Suppose you're using rise.global's Twitter Followers/Following data collector to track your users' Twitter Followers.  And you've auto-scheduled your score collector to poll every week.  In this case, your scorebook will record the number of an individual's Twitter followers every week (at a set time).  By default, rise.global saves all data entries for a period of 90 days.

So, besides giving points for someone's # of Twitter Followers each week, you may also want to give points for the % change in their # of Twitter Followers from the previous week, so that your scorebook looks like this:



To create this, you need to create a metric in the score algorithm that calculates the % change.

You do this on rise.global by using the Rise Scorebook Data Collector and formulas to create a new metric  that automatically records the difference in the score value (which depends on the score method chosen) of a metric between the two latest score periods.

Let's see how we do this.  I've set up a scorebook using rise.global's Twitter Followers/Following data collector.



With this collector, you get two metrics - Following Count and Followers Count


We now want to have a score metric that will record the difference in a player's followers count between one week's count (or whatever score period you are using) and the previous week's.

For this, create a new Rise Scorebook collector:


using the "Use raw data from another scorebook [pulled in when releasing]" option and choose the following configuration options:
For scorebook, choose your current scorebook
For the metric, choose the metric for which you want to create a time-series difference value
And in Name, choose a name for the metric that will describe it meaningfully


Next, navigate to Scorebook Settings -> Score Algorithm, select the new metric that has been just created and click on edit metric, so that you can configure it to calculate the difference in values between this week and previous week's data entries


Choose "Latest - Difference" as the score method 


Now, we can create a formula (which will also be a new metric in the score algorithm) to calculate the % change in the difference value


For this scorebook, we don't need need the "Following Count" metric, so we can delete it, and then we can add a new formula metric. But, if you will have more than 3 metrics in your score algorithm, you may need to subscribe to a paid subscription plan.


Next, in the score algorithm page, make the following changes:



Finally, we need to decide how the "Total Score" will be calculated.  In this example, we want to create a score out of the Twitter Followers metric and the % change Twitter Followers WTW metric.  Since the values in these two metrics will have a very different scale (one will be numbers in hundreds, thousands, millions and the other one will be a number between 0 and 1), it doesn't make sense to just add the two values to create the "Total Score" - even if you give the % change metric a much larger weight than the Twitter Followers metric.  So, in this case, we can use rise.global's relative ranking method.

I also gave equal weight to the two metrics (by giving them both a weight of 1 so that each metric will contribute 50% towards the Total Score) and gave the Followers Count WTW change metric a weight of zero.





You are now all set up. After the data collectors have been polled over two consecutive score periods (e.g. once a week over 2 weeks), and you then generate a bulletin, you will get a leaderboard looking like this: