MSC-related statistics project (need help)

[Shawn]

I'm gathering raw data from MSC for a statistics project but I've run into a little issue. As we all know, MSC is tough on older computers and my computer is on the verge of being useless. I cannot play MSC at all and would appreciate if others would assist me with this matter. I need the accuracy numbers (the x from the (x/number you need to hit) thing)and damage for hitting giant rats (I'd assume they have no damage reduction and would provide the base damage) with a Nova Blade (I assume most everyone has one?). Because there will be multiple data sets, I also need to know the swordsmanship power as well as the number you need to hit for each data set.

An example of someone's data might look like:

Power: 32
To hit: 21
Acc. , Dmg.
78 , 132.7
20 ,
24 , 89.1
etc...

From this data, I hope to determine
-the minimum and maximum damage for the nova blade relative to the power
-the distribution of the accuracy (if all numbers are equally likely or if the RNG does tend more towards lower numbers)
-the distribution of the damage (similar to above, focused more on if damage is normally distributed (middle is more likely))
-the correlation, if any, between accuracy and damage

Thanks in advance to anyone who helps me with this.

 

[Thothie]

*shrug* Dunno what you want it for, but it won't do you any good, as most weapons work via random adjustment before they are ratio'd by your skill's power level. You'd need a database of a 100+ attacks to get accurate estimates for a lot of them.

I'd put up the base stats for all the weaponry but MiB asked me not to. :\

But since mentioned it, I'll give you the Novablade's stats:
dmg: ( 200 +/- up to 140 ) * ( swordsmanship.power / 100 ) (req: swordsmanship.proficiency 15)
miss-chance: 25% - ( 25 * ( swordsmanship.balance / 100 ) )
second charge: x2 damage (req: fire 10 - damage calc switches to fire skill for this charge and those thereafter)
third charge: x3 damage + skull (req: fire 20)
fourth charge: x4 damage + volcano (req: fire 30)

 

[Shawn]

Thanks for the info. I realize that most of the data would probably have a flat probability distribution and would provide no interesting results but my statistics teacher doesn't care. Though it would be interesting to see if there is a defect in the RNG because axes seem to miss more than they should (though that might just be a psychological effect due to the frustration of having several consecutive misses).

 

[Tentadrilus]

Are you going to work out the standard deviation and the Spearman's Rank correlation coefficient too?

 

[Thothie]

Thanks for the info. I realize that most of the data would probably have a flat probability distribution and would provide no interesting results but my statistics teacher doesn't care. Though it would be interesting to see if there is a defect in the RNG because axes seem to miss more than they should (though that might just be a psychological effect due to the frustration of having several consecutive misses).

It uses the system's RNG function - the game doesn't have one of its own, so ye'd have to chalk it up to emotional perception. If there are any flaws with standard PC's RNG's, I'm sure you can find many comprehensive studies on the subject (IIRC, there used to be all sorts of articles theorizing as why people come up with winning lottery numbers so much more often than computer RNG's do.)