Knicks411, SeattleCuse, other stats geeks

[moqui]

I have Pomeroy's entire 2011-12 data set including standard deviations and z scores, if you are interested.

For those who don't know stats, z scores are more informative than simple rankings - for example, look at offensive turnover percentage, what I think is one of the bellweather stats when we consider potential SU opponents. Syracuse is excellent at limiting turnovers, just 15.9% of all possessions, good for 5th in the nation. Notre Dame, the only team to beat Syracuse, is also very good - 16%, 7th in the nation.

Purdue is #1, 13.6%. But that number doesn't relay just how good they are at valuing the ball. Purdue's z score is -3.15, more than a full standard deviation better than SU or Notre Dame (-2.08 and -2.03, respectively). Those of us with a background in statistics recognize that as a significantly better level of performance.

So, I have all that data if anyone is interested.

 

[PoppyHart]

I have Pomeroy's entire 2011-12 data set including standard deviations and z scores, if you are interested.

For those who don't know stats, z scores are more informative than simple rankings - for example, look at offensive turnover percentage, what I think is one of the bellweather stats when we consider potential SU opponents. Syracuse is excellent at limiting turnovers, just 15.9% of all possessions, good for 5th in the nation. Notre Dame, the only team to beat Syracuse, is also very good - 16%, 7th in the nation.

Purdue is #1, 13.6%. But that number doesn't relay just how good they are at valuing the ball. Purdue's z score is -3.15, more than a full standard deviation better than SU or Notre Dame (-2.08 and -2.03, respectively). Those of us with a background in statistics recognize that as a significantly better level of performance.

So, I have all that data if anyone is interested.

What is the null hypothesis here? Zero, the national mean, or some other measure?

 

[moqui]

What is the null hypothesis here? Zero, the national mean, or some other measure?

no hypothesis, it is just for comparison purposes. I think z scores give us a better means for making a judgement about a particular team's performance in some measure than do raw averages or simple rankings.

 

[BlackKnight76]

Purdue's very impressive, but -2.08 is nothing to shake a stick at. I've never seen an SU team limit turnovers nearly this well.

 

[OrlandoCuse]

Bump.

I never saw this thread on Monday - do you have a spreadsheet of all of the data?

I'd like to take a look at it, drop me a pm and I'll send you my email address.