a=Seasonal Ladder rating b=1v1 ladder rating c=real-time ladder rating d =(number of coins*10) e=2v2 ladder rating f=3v3 ladder rating j=boot rate k=1v1 winning % --- calculated directly by =((number of 1v1 wins)/(total number of 1v1s) l=player level g=total h=difference between days, calculated by (g(x)-g(x-1)) To calculate change in skill per day,
i=(h(x)/g(x-1) for daily difference in skill
Simple enough. Let me know if improvements could be made.
Edited 11/19/2015 18:51:50
How to calculate 'growth' in skill: 11/19/2015 19:01:34
If you want to be extra-meticulous about it, you could change k to =((1v1%+2v2%+3v3%)/3), and then average the combined seasonal ladders that you've played in. The equation's pretty easy to toy around with.
How to calculate 'growth' in skill: 11/19/2015 19:30:44
The point instead, is to see growth over time. The main idea is to use the change per day to track how you're changing as a player; thus, a higher daily percentage in i is much more useful than a high "score" in g
How to calculate 'growth' in skill: 11/20/2015 02:43:43
Nah. Since I'm not currently rated on Seasonal, 2v2, or 3v3 ladder, my only score comes from what meager coins I have + 1v1 + RT ladder. My current score is in the 6,000s; what matters more, however, is that I've grown at an average of %12.8 per day for the past three days -- which is fairly substantial, but doesn't mean much at the moment because I'm only just getting this account ranked on the ladders. Again, this is to measure growth, not total.
Edited 11/20/2015 03:10:42
How to calculate 'growth' in skill: 11/20/2015 05:28:25
Instead of aggregating all those disparate stats into one pretty arbitrary 'score', and measuring the change in that, why don't you just leave all the variables separate and measure their changes/improvements separately? That way at least the numbers you get will have some sense to them, and they can be easily interpreted.
One other thing you might want to consider is using averaging and a time lag to 'smooth' the measurement of improvements. E.g., instead of taking differences between daily stats, you could take an average of your stats over the course of seven days, and take differences between these weekly averages rather than differences between the raw scores. E.g., sum up the 1v1 ladder ratings for week 2 (days 8 - 14), divide by 7 for the avg 1v1 ladder rating for week 2. Do the same for days 1-7 for avg rating for week 1, and then subtract avg.week1 from avg.week2 to get your week-to-week improvement in your rating. Exactly the same process for all the other stats.
It's pretty easy to do if you do it up in a spreadsheet. Then you can do all sorts of graphs to visualize it to your heart's content.
Posts 1 - 28 of 28
Post a reply to this thread
Before posting, please proofread to ensure your post uses proper grammar and is free of spelling mistakes or typos.