How to Use College Basketball Betting Splits

Over the last year or so, I have habitually studied the DraftKings Betting Splits data on VSiN for several different sports. I have tried to keep VSiN readers up to speed on the strategies and concepts of using the data. I’ve done this through various articles and as part of our daily/weekly Analytics Reports. Until now, however, I hadn’t been able to analyze the data for college basketball, as before the 2023-24 season, we hadn’t collected enough results to make it worth studying. However, with the football season now in the rearview mirror and some valuable analytics time once again available, I made it my first order of business to attack the DK college basketball betting splits. I’m here to share what I found, and believe me when I say that what I have found is worth the read for any level of college basketball bettor.

To summarize the season findings in games up through Sunday, 2/11, the data sample contained 4,086 games, and the majority of bettors are losing money in every major category or essentially have a negative return on investment (ROI). These are the current results:

  • Majority handle on point spreads: 1977-1970 ATS (50.1%), -190 UNITS, ROI -4.8%
  • Majority number of bets on point spreads: 1959-1976 ATS (49.8%), -214.6 UNITS, ROI -5.5%
  • Majority handle on moneylines: 2718-1095 (71.3%). -364.57 UNITS, ROI -9.6%
  • Majority number of bets on moneylines: 2768-1056 (72.5%). -343.84 UNITS, ROI -9%
  • Majority handle on totals: 2000-1988 (50.2%). -186.8 UNITS, ROI -4.7%
  • Majority number of bets on totals: 1995-1970 (50.3%), -172 UNITS, ROI -4.3%

If you’re wondering why the results don’t add up to 4,086 games, it’s because 50/50 handle or bets splits are counted as null, and there were several ATS ties already in the season.

These numbers are collectively a great indication of why the “house always wins.” From a generic sense, this is a lot of games, and at -4.7% or worse ROI, bettors are slowly drained of their bankrolls.

This is the first study of this kind for any sports in which all major categories showed money losses. Does that mean there is never a reason in which bettors utilizing this data should follow the majorities? Of course not, and I will share some specific instances later in the systems in which the public bettors have done well. However, I would say that just blindly following a fade or follow strategy of the majorities won’t get you very far.

Why College Basketball Betting Splits Are Important

Why is this Betting Splits information so valuable? Well, most specifically, it lets us see which games are the most publicly supported each day or week. The common belief is that the “betting public” does not win over the long haul, and with the amount of money DK spends on acquiring customers, you have to assume that most of the bettors that play there would be considered “public”. In other words, the collective opinions of this group of recreational bettors are right out on display for all of us to analyze and take advantage of.

Most of the theories I believed about public bettors losing more often than not came to fruition, particularly for the NFL, a sport heavily wagered by public bettors. However, the NBA has proven to be a different story and my best explanation is that bettors of the NBA aren’t as novice of a group as the NFL. You can’t be a “weekend warrior” and bet the NBA. My theory about the college basketball findings is that the public struggles simply because there are so many games and teams to get a handle on that they don’t become experts in any of them.

I should remind you of the two common betting generalities for recreational players who wager at places like DraftKings. That is the majority bet favorites on point spread and Overs on totals. The numbers are staggering. According to the final ticket/handle so far this college basketball season for games dated through Sunday, 2/11, the majority of wagers were on favorites in at least 84% of the games and on Overs for totals in 57%. The ATS numbers are similar to other sports. To their credit, college basketball bettors aren’t nearly as infatuated with the Over on totals as football bettors.

As a little background on this exercise, I analyze the records of the “majority bettors” at DraftKings and the side/total they preferred for games. I also broke it down further, sometimes by the percentage of the majority and by some other variables. There are two fields of analysis for each wagering option that DraftKings shares the data for both sides and totals: 1) handle 2) number of bets. These can produce varied results, but I generally subscribe to the theory that the total handle is a little less “public” than a total number of bets.

I always like to use the overall data for developing systems for using these betting splits breakdowns for the continuing 2023-24 season. In this particular article, I am going to refer to the findings as “concepts” more than systems since very few of the angles cross the 55% (or less than 45%) threshold. Plus. I am trying to communicate some general philosophies you might want to employ the rest of the way rather than focusing on specific things that have been up & down for the most part.

DK College Basketball Betting Splits Concept #1:
Majority bettors for both the handle and bets on point spreads have been much better in non-conference action than conference games.
On the handle, majority bettors have gone 1015-947 ATS (51.7%) in non-conference action as compared to 962-1023 ATS (48.5%) in conference games. On bets, the non-conference numbers are 995-960 ATS (50.9%) versus 964-1016 ATS (48.7%) for conference play. Now these aren’t earth-shattering numbers for success either way, but a 3% difference in win rate can mean the difference in profit or not to a bettor.

DK College Basketball Betting Splits Concept #2:
Majority bettors for both the handle and bets on moneylines have been much better in conference action than non-conference games.
Exactly the opposite of what I just noted for point spreads, on the handle, majority bettors have gone 1391-624 (69%) for -76.55 units and an ROI of -3.8% in conference action as compared to 1327-471 (73.8%) for -288.02 units and an ROI of -16% in non-conference games. On bets, the ROI difference is even greater, 1413-604 (70%) & -55.49 units for ROI of -2.8% in conference versus 1355-452 (75%) & -288.35 units for ROI of -16% in N/C play. These approximate 13% loss margins are significant, and there were some huge upset losses in earlier season non-conference action that negatively impacted returns.

DK College Basketball Betting Splits Concept #3:
Majority moneyline bettors have done very well in neutral court games for both the handle and bets as compared to contests with hosts teams.
Moneyline handle majorities have been profitable thus far in neutral court games this season, as they have gone 262-129 (67%) for +10.04 units and an ROI of 3.5% while they have lost -374.61 units at an ROI of -10.9% in all other games. The breakdown on bets has been -5.6 units in neutral games as compared to -338.24 in all others, a similar but lesser difference. Those who know my past work on DK betting splits systems for other sports know I wouldn’t rave about only a 3.5% return, but this is the first area I’ve introduced that shows the majority of bettors winning. This could benefit readers in March when we see the return of a lot of neutral game action.

DK College Basketball Betting Splits Concept #4: Majority handle bettors have fared reasonably well on totals in conference games as compared to non-conference tilts.
Majority totals betting groups in terms of handle have nearly squeaked out a profit in 2013 games of in-conference play. They are collectively 1053-960 (52.3%) for -3 units over the course of the season so far as compared to 947-1028 (47.9%) & -183.8 units in non-conference action. This is a margin of about 9.3% on ROI. The figures on bets aren’t nearly as diverse, with both conference and non-conference records slightly over 50%.

DK College Basketball Betting Splits Concept #5:
Beware of any teams that get backed by 95% of more of moneyline majority handle and bets. These have produced massive losses for bettors.
Yes, it’s true. Some college basketball teams get backed by 95% or more of the handle and bets volume at DraftKings. So far this season, when it has happened on the handle, this majority group has gone 699-170 (80.4%) but has still lost -165.11 units for an ROI of -19%. On bets, the results are even worse: 1155-247 (82.4%) for -211.75 units and an ROI of -19.4%. You have to understand that for the most part, these 95% teams are usually heavy favorites, and every loss proves disastrous to the bankroll.

DK College Basketball Betting Splits Concept #6:
In games with point spreads of 20 points or higher, majority handle and bets groups have produced profits on point spreads so far this season.
Public bettors have had a good read on these suspected blowout games, with majority handle bettors going 160-135 ATS (54.2%) for +11.5 units of profit and majority bets groups going 160-132 (54.8%) for +14.8 units. Keep in mind, though, that in about 83% of these games, the betting public, of course, backed the chalk, although there wasn’t a noticeable performance difference between those backing favorites and those backing underdogs in such games.

DK College Basketball Betting Splits Concept #7:
In games with point spread moves opposite of the handle majority opinion, bettors have so far beaten the oddsmakers in college hoops this season.
There have been 121 games so far this season as of Monday, 2/12, in which the majority of handle bettors have been on the opposite side of a line move. Public bettors have won at a 70-51 ATS (57.9%) rate in such games, picking up +13.9 units of profit on an ROI of 11.5%. This situation can most simply be described as when Team A opens as a small favorite but becomes a bigger favorite by tip-off despite the majority handle being on Team B. In essence, this sharp group of bettors hasn’t fallen for the picture that DK oddsmakers are trying to paint.

DK College Basketball Betting Splits Concept #8:
In college basketball games this season where both teams were ranked, majority groups on the handle and bets have been sharp on almost every metric.
Through this most recent Sunday, there were 52 games this season between two ranked opponents. Majority bettors have dominated these games in every category. These are the records:

-Majority ATS handle group: 31-17 ATS (64.6%), +12.3 units
-Majority moneyline handle group: 42-10 (80.8%), +24.3 units, ROI +46.7%
-Majority ATS bets group: 32-16 ATS (66.7%), +14.4 units
-Majority moneyline bets group: 41-11 (78.8%), +22.2 units, ROI +42.7%
-Majority totals bets group: 30-21 (58.8%), +6.9 units

It’s somewhat encouraging to see that public bettors have fared well in the season’s biggest games. It will be an interesting pattern to follow in the postseason.

DK College Basketball Betting Splits Concept #9:
In college basketball games this season where only one team was ranked, majority moneyline betting groups at DK have tanked.
As compared to concept #8 above, featuring two ranked teams, the majority of moneyline betting groups have performed miserably in games where a ranked team was taking on an unranked team. These are the records for those situations:

-Majority moneyline handle group: 287-106 (73%), -201.42 units, ROI -51.3%
-Majority moneyline bets group: 286-105 (73.1%), -198.17 units, ROI -50.7%

The worst portion of the moneyline handle group was when majorities backed ranked road teams, as that group has gone just 71-50 (58.7%), losing -96.42 units for an ROI of -79.7%! It’s safe to say that readers of this article should not get behind these bettors, and alternatively, wager the opposite way on the home underdog money lines until something changes.

The betting splits on are essentially running live and will change in the days and hours before game time. If you adopt any of these strategies this season, visit for the latest numbers.

The data we get directly from DraftKings Sportsbook is updated every five minutes. We ping the DK database and see what data has changed and then indicate the change with an arrow next to the percentage.

Also, if you want to see the opening line and the line history, click on the visiting team’s logo. The updates are timestamped. When a game is over – you will get a summary that includes the final score, total result, season Over/Under record for each team, and you can compare the result to what Steve Makinen’s estimated score was.

Unlimited access to betting splits is available to VSiN Pro subscribers.

Previous articleEPL Best Bets for the February 17th-19th matches
Next articleSuper Bowl LIX Odds
As one of the original founders of StatFox, Steve Makinen has been in the business of sports betting and data analysis for almost 25 years now. In his time in the industry, Steve has worked in a variety of capacities on both sides of the betting counter, from his early days of developing the StatFox business, to almost a decade of oddsmaking consulting for one of the world's leading sportsbooks, to his last seven years as Point Spread Weekly and Analytics Director with VSiN. Steve has always believed that number crunching and handicapping through foundational trends and systems is the secret to success and he shares this data with VSiN readers on a daily basis for all of the major sports.