Analytics made easy. Release 1.0 of Advanced Stats Viewer

Both RAPM and WS have their peculiar quirks. I prefer to go to Graphs and then plot RAPM versus WS. This reduces the effects of outliers. Choosing a position helps to clear up the players who you are not interested in. You can then make a better decision as to who is more valuable in the NBA.

A Bubble view is coming. An amazing tool. Who are the best and most productive players in the NBA? It becomes obvious.

 

Well Thank you JCDenton. I did after a cooling off period take your criticism to heart.

 

This bbs just reeks of an infatuation with gossip and MSM. Just saying.

 

If you go to the default graphs of the application you can easily measure the relative worth of a player when salary is taken into consideration. Top left are the most valuable players in the NBA. Low salary and major performance. We see that Howard is what I would call a fairly valued player. If you look at WoW or other sites he is mentioned as a "fairly valued player". Because he has great performance, but is also being paid a truck load of money. By the metrics on the graph Sanders is the most undervalued center in the NBA, but he just signed a new contract and next season he will slide far to the right.

You can repeat that analysis for all of the other positions.

 

Thank you Durvasa. I have updated Tableau to reflect the problem. I coded the entire graph tab in about an hour and just did not include the minute adjusted RAPM and WS in the choices. My mistake.

And I will take a look at how you arrived at Wins Added. I have all of the information sitting in Tableau, I just need to modify the graphs to reflect it.

 

Graphs now use team logos when displayed.

 

Question: Do Lin and Harden excessively turn the ball over?

Answer: Go to Graphs and plot TOR (turn over rate) versus USG (Usage). Choose the PF and PG positions and then locate Lin and Harden on the graph. We see that neither Lin nor Harden has an excessive turn over rate compared to their peers. If they both could drop their turn over rate 2% (equivalent to one turnover every 50 possessions) they would compare favorably to the best players players at their position.

 

It actually is the sum of base RAPM and WS. It would be more accurate if it summed Minute Adjusted RAPM and Minute Adjusted Win Share. Sorry. Durvasa pointed this error out earlier.

 

I have added zoom control to the graphs. It helps make sense of crowded areas.

 

The 3rd control from the top allows you to use a magnifying glass to center the zoom. Does that fix the problem?

 

I have decided to move the normalization applied to the data from minute based to possession based. It seems to give better results from the tests I have done. A big job but someone has to do it.

 

So CLUTCH why cannot I edit this post??? If I update the cloud stored application there is a high probability that the tool will become non-functional. The multi-quote enhancement broke my ability to edit this post. And many others.