Coming soon!  We will be adding an option for users to create their own projections.

Inspired by Rob Vollman's Stat Shot. 

 

 

I applied the data from the Age Regression in Players Pts Total that I calculated for the last 10 NHL seasons to make projections for the 2018/19 season.  In the table below are Pts Total and Pts Rank projections for the 2018/19 season for all (and only) players who played in the 17/18 season.  Three sets of projections were made based on the 10 Year Average Change in Pts Total, 5 Year Average Change in Pts Total, and the Previous Years' Average Change in Pts Total.

The burning question of course is, “How good are these projections?”.  Using the same data, I made Pts Total projections for the 17/18 season and then compared these projections to the projections of the other forecasters.   On average the projections (based on the 10 year average age regression) from 2017/18 had an average absolute error of 12.89 Pts/Player for the Top 300 projections.  When compared with other forecasters, these projections based solely on Age Regression ranked 9th out of 16 sets of projections.  They didn’t create the most accurate projections, but surprisingly they did beat 4 sets of projections from other forecasters. Only time will tell how they do in 18/19. 

If you haven’t had the opportunity to read Rob Vollman’s article “Points regression for the Norris finalists”, I suggest that you do.  It is a great article that gives a little insight as to how forecasters make their projections.  He also describes where the variation in forecasters’ projections arise from.  On top of this insight, he more than backs up the points that he makes with three near perfect Pts total projections for Burns, Hedman, and Karlsson in 2017/18!

Elan and Brian run the podcast Keeping Karlsson. 

In their September 18th episode ( http://www.keepingkarlsson.com/e/no-156/ ) they took a look at the most disagreed upon projections from 6 of the forecasters that we ranked and compared this season.  I took the liberty and followed up on the outcomes of these players and which forecaster had the best projections for them.   It is important to note, that Elan normalized the projections by taking Forecasters PPG projection and then multiplied it by 82 games. Refer to the table below. 

 

Back in the Spring when I first tried to replicate Rob Vollman’s instructions on how to make projections I missed a vital step. I forgot to account for and remove the random variation in my weighted Pts Total before adding the factoring in the age regression. This didn’t happen because I didn’t know that I had to, I just didn’t know HOW to do it. Not until I purchased Stat Shot by Rob did I get a better understanding of how to do it.