^apparently not!

Trends using the noaa temperature data link here should make sure they are pulling data at about the same time if they are going to compare. Though, its still possible to grab data from the same source with enough of a time gap between each data grab that the information pulled is incorrect. Would be handy to be able to update multiple trends from one page grab.

Oh, and all the old data was deleted so we can have a fresh start. :-)

Updated the trends to collect data at the same time every day, at either 5 or 6am EST depending on wether DST is active... no way to indicate time of collection other than in UTC... Either trendy needs to be fixed for a more consistent time capability, or all the worlds governments need to hurry up and come to their senses so they can end daylight savings time practices.

you'll never walk alone

I do not manage to find the best location for my legend : either the legend is truncated , or subplot 212 becomes too small

Shows plot as broken:
Error using minmax
Too many output arguments.

However I do not use minmax to create this plot and the plot regeneration works fine. I'm confused...

fail ..I have no idea what UTC time is ..

... and the recent rescoring of solutions to eliminate the feval trick has provided a small drop in the score for some of use :-)

14-Dec-2012 I am Alone in the interval [7000-8000]. Both Richard and @bmtran seem unreacheable!

Today I posted link at vk.com group (1100 subscribers).

Badges, introduced in mid-November, provided a big one-time jump in the scores.

29-October-2012 : new player above 7000 pts!

August 27th, 2012 :
This Trendy was again discussed on the blog "MATLAB Spoken Here" http://blogs.mathworks.com/community/2012/08/27/learn-about-matlab-plots-using-trendy/

In 4 months (121 days), it happened only 3 times that the temperature was higher in Paris ...

ah surprise ! you have a new challenger ;)

Right. That's what WAS happening, but it shouldn't have that problem anymore.

Ah ok , you mean that if one Cody player is not in your playerTable , the Trendy will fail.

Thanks for the reminder...

It wasn't the NaN. My code looks up each players name based on their profile ID, and my name table didn't yet include Khaled, because he has made such a meteoric rise. Should be a little more robust now.

Your plit is broken , maybe it comes (like me) from the bug of 14 Sep 2012 06:07:17 . Trendy has returned NaN .

I am fair play : I did not delete the first data which show that there was some rain at home and not in Paris !!
Anyway it is so rare that this scenario occurs ...

wow I finally managed to show correctly the evolution of score .
Until this fix , legend was wrong and all data were linear.
Now we can see that when a Cody player has scored above 2000 points , the red line decreases whereas the black line increases ...

If we look at the mean temperature, we can use the formula "Six degrees of separation" !!

I find it interesting that the level of US debt per capita is around the same as the approximately $51k median household income in the USA... so that $51k median household income gets divided amongst the household, but the debt burden is for EACH individual in the household. For some reason, economists think the median is more useful than the mean... so Im not finding that number right now, but it would be more useful for this discussion. So instead, lets consider GDP per capita.

In terms of GDP (sorry, only found 2011 data), the US has a GDP per capita of ~$48k. ~$39k for the UK, and ~$50k in Canada. So that begs some questions both about affordability of our debt here in the USA, vs that of our neighbors to the north and across the pond. From eyeballing the graph, thats a GDP per capita of ~ 4x the debt per capita in the UK. ~2.7x in Canada. ~1x in the USA. Of course thats comparing 2011 GDP to the beginning of the year.

So, it also just occured to me, that yes the data is normalized... but it was still collected in the respective currencies of the countries in question. And each trend is only normalized to its first data point... In fact, might be more interesting to see the actual numbers, translated to a common monetary unit.

Will have to collect the data with respect to a common unit, based upon each day's exchange rate. Also, it would be interesting to see another with just the current exchange rate. With roughly $1.6 USD per british lb, this picture, visually, should look a little different than it does.

With the granularity given by the US data, any idea what the jumps/drops can be attributed to?

If anything significant, would be cool to drop the annotations on the plot too. In particular, roundabouts where it starts back in Feb, again in May, the up and down again in Jun, then the upward shift in July.

Also, since everyone is focused on cutting spending, maybe some of the major inflection points for spending rates... If you dont get to it anytime soon, maybe i'll do some modified versions of the plot after my move. :-)

Romney has taken a dive through July and August...

Fixed, error due to website change.

If Cody was on the Olympic games , I would have had the bronze medal !

@29 Jul 2012 06:06:46 , all the scores are reset for the new season 2012-2013

3.21 yeap let's go to upgrade!

cool! It looks like very nice now with the legend.

Thanks for pointing out the problem Aurelien! I meant to cycle through the current leaders, but instead I was cycling through the top ten when I started gathering data. Should be fixed now.

Ned,
Is is correctly updated ? I mean I see that for example the trend of Mark is not updated anymore (he is currently 11th) , we should see the other Cody player James instead.

I so look forward to seeing how this trends. Thanks Bob!

very nice plot!

Bien fait! See http://www.mathworks.com/matlabcentral/trendy/plots/917

12-Jul-2012: same player (Richard b. ) has given 7 different correct solutions which explains the jump of the blue line

Yes I am in the big four !

See http://xkcd.com/1071/

This Trendy was also discussed on the blog "MATLAB Spoken Here"
Multipoint data gathering in Trendy:
http://blogs.mathworks.com/community/2012/06/18/multipoint-data-gathering-in-trendy/

Updated 6/15/2012, Dropped Rick Santorum from the plot and fixed trends for Obama and Romney which were getting spurrious data due to politifact.com changing their display.

John, in my case +2=absolute Knight and -3=absolute Knave. Above zero would indicate that they're more truthful, below would indicate that they're less truthful. Obviously I'm not suggesting that you base any real decisions on this graph. The truthiness numbers that are assigned are compiled from ratings on Politifact.com, who analyze statements made by candidates on a regular basis. This program does no analysis of the statements, rather it grabs the analysis that has already been done on the statements and compiles it.
Ultimately, I've found that whether you believe this graph or not depends on your political leanings. My conservative friends wave it off saying anything from it's 'bad analysis' to it's a 'vast left-wing conspiracy'.

Fanboys always vote first!

This is nice; shows a great visual correlation of how the opinion of the movie trends against the number of viewers.

Of course, anyone who ranked the movie below a 9 is wrong anyway... ;)

This plot has been quoted on the blog "MATLAB Central Release Notes – May 2012" : http://blogs.mathworks.com/community/2012/06/04/matlab-central-release-notes-may-2012/

@Jan - Explained what an activity is!

I'm assuming 1=Knight, -1=Knave, correct? So basically, you can't trust anyone. Probably can't even trust this graph, because of how the compiler numeralizes(assigns a truthness value) a statement or action by the person and how closely they are followed.

I finally managed to solve this problem!!! Good luck for the other players!

Thanks for the note. Some NaNs in the data caused the colors to get bleached out of the plot. Should be fixed now. Yes, there is plenty of room to make Answers more efficient. On the plus side, more than 70% of questions are getting answered.

I see 3 colors in the legend, but only grey edges in the diagram. I assume the Y-axis contains percent values.
I think the diagram means, that on 22-May-2012 15% of the questions asked on 15-May-2012 got an accepted answer. If so, there is a large potential to make the Answers forum more efficient.