Back-to-Back College Basketball Games Show Value in not Overreacting

Back-to-Back College Basketball Games Show Value in not Overreacting

March 16, 2021 – by Jason Lisk

Going through two of the same opponent is harder than facing just one (Photo by Chris Williams/Icon Sportswire)

The Coronavirus pandemic has created a unique scheduling situation in college basketball this season. Prior to this year, it was pretty rare for teams to play two games in a short period of time at the same location. 

Traditionally, conference teams, if they play two games against the same opponent, have almost universally done so by each getting a home game. Usually, those games are several weeks apart as well, with games against other teams taking place between the two matchups. 

The schedule this year for some conferences, though, has featured teams facing off twice in a short span at the same venue, in order to minimize travel. These pairs of games provide new insight into college basketball teams and their performance fluctuations.

The Overall Data 

In games completed by February 17, 2021, there were 466 back-to-back matchups played at one team’s home venue. For our purposes here, “back-to-back” means two games played with two days of each other on the calendar, which mostly includes teams playing on consecutive days, but also includes some series played with one day of rest in between (The Mountain West, for example, generally played two games in a three-day span). 

Here is a general summary of the data:

	Game 1 Spread	Game 1 Margin	Game 1 Win Pct	Game 1 ATS Pct	Game 2 Spread	Game 2 Margin	Game 2 Win Pct	Game 2 ATS Pct
All Home Teams	-2.30	+2.26	59.2%	49.4%	-2.01	+2.13	55.8%	49.8%

Colgate: A Case Study in Crests and Valleys

The above data is a composite summary of the overall results of the back-to-back games. But the devil is in the details. 

Take Colgate as one extreme case. Colgate is an anomaly in many ways this year. They have only played three opponents as the Patriot League went to a conference-only schedule and further divided the conference up into groups to reduce travel. Colgate is currently inside the Top 10 in the NET rankings as they’ve broken the algorithm with the combination of some extreme results and lack of connectivity with any other basketball programs.

Here are Colgate’s results in games played on consecutive days against the three opponents they have faced so far:

Opponent	Game 1 Margin	Game 2 Margin	Difference
vs. Army	44	-2	46
at Boston U	7	44	37
vs. Holy Cross	40	9	31
at Holy Cross	11	18	7
at Army	10	9	1
vs. Boston U	10	15	5

That’s an average swing of 21.2 points between the Game 1 and Game 2 result played a day later. In three series, they have won one of the games by 40 or more points, while the other was decided by single digits against the same team.

If Colgate has had those massive swings in games against the same teams, and has only played three teams all year, how should we view them? And how should we view the impact of one game result on what it tells us about a team?

 

Interested in pick analysis for your 2021 NCAA Tournament Bracket Picks? Visit our NCAA Bracket Picks 2021 and see why an average of 65% of responding subscribers per year have reported winning at least one bracket pool prize over the last five tournaments. 

To see additional articles on the NCAA Tournament and NCAA Bracket Picks, check out our Articles page.

Home Teams that Cover the Spread in Game 1, Versus Home Teams That Do Not 

Of course, Colgate is one example, and perhaps an extreme outlier. Here are the results grouped by whether the home team was favored or an underdog in Game 2, whether they won, and whether they covered. 

Category		Game 1 Spread	Game 1 Margin	Game 1 Spread Margin	Game 2 Spread	Game 2 Margin	Game 2 Spread Margin	Game 2 Win Pct	Game 2 ATS Pct
Home Fave/Won+Cover		-6.3	+15.3	+9.0	-6.6	+5.4	-1.3	67.8%	47.5%
Home Fave/Won+No Cover		-10.6	+6.2	-4.4	-9.7	+9.4	-0.3	78.5%	50.8%
Home Fave Lost		-4.3	-7.9	-12.2	-3.3	+3.9	+0.6	59.3%	55.2%
Home Dog/ Won		+3.8	+8.3	+12.1	+3.4	-2.7	+0.8	38.2%	47.1%
Home Dog/Lost+Cover		+8.2	-4.3	+3.9	+7.4	-7.4	0.0	29.4%	47.1%
Home Dog/Lost+No Cover		+5.6	-15.5	-9.9	+6.4	-4.9	+1.6	34.5%	50.0%

Overall, home teams that covered in Game 1 only covered in Game 2 47.2% of the time

Home teams that did not cover in Game 1 covered the spread in Game 2 52.2% of the time.

Which means that the team that did not cover the spread in the first game did so more often than not in Game 2, covering the spread 52.4% of the time.

The group that performed the best in Game 2 were home favorites that lost outright in Game 1. When that happened, they bounced back and covered 55.2% of the time in the second matchup.

What if the Line from Game 1 Did Not Move?

Of course, what tends to happen when a team covers the spread in the first game, particularly if they cover by a comfortable margin, is that the spread tends to adjust for Game 2. 

On average, the spread moved 0.74 points in favor of the team that covered the spread in game 1. For example, if the favorite won and covered the spread, they would on average be favored by 0.74 more points in the second contest. 

As you’d expect, the size of the spread movement was influenced by how many points a team covered the spread in the first game. Results that were further from the spread tended to result in bigger closing line differences between Game 1 and Game 2.

So how much of the Game 2 value in favor of the team that did not cover the spread in Game 1 is based on that line movement? As it turns out, most of it.

If every single matchup in our data set had used the Game 1 point spread for Game 2, without any adjustments, the results would have been:

When the Game 1 favorite covered the first spread, they would have also covered the same spread 49.8% of the time in Game 2.When the Game 1 favorite did not cover the first spread, they would have covered the same spread 50.6% of the time in Game 2.Overall, the team that did not cover in Game 1 would have covered the same point spread 50.4% of the time in Game 2.

The spread cover rate for Game 2 would have been very near 50% if the Game 1 spread was used again.

That teams that failed to cover in Game 1 covered an additional 2% of the time (52.4% overall) after line movements between the two games provides some evidence of the market overreacting to the results of one game, even between the same opponents and at the same venue right away, can provide value if you avoid overreacting to one game.

The Biggest Covers in Game 1 Have Led to Game 2 Opportunities 

Earlier, we discussed Colgate’s extreme splits from Game 1 to Game 2. But Colgate is not alone. Here are the results when one team covers the spread by at least 15 points in the first contest. 

When the home team covers by 15+ in Game 1: home team is 63.6% SU and 48.9% ATS in Game 2 (n=44)When the road team covers by 15+ in Game 1: home team is 50.0% SU and 56.3% ATS in Game 2 (n=48)

In the situations where a home team has been embarrassed at home, and they get an opportunity to come back and play the same team right away, they have covered 56.3% of the time. 

Those results aren’t because of arbitrary line drawing. In fact, home teams that failed to cover by at least 16 points were 24-14 ATS (63.2%). Fifteen points was just a nice, round number. But the sample sizes become smaller and smaller as you continue to move out. For example, in games where the home team failed to cover by more than 25 points in Game 1, they were 7-2 ATS in Game 2.

Back-to-Back Results Have Varied Just as Much as Home-Away Splits From Last Year

The average difference between the Game 1 scoring margin and the Game 2 scoring margin, across all the back-to-back games played at a home venue, is 12.7 points.

For example, Southern Illinois beat Evansville by 6 points on December 27th, but then lost to them by 12 points on December 28th. That would count as an 18-point swing in the two scoring margins.

That 12.7 number, then, is the average across all such back-to-back games we examined.

Let’s put that number in some context. In the previous season (2019-2020), across all conference games where the opponents played twice in the regular season (home and away), the averaging scoring margin difference between the two games was … 12.7. 

But those were games played at two different locations (and with fans), where you would expect home court advantage swings to naturally create some variation in the two outcomes. They were also games that were generally played several weeks apart, with lots of other opponents in between. That’s more time for injuries to happen for either team and for other factors to change how teams play.

Why, then, are we seeing similar variation in game results in these back-to-back games? It could be some evidence of how strongly motivation factors impact college basketball, where teams get another chance to play the same opponent right away. It could also be some evidence of teams being able to make immediate adjustments to what worked and did not work in Game 1.

Whatever the reason, the games have had a nearly 13-point swing in the outcomes when played under nearly identical conditions. That’s again a reminder of just how little one individual basketball game can tell us about the quality of the two teams, and how we should be careful making judgments from any one game, even a game between the very same opponents in what seem like the same conditions. 

 

Get access to all betting content, picks and models

Free subscription offers now available!

Subscribe now

Already a premium subscriber? Sign in for access