The following insight was derived from The ULTIMATE Dataset. I’ve debated whether or not to publish some of my own explosive findings, but ultimately decided the tremendous potential of this data is best understood by example.

The numbers used for today’s piece are from DraftKings, but the FanDuel implications are the same

Giannis burned you- yes, you the general public. “I like Giannis tonight,” you said to yourself while dreaming of all the gently used DVDs you’d buy with your winnings. But things didn’t go according to plan. After a nice first half, the points stopped coming. By the end of the night you were looking up at the min-cash line like Bane from his dark, prison-like hole. What went wrong? The matchup was good. Giannis was 20% owned- it’s not like you were the only sucker here. I suppose the goddess Fortuna simply did not favor you. If only there were a way to better avoid those Giannis bust games that sink any chance of taking down a GPP…

Enter the power of historical analytics. Let’s start with the basics- we consider 6.0x value to be a good return on your GPP investment on DraftKings. This would put you solidly into the cash most nights. This is even more true for a high-salary player like Giannis, who allows you to hit value with a large chunk of your salary cap.

Now, let’s look at the single biggest key factor in predicting Giannis success- the margin of victory. We’ll look at 120 games over the past 2 years (there are a handful of additional games, but without salary data as Giannis was not on the main slate), and let’s compare the numbers:

Bucks win by 16+ (40 games) – 6.0x value hit: 12.5%
Any other win/loss margin (80 games) – 6.0x value hit: 27.5%

Wow- that’s a big difference- more than twice as often to hit significant value in non-Bucks blowout games! “Well, maybe 6.0x is an arbitrary line and the difference isn’t really so stark,” you blubber as you realize how much money this single fact has cost you. OK- let’s look at overall value:

Bucks win by 16+ (40 games) – Overall Average Value: 4.7x
Any other win/loss margin (80 games) – Overall Average Value: 5.4x

That’s powerful – Giannis must be avoided at all costs in Bucks 16+ blowouts, yet is nearly a must-play in other scenarios. “Yeah, but sometimes the cash line is like, super-high, and 5.4x isn’t THAT much,” you blurt through misty eyes. Fine, let’s look at where the big money is made- the 6.5x value line:

Bucks win by 16+ (40 games) – 6.5x value hit: 2.5%
Any other win/loss margin (80 games) – 6.5x value hit: 20.0%

6.5x is an outstanding return, and highly likely to give you an elite finish. If you didn’t own Giannis on these nights, you had little hope of actually winning your GPP. “B-b-b-b-but who can predict a blowout? They’re so random!” you manage to get out as you shake like peak Don Knotts. I’m not enjoying your demeanor, but let me address your question- what if you kind of *COULD* predict them?

Bucks win by 16+ (40 games) – Average Vegas Point Spread: Bucks -9.7
Any other win/loss margin (80 games) – Average Vegas Point Spread: Bucks -6.9

Yukon Cornelius! That, my digital friends, is gold- the higher the spread, the more likely we are to see a Bucks 16+ blowout. When we get a Bucks blowout, we’re much more likely to get a Giannis bust that sinks our night. Whereas in the tighter games with tighter spreads, Giannis is often the meal ticket to the top of the GPPs. “Oh, that’s it? Well EVERYONE knows not to play stud players in blowouts!” you mutter as you haughtily straighten your ruffled shirt. First of all- this rule is false- it is very player-dependent (if there is enough interest, maybe I’ll do a piece on a certain stat-hunter who actually performs *better* in these games).

Second, very clearly- everyone does NOT know this:

Bucks win by 16+ (40 games) – Average Giannis GPP Ownership: 20.2%
Any other win/loss margin (80 games) – Average Giannis GPP Ownership: 19.1%

Well, we don’t have gold anymore. We have gold made of diamonds and rubies and Tiberium and Bitcoin in a bull month. The incredible weight of this finding is that is both a value play- helping you to make better +EV lineup decisions independent of the field- AND a leverage play- since the field is not adjusting for this variable, you can own Giannis in his most-likely-to-succeed games without any chalkier ownership, and you can avoid him when he is still (incorrectly) significantly owned by the rest of the field. THIS is the power of historical data analysis.

Well- I hope you’ve enjoyed this edition of NBA Analytics Master Class. Happy to hear any feedback/questions in the comments or via email. I’m certainly not giving away all I know, so find true value by picking up the ULTIMATE Dataset for yourself.

Learn the Game. Beat the Game. Thanks for reading!

The primary purpose of this website is to offer you the data to find your own valuable insights. Therefore, do not expect these insights to be regularly updated- this is not a DFS tout site. However, these articles demonstrate the incredible value made possible by analyzing the data in the right way and creating an informed decision-making process