Betting totals when bad college basketball teams play

We earn a commission if you open a Bovada account through links on this page.

When two weak college basketball teams meet, adding up their scoring averages is a poor way to project the total, because those averages are skewed by pace and by lopsided games against much better opponents. Project these games from opponent-adjusted numbers instead. A common betting idea is to take the under when two bad teams meet. Test it before you rely on it, because an angle this well known may already be built into the totals.

Why raw averages mislead

Points per game mixes two things: how many possessions a team’s games have, and how many points it scores or allows on each one. A slow team can look like a good defensive team just because its games are short, and a fast team can look like a bad one.

Raw averages also depend on the schedule. A few lopsided games against much stronger opponents can pull a weak team’s season averages a long way, especially early in the season when it has played only a handful of games.

An example with made-up numbers

Team A has played 14 games, scoring 64 points a game and allowing 74. Four of those games were road losses to top teams, in which it scored 55 and allowed 88 on average. In its other 10 games it scored 67.6 and allowed 68.4.

Team A	All 14 games	Without the 4 mismatches
Points scored	64.0	67.6
Points allowed	74.0	68.4
Average game total	138.0	136.0

Look only at the defense and Team A seems 5.6 points better than its average. Its offense also looks 3.6 points better, though, so its typical game total drops by just 2 points. If you adjust one side of a team’s numbers, adjust the other side too.

Throwing games out is a crude fix. Opponent-adjusted stats do the job properly, by weighing every game against the quality of the team on the other side.

Project the game from adjusted numbers

Tempo-free stats sites such as KenPom publish adjusted offensive and defensive efficiency, in points per 100 possessions, along with an adjusted tempo. A simple way to combine them is to start from the average efficiency, then adjust it for how far the team’s offense and the opponent’s defense are from average.

Here’s a new made-up matchup, Team C against Team D, in a season where the average is 105 points per 100 possessions. Team C’s offense rates 96, nine points below average, and Team D’s defense allows 110, five points worse than average. That puts Team C at 105 − 9 + 5 = 101.

Team D’s offense rates 94 and Team C’s defense allows 108, so Team D comes out at 105 − 11 + 3 = 97.

Team	Its offense	Opponent’s defense	Expected per 100 possessions	At 66 possessions
Team C	96	110	101	66.7
Team D	94	108	97	64.0
Total				130.7

If both teams average about 66 possessions, the projection is 130.7. If the book’s total is 135.5, you’re almost 5 points under it. That’s a big enough gap to look into, starting with injuries and lineups.

Adding and subtracting like this is a simplification, but it’s close enough to tell you whether a total is in the right range. Our guide to betting college basketball totals covers pace and the rest of the projection.

The case for the under, and its limits

The usual argument is that bad teams are often bad mainly because they can’t score. When two of them meet, neither offense is good enough to take advantage of the other’s weak defense.

The adjusted method above already allows for weak offenses. For the under angle to add anything, these games would have to finish below even an adjusted projection, and below the closing total often enough to cover the vig.

How to test it yourself

1. Decide what “bad” means before you look at any results, for example both teams in the bottom 75 of an adjusted efficiency ranking.
2. Use each ranking as it stood on the day of the game. End-of-season rankings include results nobody knew when the bet was placed, which makes almost any angle look better than it was.
3. For each qualifying game, note your adjusted projection, the closing total and the final score, over at least two seasons.
4. Check whether these games finish below your projection on average. If they don’t, the under angle isn’t adding anything beyond what adjusted numbers already tell you.
5. Count unders and overs against the closing total, leaving out pushes, and compare the under rate with 52.38%, the break-even rate at -110.

A good-looking record can still come from luck. Our guide to short rest and college basketball scoring works through how big a record needs to be before luck stops being a likely explanation.

Other things to check in these games

• Lineup news. Weak teams get less coverage, so injuries and suspensions can be hard to find. The school’s athletics site and the box score from the team’s last game are good places to check.
• Early-season numbers. A few games tell you little, and last season’s figures may describe a different roster.
• Late fouling. A close game can still end with intentional fouls and free throws, which add points however poor the two teams are.
• First halves. The same raw-versus-adjusted problem applies to first-half totals, where the line is smaller and each mistake is a bigger share of it.
• Betting limits. Bovada says its limits vary by event, so check the maximum on a low-profile game before you plan a bet.

Our Bovada review covers how the site works, including what it means that it’s an offshore book.

Only bet what you’ve set aside to lose. If gambling has stopped being fun, call or text 1-800-MY-RESET (1-800-697-3738) for free, confidential help, 24/7, or chat online with the National Problem Gambling Helpline.

Sources

• KenPom: Help with team page (adjusted efficiency and tempo)
• Bovada help center: Betting limits