The Oracles (and shams) of Mock Draft Predicting

Deandre Ayton may have gone first in the draft but critics argue if Michael Porter Jr. is healthy he could be worth a top two pick, yet due to concerns about his back he fell all the way to the Denver Nuggets with the 14th pick. (Clutchpoints / Virgil Villanueva)

Last week we took a look at 20 NBA mock drafts in an effort to compile the combined data into one to create a meta-mock draft, or what’s known as a “consensus draft.” This week we’re looking back to see who was right and wrong.

The hypothesis? There is strength in numbers. Maybe, just maybe, if you get a bunch of NBA fanatics together you can predict erratic GM behavior better than individual mock drafts can. So was it true?

The methodology was simple. A spreadsheet was created juxtaposing players to where each individual mock draft had them falling, then the numbers were averaged, and our consensus draft order was created. From there, we were able to compare mock draft correlation to the actual draft order in order to rank the mock drafts in terms of correctness.  Note none of this is adjusted for trades, just the raw original picks. To find any of the mock drafts links, mock draft orders through the first round, the actual draft order, or any of the data analysis discussed here please follow the link to the spreadsheet.

Loser:

The biggest sham in these mock drafts? None other than NBA Insiders multiple day front page feature on ESPN.com which had a correlation to the actual draft of 0.697. The NBA Insiders would have had the draft going as follows for just the lottery:

Actual Draft Order NBA Insiders Predicted
Deandre Ayton 1 1
Marvin Bagley III 2 5
Luka Doncic 3 2
Jaren Jackson Jr. 4 3
Trae Young 5 9
Mohamed Bamba 6 4
Wendell Carter Jr. 7 7
Collin Sexton 8 11
Kevin Knox 9 14
Mikal Bridges 10 10
Shai Gilgeous-Alexander 11 8
Miles Bridges 12 15
Michael Porter Jr. 14 6
Lonnie Walker IV 18 12
Robert Williams 27 13

Needless to say this was surprising. You would think the Stephen A. Smiths of the world would be better suited than an outsider in the realm of draft predicting. If it’s a testament to anything, it is that the ESPN Insiders might just be a clad of clickbait composers. A majority of the site is just player rankings, and the rest is what if scenarios – it is fun to think about, but probably not worth paying for.

Winner:

The best mock draft in the analysis was the Los Angeles Times’ Clippers beat reporter Broderick Turner.  The winning mock draft order for the lottery was as follows:

Actual Draft Order Broderick Turner Predicted
Deandre Ayton 1 1
Marvin Bagley III 2 2
Jaren Jackson Jr. 4 3
Mohamed Bamba 6 4
Luka Doncic 3 5
Trae Young 5 6
Wendell Carter Jr. 7 7
Michael Porter Jr. 14 8
Mikal Bridges 10 9
Kevin Knox 9 10
Collin Sexton 8 11
Shai Gilgeous-Alexander 11 12
Miles Bridges 12 13
Robert Williams 27 14
Zhaire Smith 16 15

For an impressive 0.888 correlation to the actual draft order and a first place ranking in terms of correctness, we would like to present you with this, Mr. Turner for your “NBA mock draft 2.0“.

Broderick Turner Trophy

Congratulations and I hope this trophy finds you well. Your prodigal picking abilities and succinct analysis made for a great mock draft.

Consensus Analysis Results

The consensus analysis garnered a correlation score of .833 and is ranked 8th out of the 21 mock drafts (20 mock drafts and the meta mock draft) surveyed. For the first fifteen picks the order was close to correct, but not quite close enough:

Actual Draft Order Consensus Draft Predicted
Deandre Ayton 1 1
Marvin Bagley III 2 2
Luka Doncic 3 3
Jaren Jackson Jr. 4 4
Mohamed Bamba 6 5
Michael Porter Jr. 14 6
Trae Young 5 7
Wendell Carter Jr. 7 8
Collin Sexton 8 9
Mikal Bridges 10 10
Kevin Knox 9 11
Shai Gilgeous-Alexander 11 12
Miles Bridges 12 13
Robert Williams 27 14
Lonnie Walker IV 18 15

This places it towards the middle of its competition, which negates the hypothesis that there is predictive ability of compiling a group of mock drafts into one. It makes sense that since the consensus draft was based off averages that the consensus would fall towards the middle of the ranking because there were drafts that were considerably more correct than others. Possibly if the consensus was based off of modes instead of averages it would’ve had a different result.

Between the oracles and the shams, fell the consensus. Everyone has a different way of trying to predict draft order, and whether it is based on player talent, athlete potential, team needs, league rumors, or compiling all them into one. It can still be wildly unpredictable what sort of moves a GM is going to make on draft day, but it seems like there was a good amount of truth in this sample of drafts since the correlations across the board were so high.

 

 

 

 

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s