanded).



I'm not sure whether or not this is necessary-- that page is really only relevant if you have your own quizzes waiting to get ranked for the 5/10/15 Categories badge. Linking it for everyone, authors and non-authors, may only complicate things a bit.

That graph certainly has some inconsistent use of decimals on its horizontal axis, but it shows scores between 0 and 10, with a number of players on each score that looks at a glance to correspond to the 38 plays the game has received so far. I don't see how it reflects "imaginary numbers of being outside the realm of other quizzes".

I think the ranking algorithms needs to be checked.
How can the rank be-114625 of 111000!!
This makes no sense!

Read the earlier posts in this thread. The number 111000 is not accurate, as Terry only updates that figure at intervals. There are actually over 117,000 quizzes at the moment - see post #972608 for information on how to see the current number.

If this is really going to be considered a 'problem', perhaps the easiest thing to do is remove the "of 111000".

Then it'll just read "Rank: 114625". Sorted.

I've taken three quizzes over the past day (two older ones and a new one) and they all have that x-axis decimal issue. This was solved about a year ago and now seems to have returned. Did something get undone, a reset to former definitions take place or is it a new problem causing the decimals to appear where whole units are required?

I think a rough estimate is good. Someone might think a quiz rated 5,000 was terrible unless they knew that there were over 110,000 quizzes on the site.

OOOOOOOOOOOOOOOOOOOO

All of my quizzes are new now!

Today the system ranks quizzes early in the morning. They'll go back to normal.

You are quite right