Conditional formatting of line chart

[kadapavel]

Dear Experts

Is there any way to change colour of part of line chart that exceeds certain level/value.

I would colour red the part of temperature chart that exceeds 40C.

Thanks in advance!

[Greg_Deckler]

Not to my knowledge with the OOTB line chart. Might look into a custom visual or R visualization. Would be a really nifty feature, you should suggest it as an Idea if it is not already out there. I'd vote for that.

One thought, could you potentially break up your value that you are charting such that it is the same metric but filtered to either be below or above 40C? Then you could put both measures into the line chart and color them differently. Would need to see sample data though.

[kadapavel]

Hello 

I tried to split... but it seems I do something wrong as result is 2 identical charts

Measure1 = if(MAX([Value])>40;BLANK();MAX([Value]))

Measure2 = if(MAX([Value])<40;MAX([Value]);BLANK())

[Greg_Deckler · Accepted solution]

OK, I created an Enter Data query like this:

DateTemp

Tuesday, August 22, 2017	30
Wednesday, August 23, 2017	35
Thursday, August 24, 2017	36
Friday, August 25, 2017	37
Saturday, August 26, 2017	38
Sunday, August 27, 2017	39
Monday, August 28, 2017	40
Tuesday, August 29, 2017	41
Wednesday, August 30, 2017	42
Thursday, August 31, 2017	40

I created two measures like this:

Temp1 = IF(SUM(Temperatures[Temp])>40,BLANK(),SUM(Temperatures[Temp]))

Temp2 = IF(SUM(Temperatures[Temp])>40,SUM(Temperatures[Temp]),BLANK())

Put both of these as Values in a line graph with an axis of Date. Not the exact effect I was going for but did display both lines in different colors.

[kadapavel]

It works. Not very elegant solution with continous X axis, categorical for some reason cuts big part of data. Hopefully in next revisiont that will look better.

Thanks for support!

[Greg_Deckler]

Hmm. Really tough to be specific here, I may have to invent some sample data to play with. Perhaps if you filtered your data on import to above and below 40 and put them in separate tables? I'm going to play with this one, should be easy enough to generate some sample data, I assume that your x-axis is date/time based?