How do you get them in the green? Carrying?
What makes up the measurement? Is speed ignored?
What makes up the measurement? Is speed ignored?
Meatdawg
offline
offline
A green rushing bar = speed and agl
Speed increases it more then agl does.
Str does nothing.
Speed increases it more then agl does.
Str does nothing.
Nuge20
offline
offline
Originally posted by Hikariu
What about Carrying?
carrying is very small part, once got my bar to go up from training carrying but mostly speed/agility. Receiving bar=agility as well (and catching duh)
What about Carrying?
carrying is very small part, once got my bar to go up from training carrying but mostly speed/agility. Receiving bar=agility as well (and catching duh)
Octowned
offline
offline
I sent this to squab, based on data he collected. I figured out the blocking bar pretty well. This is a very crude analysis of the rushing bar.
If anybody wants to collect a data set of ~ 100 builds including their bars, just PM me the data set and I'll run an analysis.
----
I got very poor results with "jumping" in the mix. It was by far the largest coefficient. This is likely because everybody has not too much of it, so it got absorbed as a constant at the end. I then took it out, and got..
Strength 0.13
Speed 0.38
Agility 0.3
Vision 0.05
Carrying 0.05
This makes sense, though some of the residuals, especially toward the bottom range, were incredibly large indicating that this is NOT correct.
Adding an intercept, I got
Strength 0.1
Speed 0.33
Agility 0.25
Vision 0.07
Carrying 0.03
+ 6
The residuals were all a bit better. Carrying was actually not significant in this model. If I take carrying out, the analysis gives...
Strength 0.11
Speed 0.33
Agility 0.26
Vision 0.08
+ 6.2
I do not believe there is actually a +6.2 at the end of whatever formula. Rather, this is an approximation of whatever jumping, stamina, confidence, carrying, etc., might carry in this formula. Most people have similar numbers in these proportional to their level, so you can imagine all of that data just getting lumped into, on average, about 6 points worth of the bar.
A few things that you can conclude though, is that strength and vision both factor less than speed and agility. The confidence interval around speed does not overlap the confidence interval around agility, so you can reject that speed = agility at the 95% confidence level.
I think the data set is too small, and that carrying/jumping/stamina/confidence all play a very small part in this formula. However the above was right about 85% of the time, and gives a good approximation to how the bar works. If those mentioned 4 minor attributes were in the formula, their coefficients would be pretty small.
I hope that helps.
If anybody wants to collect a data set of ~ 100 builds including their bars, just PM me the data set and I'll run an analysis.
----
I got very poor results with "jumping" in the mix. It was by far the largest coefficient. This is likely because everybody has not too much of it, so it got absorbed as a constant at the end. I then took it out, and got..
Strength 0.13
Speed 0.38
Agility 0.3
Vision 0.05
Carrying 0.05
This makes sense, though some of the residuals, especially toward the bottom range, were incredibly large indicating that this is NOT correct.
Adding an intercept, I got
Strength 0.1
Speed 0.33
Agility 0.25
Vision 0.07
Carrying 0.03
+ 6
The residuals were all a bit better. Carrying was actually not significant in this model. If I take carrying out, the analysis gives...
Strength 0.11
Speed 0.33
Agility 0.26
Vision 0.08
+ 6.2
I do not believe there is actually a +6.2 at the end of whatever formula. Rather, this is an approximation of whatever jumping, stamina, confidence, carrying, etc., might carry in this formula. Most people have similar numbers in these proportional to their level, so you can imagine all of that data just getting lumped into, on average, about 6 points worth of the bar.
A few things that you can conclude though, is that strength and vision both factor less than speed and agility. The confidence interval around speed does not overlap the confidence interval around agility, so you can reject that speed = agility at the 95% confidence level.
I think the data set is too small, and that carrying/jumping/stamina/confidence all play a very small part in this formula. However the above was right about 85% of the time, and gives a good approximation to how the bar works. If those mentioned 4 minor attributes were in the formula, their coefficients would be pretty small.
I hope that helps.
You are not logged in. Please log in if you want to post a reply.
