A study on ping: Part 1 (Ping vs Wins)

[Ninetailefox92]

I like what this thread attempts to do, since math and facts are nice to have.
However, I feel that the study and the statistics involved is not sufficient to make any valid conclusions.

• The sample size here is much too small for the community's general size. By my estimates, we would need at sample size of at least 300 individual cases in order to be representative of a community our size. Plus the samples being taken are not random, thus they may not be representative of the entire community.
• I do not believe that the conclusion, "the higher ping you have, the more likely you are to win". Not only is Ping a very dynamic data point, but the conclusion itself has the issue on its extremes: having low ping means you can't win at all, and having high ping makes you the best player ever. The logic does not make sense when practicality is factored in.
• The case ignores a very important statistical analysis data point, the Pearson product-momentum correlation coefficient (Or as I learned it, the r Value). The r value checks to see how far the individual data points are from the average, averages out all of the displacements, and then comes up with a value between 0 and 1. That value determines how accurate the average is, and thus how representative the average is to the data. Based on the graph shown, I do not think that the linear regression of the graph is very important, considering how far displaced the data points are. This would further explain the previous point.

Again, I like using facts and research to find information like this. However, statistical surveys require a very particular way of proceeding in order for their results to be valid. You'll understand why such little things matter once you take a college level Statistics course.

Col! You pretty much resaid what I already said! stealing my points!

 

[jtmboy]

Ping:40
Wins:11
I know, I am so great at sg

 

[IronOre12]

Lots of math and stuff.

 

[Ex0rzioN99]

IGN : chickenboy123
Ping : 80:2 = 40ms

 

[Kalzo]

(If you don't like math don't read this)
Hello everyone!
So if you saw my other thread, I was collecting data to run a study on how ping effects stats, etc.
I understand that atm the data-set is quite small but here are the current xy values (Ping vs Wins)
75,11
75,802
41,3237
13,541
40,802
96,454
70,941
100,11
150,1536
105,371
21,91
64,630
85,325
68,66
78,94
38,56
73,1754
220,1530
25,6
85,1474
54,58
75,618
110,2127
75,148
150,257
25,330
9,20

Plotting these points on a scatter plot gives some interesting results
Ping on the x
Wins on the y

What you can see is that as your ping increases (generally) your wins increase (generally).
The opposite of what most people would expect.
Some interesting things
-No one in the data-set with less than 40ms has more than 800 wins
-It is perfectly possible to have 1500 wins with 150ms

We can use the formula y=331.47+4.62x to calculate y off of x or vise versa
Someone with 50ms would have (on average) 562 wins
Someone with 100ms however would have (on average) 793 wins
Now don't go say "LOL SO IF SOMEONE HAS 250MS THEY AUTOMATICLY HAVE 1500 WINS!?!?!11" Remember that these are averages.

Keep in mind not everyone will correspond exactly to these values and many even on the data table are far off (again, these are averages)

The slope is 4.62, meaning for every ms up you will have another 4.62 wins

Conclusion: Ping does not directly effect wins (So far with this limited data)

REMEMBER THAT THIS IS WINS, NOT SKILL OR KD OR ANYTHING.
THAT IS COMING IN PART TWO.

IF YOU WOULD LIKE TO CONTRIBUTE COMMENT DOWN YOUR
-IGN
-PING

-Tene

I HAVE 60 PING THATS WHY IM ASS LOL

 

[Budderunner]

ign: xNeh
ping: 150ms

 

[Giggums]

Thread locked due to necro.