With access to the NBA player stat projections that drive our winning DFS player point projection models, you can apply advanced sports analytics to player prop bets and get an edge on the sportsbooks!

Our NBA projections are managed by Stokastic’s team of experts, including Alex Baker, Rich Green, Shawn Zhan, and Steve Buzzard, and updated throughout the day as more information becomes available up until the last game tips.

Key Metrics:

Win% – Our predicted % of the time you win this bet based off of our projections, book lines, historical results, mean/median distributions and simulations.

EV – Expected value – also known as ROI. This is the weighted average of returns. For example 60% chance of winning a +100 bet has an EV of 20% (.6 * 1 + .4*-1)

sBet% – This is our refined version of Kelly Criterion optimal bet percentage that is designed to maximize your bankroll in the long term. For example, if you had a 20% ROI bet, betting 100% of your bankroll would cause you to inevitably go broke, but betting 1% might not be maximizing your long term growth. We do the math to help you determine the correct % of your bankroll to bet.

sEV – This metric measures the % your bankroll is expected to grow if you were to place this bet. We factor in how much of your bankroll you should invest (sBet%) as well as negative geometric drag to calculate this. This is the default sort because at the end of the day, as a bettor you are trying to maximize the growth of your bankroll and this captures that. 

What is the difference between EV and sEV?

The key difference is that EV measures the growth of a single bet, whereas sEV measures ROI in terms of the growth of your entire bankroll.

Why use sEV?

Example 1:

Would you rather place a bet where you have a 100% chance to win and an EV of 20% or a bet with a 1% chance to win and the same EV of 20%?  Clearly the first option is better because you can safely invest every last penny of your net worth and you would be able to fully realize that 20% EV whereas with the 1% win bet you could only invest some amount < 100% of your bankroll.

Example 2:

Lets begin with a few assumptions –

You have a 0 EV bet where you have a 50% chance of doubling up and 50% chance of losing your bet.

You are betting 10% of your bankroll on this bet.

Now we can say that the two possible outcomes, each with a 50% chance of occurring, are that you end up with 110% of your bankroll or 90% of your bankroll.

If you were win to this bet once and lose once you would end up with 110%*90% = 99% of your bankroll for a net loss of 1%! The general way to calculate bet returns this way is winReturn^(win%)*lossReturn^(loss%). This is the essence of the concept of negative geometric drag also known as volatility tax in the finance world which we’ve factored into our sEV calculation.

Considerations – 

Be sure to check if there are any key questionables on a player team before placing a bet. Our current projections are assuming questionable players are IN until they get ruled out. For example, if the tool is saying that Jaylen Brown under points is a good bet, but Tatum is questionable then avoid taking the bet unless you are confident Tatum will play.

We are currently working on fixing a bias towards the over in the underlying algorithms. For the time being  this means is that you should focus primarily on unders and only take overs that really stand out. This will be fixed in an upcoming update.