I sent my AC/DC pinball statistics spreadsheet off to one of the academics I work with. I mentioned the little skills I had in creating cool stuff with the statistics I had and she sent me back this...
"I've had a look at your pinball data. You must have mad skillz haha. Pivot tables are for any kind of data summary especially across multiple variables and they make it easy to apply filters etc. as well. So your main sheet where you've got average score and the % of games that you got the different multiballs these can all be done in pivot tables too.
In the red tab I had a look at comparing the average score for games where you did or did not get each of the types of multiball. Seems like scores are about 100 million higher on average whenever you get at least one multiball. I'm sure this isn't news to you though!
If you click in the pivot table it should come up with the Fields option window to the right and you can see the choices I made for the 'values' and 'columns'.
in the blue tab I had a look at whether game order on the day was related to winning Awards. The overall award rate was 10% (I had to create a new variable in the base dataset that simply flags whether or not there was a comment in your original Award column.) Looks like most of the time you play 3 or 4 games on the day, and for some reason the really big outlier scores tended to happen more often on game #1 or game #3.
The scatterplots on that page show the scores over time with the red markers showing games associated with Awards. It looks like the blue markers (normal games) have a steady amount of variance but then there are these outlier games where you get massive scores and Awards. And it does seem as though these outlier games have become more likely over time since the start of the dataset.
In the tab for Daily average score you can see the pattern as well.
And lastly I had a look at whether you were getting the multiballs more frequently over time. Extra ball seems to be the rarest and worth the most points and the proportion of games per month that you're getting the extra ball has increased over the three months. There's a bit more daily variation though. But you can look at this type of thing for the other types of multiballs too.
If you keep recording the data it should be easier to see if any patterns are emerging."
This image is average score by balls
this one is "Characteristics of a Winning Game"
And this one is called "balls frequency over time"
Im still not sure what they all do/mean but its good to know I have a statistician I can call on to help me with my data 🙂