BigProject Final Presentation

[DerekChibbaro]

No description provided.

[ernbilen]

Super cool project. In terms of improvements:

• I'd try to keep numbers as minimal as possible, but highlight 1-2 key statistics. Like, "it used to be that X percentage of balls went to left, now it's Y percentage" etc. Having too many numbers make it difficult for the audience to follow the narrative.
• Adding a baseball field in your dashboard would be really cool + unique!
• In terms of implications, it might help to look at results by team, to see if any particular team started doing better or worse.
• If you end up working more on this project, you should definitely use differences in differences.

[ernbilen]

I really like your data, the fact that it is play-by-play for NCAA is great!

• Just to double check, does your analysis use games that Dickinson played only? You can consider EDA with line charts where on the x axis you have game # in the season, like 1st game, 2nd etc. and y axis could have different things, like points, fouls etc. One line per team. Of course showing all teams would make this pretty cluttered, so maybe you can pick 5 teams that are Dickinson's rivals.
• The PPP graphs: I think instead of scatterplots here you can consider group by PPP Rating and take average PPP. So it would be a bar chat with categorical values of PPP Rating on the x axis and numeric PPP scores on the y axis. This would be similar to your offense/defense graph in the next slides (you can keep those at is because it is cool to look at offense/defense separately) but the difference would be you'd be reporting averages under each category.
• Another cool thing to consider is player level analysis. Who are Dickinson's all time greatest players? By position?