The Stats Behind Back to Back NBA Games
We constantly hear announcers or pundits talk about players resting because of back to back NBA games. If we’re resting players, it must be because they are extra tired the day after a game. If they are extra tired, reason would suggest teams would perform worse the second day of back to back NBA games.
But do the stats back this up? We want to look at the stats surrounding back to back NBA games to see if there actually is a difference. We’ll look at a few different years to see if things change over time as well.
Click here to jump ahead to our conclusions!
To receive email updates when new articles are posted, use the subscription form below!
How Much Worse are Teams in Back to Back NBA Games?
The easiest way to measure how back to backs impact NBA teams is by looking at the average margin of victory in the second night of a back to back. We started by looking at the data from the 2022-2023 NBA season and found the data below (This data is not our final conclusion, in fact it is an anomaly. Keep reading to see why!):
2022-2023 | Margin (All) | Margin (B2Bs) | Diff |
|---|---|---|---|
Home | 2.6 | 1.6 | -1.0 |
Away | -2.6 | -5.1 | -2.5 |
The first thing we notice is that home court advantage was worth about 2.6 points in the 2022-2023 NBA season. This is squarely in line with the 2.5-3 number that most NBA models use.
Now look at the margin in back to back NBA games. Playing at home in the second night of a back to back only results in a +1.6 point per game margin. This means that at home, teams played 1 point worse the second night of a back to back.
On the road, the difference is much larger. Teams playing away from their home court in the second night in a back to back are 2.5 points worse than in generic road games.
Combined, teams were a net 1.9 points worse in back to back games than in other generic games. This was computed over a sample of 401 back to back games in the 2022-2023 season. The dataset of margins of victory has a standard deviation of 13 points, so this results would be considered statistically significant.
What about other years? Does the data change in other seasons?
Data from Other Seasons
How specific is the impact of a back to back by season? Lets look at data from a few other seasons to check. First, the 2021-2022 season:
2021-2022 | Margin (All) | Margin (B2Bs) | Diff |
|---|---|---|---|
Home | 1.8 | -1.3 | -3.1 |
Away | -1.8 | -3.1 | -1.3 |
This data shows a similar overall trend: the performance difference in back to back NBA games is worth about 2.0 points. However, this time the difference at home is larger than the difference away.
We’ll skip a few seasons to avoid the COVID time frame, and look at data from the 2015-2016 season:
2015-2016 | Margin (All) | Margin (B2Bs) | Diff |
|---|---|---|---|
Home | 3.1 | 2.9 | -0.2 |
Away | -3.1 | -3.6 | -0.5 |
Here, the difference almost entirely disappears. Teams were only about a half point worse in the second night of back to backs! Is this a fluke, or did something change?
2014-2015 | Margin (All) | Margin (B2Bs) | Diff |
|---|---|---|---|
Home | 2.4 | 0.0 | -2.4 |
Away | -2.4 | -2.3 | +0.1 |
The average difference here (weighted by number of games played) is only 0.6 points worse in back to backs. Let’s check one more…
2013-2014 | Margin (All) | Margin (B2Bs) | Diff |
|---|---|---|---|
Home | 2.5 | 1.4 | -1.1 |
Away | -2.5 | -4.1 | -1.6 |
Here, the average difference is closer to one and a half points.
What does all this mean?
Hypothesis on Back to Back Trends
There are two hypotheses for why teams could perform worse in back to back NBA games. The first is tiredness, but the second is kind of a self fulfilling prophecy. And, as we’ll see the data backs up that both of these hypotheses are probably true!
The basic hypothesis is that teams playing the second night of a back to back NBA game are tired. Basketball is exhausting, and with stars playing roughly 40 minutes a night, it is hard to expect guys to be as good the second time around.
The second hypothesis is that teams – expecting that they are going to perform worse in a back to back – rest their starters and/or don’t play quite as hard. We know that this happens in the modern NBA. Players in the modern NBA get a lot of games off. The second night of a back to back is the perfect time to rest stars because it is so physically demanding.
Stars resting is a relatively new phenomenon, though. The NBA is currently in an era often called the player empowerment era. This era is headlined by: dictating where you play and who you play with and more rest for stars.
My hypothesis is that we can “see” the player empowerment era in this data. In 2023, teams were much worse in back to backs than in the mid 2010s. This isn’t because players are more tired, my idea is that this is because players rest more often in back to backs.
Back in the 1990s, hardly anybody rested. Take a loot at the best players (go here and sort by VORP) played almost the whole season, some all 82. Guys just didn’t take time off. Now look at the data from back to back NBA games from the 1990 season below:
1989-1990 | Margin (All) | Margin (B2Bs) | Diff |
|---|---|---|---|
Home | 5.1 | 5.0 | -0.1 |
Away | -5.1 | -5.7 | -0.6 |
The average overall team quality decreased by only about a half a point in the 1989-1990 season on the second night of a back to back. This is consistent with some of the numbers from the mid 2010s before players started resting regularly.
Point Difference in Back to Back NBA Games
Looking at all the data above, we think that “teams being tired” the second night of a back to back makes the teams play about a half point worse. This difference isn’t huge, but it is noticeable.
In the early 90s and even the mid 2010s, we didn’t experience players resting the second night of back to backs nearly as often. Fairly universally in this era, teams were a half point worse the second night of a back to back.
The data from post-2020 or so can be explained by stars resting. It isn’t that players are getting more tired, it’s that teams are purposely handicapping themselves the second night of back to backs more often. This makes it so that the point differential we measured in the most recent seasons is more extreme.
To summarize, back to backs hurt teams by about half a point. In the modern NBA, the effect is larger because starters tend to rest the second night of back to backs.
We’ll close out by looking at some data regarding how common back to back NBA games are.
How Often do Teams Play Back to Back Games?
We looked at all games in the 2022-2023 season and counted how long teams spent in between any two games. The distribution below is what we found:
Notice that “1 day between games” means games on back to back days. We can see that overwhelmingly, games are played with one day of rest in between. This accounts for about 63% of games. About 16% of games were the second night of back-to-backs. The same amount were played with 2 full off-days between. Longer was rarer.
This is actually fairly common. This means that teams can’t rest their best players in all back to back NBA games because that would nearly guarantee a loss in 16% of games! Teams must pick and choose where they rest their starters.
We also looked at the distribution of time between NBA games back in the 1990 season for comparison.
Back before the player empowerment era, back to backs were much more common. A full 25% of games were played by guys who just got off the court the night before.
You can kind of tell that the schedule in 2022 was made by a computer and was “optimized” to make 2 days between games as common as possible. The 1990 schedule and rest day distribution looks much more natural, like something made by hand.
Future Study
We reached the conclusion that teams played about 0.5-1 points worse the second night of back to back NBA games. We determined this by looking at multiple years of data and realizing that the larger differences in recent years were due to starters resting more than anything else.
One could formalize this study with quite a bit of extra work, though. Using modern NBA model (like our Sparse Impacts NBA Model) we can “factor in” that starters rest. For example, we could figure out that Kawhi Leonard is worth about 7 points to his team. Then, we could look at how the Clippers played the second night of back to backs.
We would expect them to be about 7.5-8 points worse if Kawhi sat. 7 of these points are from Kawhi not playing, the other 0.5-1 points would be due to the guys being tired.
Going through all the games and factoring in starters resting can help us evaluate whether or not the lack of rest is actually worth about a half point.
To receive email updates when new articles are posted, use the subscription form below!
