Chap 11 Sports Stats

Sabermatrician

someone who uses statistical methods to analyze baseball

residual

difference between actual value of a variable predicted value of a variable

SSR

One way to measure how well a model makes predictions

least squares regression line

mom relationship between an explanatory variable and a response variable

interpret the slope

for each (explanatory), there is a (slope) (increase/decrease) in predicted (response)

interpret the Y intercept

when (explanatory) is zero, the predicted (response) is (y intercept).

extrapolation

when we predict the value of the response for a value that is much larger or smaller than the other data set values for the explanatory variable

True slope

slope we would get if each team was able to replay the season millions of times and we calculated the creation of the least squares regression line

observed slope

The slope we get by calculating of LSRL for a performance or set of performances

regression to the mean

when measuring the same variable in two different time periods, the tendency for better performances to follow poor performances and for worse performances to follow good performances

Standard deviation of the residuals

typical distance between the actual values of the response variable and their corresponding predicted values

coefficient of determination (R squared)

used to measure how well a model summarizes the relationship between two variables

interpret the standard deviation

when you are using (explanatory) to find (response), our predicted values will typically be about (SD) ( label response) away from the actual values.

interpret the P value

assuming that null, there is a (P value) percent chance of getting an observed slope of (B or value of the slope) or higher by random chance.

interpret the residual

The residual indicates that the ( subject)'s (response variable) was (residual) than what we predicted based on (exaplanatory).

null hypothesis

The true slope between (explanatory) and (response) is zero.

alternative hypothesis

The true slope between (explanatory) and (response) is positive/negative.