LA consistency, SDLA, Swinggraphs

[Westwind]

Can someone put this chart into layman’s terms for me? Especially the line about 2 standard deviations?

I tried to talk my math major wife into sitting down to explain it to me but she rolled her eyes and told me she could explain it but I still wouldn’t understand....

(Well maybe she wasn’t that harsh, lol)

 

[jdint]

I'll give it a shot, but college was many moons ago. So this study/survey found the mean (average) of the Launch Angle used for a data set. I'm assuming sample of major or minor league baseball players or something similar.

A standard deviation would measure the dispersion of data relative to the mean. The more spread out the data is, the higher the standard deviation.

Since they are talking about 2 standard deviations you can think of it as a bell curve and use the 68-95-99.7 rule, which you can think of as 1 standard deviation will contain 68 percent of the data points, and 2 deviations will contain 95 percent of the data points.

So, I think what they are saying is, as long as you stay within 2 standard deviations from the mean launch angle, there is no benefit to trying to get a launch angle closer to the mean. So basically eliminate the extreme highs and low launch angles and results will be similar.

Now someone better at math and stats. can explain why everything I just typed is wrong.

Edit: Seems like this would be Pattar's area of Expertise.

 

[_MIA_]

@jdint got it. In terms of how to read the graph to come to the conclusion they made, in the 2nd graph the (regressed) line is more or less flat (in particular between 15 and 25 degrees) which means that xWOBAcon doesn't really change as long as your LA is within those two LA standard values.

To be honest not a fan of how those two graphs, which I presume represent the same data but over different abscissa values, were presented. In the first graph for some reason they don't even show the complete range standard deviations which are represented in the second graph and the title of the graphs should be xWOBAcon vs Std. Dev Launch Angle not the other way around.

 

[Golfclubgreg]

I knew I shouldn't have slept through all those classes haha

 

[Rolling Hard]

OK, I need it real simple. How would I use this info to help my 13 Y/O DD?

 

[_MIA_]

OK, I need it real simple. How would I use this info to help my 13 Y/O DD?

You wouldn't...

 

[jdint]

@jdint got it. In terms of how to read the graph to come to the conclusion they made, in the 2nd graph the (regressed) line is more or less flat (in particular between 15 and 25 degrees) which means that xWOBAcon doesn't really change as long as your LA is within those two LA standard values.

To be honest not a fan of how those two graphs, which I presume represent the same data but over different abscissa values, were presented. In the first graph for some reason they don't even show the complete range standard deviations which are represented in the second graph and the title of the graphs should be xWOBAcon vs Std. Dev Launch Angle not the other way around.

This account suffers from this repeatedly. They make some pretty interesting assertations, but it's generally not presented very clearly. I figure some of this is do to it being on Twitter, and I believe they have a pay site now where they probably do a better job of stating their case.

And you know who stops by now and then to tell them they are wrong which is always delightful.

 

[Rolling Hard]

You wouldn't...

Well, thank a lot. I wanted something to work on today.

 

[_MIA_]

Well, thank a lot. I wanted something to work on today.

Post a swing..I am sure you will get plenty of suggestions

 

[Bonesaw]

"Hit the ball over the outfielders' heads". That is how I read those graphs.