NBA DFS (daily fantasy sports) players are often concerned with the risk of a blowout. Particularly for star players, the idea is stars can’t hit value if they’re only playing 3 quarters.
I used the complete 2017-18 season, spread data from OddsShark.com and box score data from Basketball-Reference.com. The scoring is based on DraftKings (DK) scoring system.
There are three different ways to look at the spread set by Vegas:
- Spread – this is the spread for the game
- absSpread – absolute value of the spread for the game
- netSpread – game spread minus the team’s average spread for the season
Players have also been put into four different groups by their production, with the count in parentheses and a minimum of 10 games played:
- Star (24) – Min. 25 minutes per game (MPG) and 1.225 DK points per minute (DKPG)
- Starter (139) – Min. 25 MPG, less than 1.225 DKPG
- Bench (206) – Less than 25 MPG
- Scrub (83) – Less than 12.5 MPG
Using these qualifications, there does not appear to be a significant risk to star’s production in games with large spreads.
NBA DFS production is viewed two ways: DraftKings points and net DraftKings points (this is the difference between a player’s DraftKings points and their season average).
The top left graph shows DraftKings points vs Spread and indicates a clear negative relationship. However, this is misleading. Not all stars are equally productive. Net DK Points adjusts for the players expected production, and is therefore more accurate.
Note the Net DraftKings points vs Spread graph (upper right). The trend line is only slightly negative. This difference indicates that more productive stars tend to be on better teams, aka teams with lower spreads – indicating they are the favorites. It also indicates that stars tend to perform near expectations, regardless of spread.
For both DK Points and Net DK Points vs absolute spread (middle row), there are essentially straight trend lines. This indicates that a game having a large spread doesn’t necessarily affect star production.
In the bottom row, you’ll see production vs net spread (the difference between the game’s spread and the team’s season average spread). While there is a slight negative trend, it is hard to make a strong statement off those graphs.
Looking at the correlation numbers, you will see that there are very weak trends for each group.
DraftKings points vs Spread, grouped by player role:
| Role | Spread | absSpread | netSpread |
|---|---|---|---|
| Bench | 0.017 | 0.029 | 0.014 |
| Scrub | 0.057 | 0.012 | 0.062 |
| Star | -0.074 | 0.026 | -0.087 |
| Starter | -0.026 | 0.004 | -0.032 |
Minutes vs Spread, grouped by player role:
| Role | Spread | absSpread | netSpread |
|---|---|---|---|
| Bench | 0.057 | 0.022 | 0.068 |
| Scrub | 0.088 | 0.001 | 0.098 |
| Star | 0.01 | -0.074 | 0.012 |
| Starter | 0.035 | -0.042 | 0.045 |
These relationships are extremely weak. Overall, it appears the blowout risk is overstated in the NBA DFS community. If you do want to watch the spread, being a large underdog does appear to be more of an issue than being a heavy favorite for star players.
Other NBA DFS Links: