2026 College Football Analysis, Part 8: Season Bad Betting Advice

Over the past few weeks, I have taken you on a tour of the college football preseason landscape.

I reviewed the lessons of 2025, examined Michigan State's schedule and analyzed the races in the Big Ten, the rest of the Power Four and the Group of Six. Most recently, I projected the 12-team College Football Playoff.

Across the first seven installments, I shared odds generated by a simulation of the full college football season. In today's finale, we will see whether those numbers might help us win a little money.

That's right: It is time to kick off the 2026 edition of Dr. Green and White's Bad Betting Advice.

The Big Picture

My full-season simulation produces odds that can be compared with futures available at various online sportsbooks. The types of possible wagers include winning a division, reaching a conference championship game, winning a conference, making the College Football Playoff and winning the national championship.

Table 1 summarizes those odds for the consensus preseason top 25.

Table 1: Conference championship, playoff and national title odds for the consensus top 25 in 2026. Simulation probabilities are compared with DraftKings odds to estimate potential return on investment.

Table 1 includes each team's preseason ranking, strength of schedule, and strength-of-schedule ranking. It then lists the probability of each outcome from my simulation, followed by DraftKings odds as of Aug. 24, 2026. I converted the moneylines to implied probabilities for an easier comparison.

The right side shows the projected return on a $100 wager, assuming my simulation reflects the true probability. Bets with a positive projected return are shaded green.

As expected, national title odds track closely with the preseason rankings. No. 1 Ohio State has the best chance to win the championship at 13.7%, followed by No. 2 Oregon (11.5%), No. 3 Notre Dame (8.7%), No. 4 Texas (8.6%) and No. 5 Georgia (7.0%).

Playoff probabilities for the top 10 cluster near 50%, suggesting that roughly half of those highly ranked teams will make the field.

Some will fall short of expectations; others will absorb one loss too many. In my disruptive scenario, No. 6 Indiana (10-2), No. 8 Texas A&M (7-5) and No. 9 Oklahoma (7-5) all miss the field despite being three of the best nine teams in the country.

About 75% of these bets carry a negative projected ROI. That is no surprise: Sportsbooks build an edge into their prices, so the house wins on average.

Still, a few teams show a positive ROI in at least one of the three categories. Texas A&M, Oklahoma, BYU, Iowa and Clemson each post positive returns in at least two.

Suggested Conference and Division Champion Bets

Table 1 captures only a sample of the wagers generated by my simulation. I also ranked the most promising options by projected ROI in several categories.

Table 2 lists the 25 highest-ROI wagers on teams to win a conference or reach its championship game.

Table 2: Conference and division champion (or other conference championship game participant) wagers with the highest projected ROIs in 2026, based on a 100,000-run Monte Carlo simulation and DraftKings odds.

Most are high-risk, high-reward bets on teams with probabilities below 5%, but a few options stand out.

Several are dark-horse picks to reach a title game. I am intrigued by North Carolina (projected ROI: $168.34) and Boston College ($141.00) in the ACC; Minnesota ($175.50) and UCLA ($158.32) in the Big Ten; and Colorado ($109.99) in the Big 12.

A few Group of Six wagers farther down the table also caught my eye. UAB to reach the American Conference championship game (+12,000; projected ROI: $80.05) is compelling. So are Georgia Southern to reach (+1,300; $65.19) and win (+2,200; $48.21) the Sun Belt championship game.

Conference futures have performed well with this methodology. Last year's top 20 by projected ROI included Indiana to win the Big Ten (+3,500) and BYU to reach the Big 12 championship game (+700).

Most impressively, last year's top projected-ROI wager was Kennesaw State to win Conference USA at +40,000.

Even so, in 2026 the sum of the individual hit probabilities in Table 2 is only 0.64 — less than half of last year's total. My data therefore suggests only about a 50-50 chance that at least one of these wagers cashes.

Playoffs and National Title Bets

Table 1 includes selected national title returns. Table 3 expands the view to the 30 most promising wagers on teams to make the playoff or win it all.

Table 3: Playoff and national champion wagers with the highest projected ROIs in 2026, based on a 100,000-run Monte Carlo simulation and DraftKings odds.

As in Table 2, many of these are high-risk, high-reward picks. The sum of the hit probabilities is slightly better but still trails last year at 1.09, implying roughly one winner across the listed wagers.

The most interesting options are dark horses from slightly weaker conferences. Colorado (projected ROI: $212.31), Georgia Southern ($164.53), North Carolina ($158.06) and Colorado State ($133.47) all offer intriguing potential wager on the College Football Playoff.

I must also note that Michigan State to make the playoff at +15,000 carries a projected ROI of $227.97.

Win Total (Over/Under) Bets

Season win totals offer a lower-risk approach. Table 4 lists the 30 best over/under wagers, according to my model.

Table 4: : Season win total wagers with the highest projected ROIs in 2026, based on a 100,000-run Monte Carlo simulation and DraftKings odds.

The potential returns are smaller, but the hit probabilities are much higher than for the futures discussed above. The sum of those probabilities is 17.6, so a healthy share of these wagers should land.

The top plays are Texas Tech under 10.5 wins (projected ROI: $57.39), North Dakota State under 8.5 ($42.12) and Louisiana-Monroe under 3.5 ($39.27).

For the optimists, the best over is Colorado to top 4.5 regular-season wins (projected ROI: $29.34).

Michigan State ranks No. 26 on the list, with a projected ROI of $15.98 on over 4.5 wins at +130.

The model's expected win totals also align closely with DraftKings' adjusted totals. By "adjusted," I mean that I used the moneyline prices to estimate fractional win totals. Figure 1 shows the comparison.

Figure 1: DraftKings' adjusted win totals for all 138 FBS teams compared with expected regular-season win totals from my Monte Carlo simulation.

Using preseason rankings alone, the model predicts win totals with impressive accuracy. The standard deviation is 0.50 wins, and only five of 138 teams differ by more than one win. The farther a point sits from the line, the larger its projected ROI in Table 4.

Most Promising Big Ten Bets

For Big Ten fans wondering where the best potential returns lie, Table 5 provides the model's answer.

Table 5: Big Ten wagers with the highest projected ROIs in 2026, based on a 100,000-cyle Monte Carlo simulation and DraftKings odds.

The model sees potential value among teams near the bottom of the conference projections. Long-shot wagers on Rutgers, Purdue, Michigan State, Maryland and Minnesota to make the playoff, reach the Big Ten championship game or win the conference all carry strongly positive projected returns.

But the sum of the hit probabilities is just 0.59, again suggesting only about a 50-50 chance that at least one of these picks hits.

The higher-probability options include Minnesota, Nebraska and UCLA to make the playoff or reach the Big Ten championship game after a surprisingly strong season.

Bettor Beware

Before I wrap this article and the preseason series, one warning is in order: Although this piece centers on betting, I do not wager on sports. I find the math behind betting markets fascinating, but my interest ends there.

This analysis is for entertainment. I will track the picks during the season, but mostly because I find the results interesting.

I believe the probabilities are sound, but the projected returns depend on one key assumption: The preseason ratings from Phil Steele, Lindy's, Athlon, SP+ and ESPN must be slightly more accurate, on average, than the ratings embedded in the market. That may or may not prove true.

Serious bettors should treat this information as one input, not a complete strategy. If you already have a few wagers in mind and want a quantitative cross-check, the data may be useful.

Either way, another college football season is nearly here. Welcome back, college football. We missed you.