Use Our New Metric To Predict Minutes In Fantasy NBA

In Daily Fantasy NBA, figuring out who is going to play a lot of minutes is half the battle of figuring out the top plays of the day. The most obvious way to predict minutes is to look at the past game or couple games and simply look at how many minutes a given player has played. But that method can be problematic, because blowouts, overtime games, and other factors can cause minutes to drastically fluctuate and will mislead you on future minute predictions. At Daily Fantasy Winners, we’ve developed a new tool that sorts through the noise and gives you a great predictive stats of minutes for NBA players.

Why Predicting Minutes Is So Important

If you’re not on the floor, you can’t score fantasy points. The amount of minutes a player spends on the floor is the amount of opportunity they have to score fantasy points. As their opportunity increases, so should their expected fantasy points.

Let’s assume a player averages 1 fantasy point a minute, which is what you’d find for a good starter on an NBA team. Let’s say for various reasons we expect this player to play 5 more minutes than usual. Assuming a neutral matchup, we can expect this player to score 5 fantasy points above their average. If our target fantasy points per $1000 spent is 5, than this player gives us $1000 in cap value, which is huge. If we can predict minutes, we can strongly predict fantasy performance.

Why Predicting Minutes Is Hard

The hardest part of predicting minutes is that past minutes aren’t always indicative of what we can expect from a players future minutes. Sometimes a player will have played in one or two overtime games in his past couple of games, which will drastically inflate his minutes played and make him seem more appealing. Sometimes a player will have played in one or two blowouts, causing him to play much less minutes. But that shouldn’t make our expectations for his minutes in the next game be any lower. Looking at minutes leaders in the past couple games will be dominated by players who played in an exceptional amount of close games or overtime games, which does not help us predict who will get big minutes in the day’s set of games.

Introducing Minutes Above Starter Average

What we really want to look at when we are predicting future minutes from past data is this: How did a player’s minutes compare to other starters? In overtime games, every starter will play a lot of minutes. If Kevin Love played 38 minutes last night, that looks like a lot. But if Lebron James played 45, Kyrie Irving played 46, Timofey Mozgov played 42, and JR Smith played 40, does that number look good anymore? Love actually played less minutes than the average starter. Therefore, we should attribute Love’s minutes in the past game to luck factors  (overtime or excessive closeness), not an indicator of high future minutes. If anything, we should not expect high minutes from Love in a more reasonable game.

Comparing minutes to other starters also works well when looking at blowouts. Andrea Bargnani getting 28 minutes in a game on March 14th against Golden State doesn’t look like a lot. But the game was over by the beginning of the third quarter, and no starter played over 28 minutes. Bargnani was actually second in minutes on the team, and we should probably expect his minutes to be quite good in a game that isn’t a total blowout.

At the tools section of Daily Fantasy Winners, you can find a tool called NBA Minute Trends that shows precisely this data. The data in this tool shows you how many minutes a player has received above an average starter for their team. The first column gives you a 3 game moving average, while the 2nd column shows you the last game alone. Players with high positive values in either column should be expected to have high minutes in their next game, while players with high negative values are at risk to play little minutes. The last column shows the difference between the 3 day and 10 day moving average, which will show you lower minute players who have recently received big minute increases.

Flaws Of This Metric

While this metric is a significant upgrade over looking at minutes for each given player over the past couple games alone, it still has some issues. It probably has some issues I’m not even aware of, so don’t be afraid to leave feedback on what you see as this metric’s flaws.

First off, although I have used the term “starter minutes” throughout this article, the actual data used for average starter minutes is simply the players with the five highest minutes for that team in that game, even if they weren’t a starter. This actually helps in situations where starters are injured early in the game, as if we used actual starters it would cause some players to look significantly better than they should.

Secondly, teams with bigger rotations will have a lower average starter minutes, and will cause the big minute players on their team to look better. So when using this stat, it’s important to consider what you can expect from the top five minute-receivers on any given team. For example, today Khris Middleton has a 6.09 EMAMinutesASA3 (look at tool for definition), but he plays on a team with a large rotation and lower top five player minutes than most NBA teams. Victor Oladipo has a lower EMAMinutesASA3 at 4.42, but has been arguably more impressive in the minutes department because he plays on a team where starters play very heavy minutes. Still, Middleton and Oladipo have both been huge in the minutes department, and both should have high minutes projections for their next game.

Lastly, this metric can get messed up by mid-game injuries. If a starter get’s injured in the first quarter and leaves the game, he is going to have shown up as playing very few minutes. This will bring down the average starter minutes and make the rest of the starters look more appealing. This will also bring down his averages and make him look less appealing once he returns from injury.




  • A small increase in projected minutes can result in a large increase in expected fantasy points. Minutes are one of the strongest projection signals you can use.
  • Looking at minutes over the past several games for any given player can paint a poor picture of future minute expectations because of factors in the previous games, such as overtime, that caused a luck based minute increase.
  • While our analysis is not perfect, you can use our Minutes Above Starter Average metric to predict minutes for any given NBA player better than looking at the past few games of data will.

View all posts by Daniel Steinberg
Daniel Steinberg

About the Author

Daniel Steinberg Daniel Steinberg is a former bond trader at a multi-billion dollar proprietary trading firm in Chicago. He uses his knowledge of statistics and his creativity from his career as a poker professional to create the most advanced Daily Fantasy statistical analysis that you will find anywhere. Follow him on twitter @DanielSingerS

no banner found

Daily fantasy sites