Projecting the ACC Tournament: Probabilities and Odds

The ACC Tournament opens in Brooklyn on Wednesday — Welcome to New York City, huckleberries! — and the league’s premier event has, at its core, one of the most watchable conference championship products this year:

• On a macro level, the ACC’s favorite — North Carolina — is not a slam-dunk guarantee to hoist gold on Saturday, allowing for potential anarchy to wrap itself around a tournament where a not insignificant number of very good teams have an opportunity to hang a championship banner; and
• On a micro level, a handful of programs are placed in pressure-filled situations against teams of importance that yield toss-up results, helping to add a layer to Selection Sunday’s opaqueness.

The idea that “Anything can happen!” is a chorus that tends to find its refrain most often at this stage of the season — it’s true throughout the year, but as bingeful consumption of college hoops rarely occurs until March, it tends to be heightened as NCAA bids start getting earned — and it may be especially true as the ACC plants its flag in New York for the first time. This is going to be insanely fun.

TITLE AND ROUND PROBABILITIES: IT’S THE PATH AND THE TEAM

Building a projection requires both an assignment of relative strength and understanding the path to which a team must traverse in order to come out of a tournament unscathed. This is why seeding is more than just a secondary concern: A painful road is going to create deflated win probabilities; the concept of “If they were good, they’d just win!” is borderline insane in that the concept of forward potential is influenced by the strength of potential opposition. This is why, despite the Tar Heels’ dominance, North Carolina is somewhere around a 2:1 to 3:1 favorite to win the ACC Tournament and a team like Virginia has the second-best odds to triumph despite being seeded sixth in the bracket.

To determine title and round advancement probabilities, I pulled up the Sagarin (Predictor) ratings and simulated the bracket 20,000 times using those ratings as the basis for win expectation.* Not surprisingly, Tuesday teams all have less than a 1% chance of winning the title — hell, only Wake Forest has more than a 20% chance of advancing to the quarterfinals — and those that start on Thursday are at a decided advantage to play on the weekend. Here’s the full picture:

(Click here to embiggen.)

Some brief thoughts on the simulation:

• Wake Forest is dangerously situated in the bracket. The opportunity to pound Boston College and face an overseeded Virginia Tech team in the second round gives the Deacons and incredible advantage to progress to the quarterfinals. If Wake were to advance to meet Florida State, it would face a Seminoles team ranked fifth in the league according to the Sagarin (Predictor) ratings despite earning a second seed. It would not be shocking to see the Deamon Decons push through this thing a bit despite starting at the Tournament’s sunrise. It is a path that is conducive to progression compared to what Syracuse and Miami need to cover, two teams seeded immediately ahead of Wake Forest. Directly and simply: Wake Forest has an amazing draw, an underseeded team that has a preferential path to success.
• If you’re basing Syracuse’s NCAA Tournament participation on needing two wins at Barclays, there’s isn’t a ton of good news in that simulation. Facing a toss-up game against Miami and an uphill haul against North Carolina, the Orange would need to overachieve to hit that mark. The Hurricanes aren’t imposing, but the waiting Tar Heels is what makes a potential rush incredibly difficult.
• The top half of the bracket is stronger than the bottom half of the bracket, and the relative weakness of the bottom portion of the bracket increases the overall odds for those teams despite being weaker than those sitting in the upper portion. In function, the winner is more likely to come out of the top cut than the bottom but there is a higher likelihood of seed upsets in the lower half because of how the standings shook out.
• If you were to utilize Sagarin (Recent) ratings, most odds stay about the same from a title likelihood standpoint. Virginia’s probabilities are a little lower, Notre Dame’s are a little higher.

ODDS VALUE: RETIRE AFTER THE WEEKEND

What’s the point of carefully managing retirement assets in a 401(k) when you can just bet all of your money on college basketball games and retire before April? Screw those idiots at Vanguard — you know hoops and hoops is going to make you a billionaire because there’s is no way that Vegas is going to beat you.

OddsShark has issued odds for the ACC Tournament, and while none of them are exceptionally tasty based on the Sagarin (Predictor) simulation, there are some reasonable investments in there.

AVOID

• Do not touch any of the Tuesday teams. Even Wake Forest, which is +4000, is giving away value based on the team’s projected win probabilities (OddsShark is saying that Wake is 40:1 and the team is closer to 86:1). These are awful bets.
• Notre Dame (+1200): I’d just want more; the bracket is useful for the Irish, but the payout is shorter than the Sagarin (Predictor) odds. Too much money is left on the table.
• Duke (+450): Same deal as Notre Dame -- payout is shorter than true odds. The idea is to maximize the board, and Duke -- while with decent title probabilities — isn’t going to give you value on the payout because Vegas is shorting the odds.
• Virginia Tech (+4000): Payout is tempting, but the Hokies having only toss-up probabilities to get to the quarterfinals. You’re flushing cash down the toilet without the necessary incentive.

CONSIDER

• Virginia (+400): The bottom half of the bracket is weaker than the top and the Cavs are getting about what they’re worth (again, nobody is getting more). The 20% title probability is true and the path is better than that of Notre Dame or Florida State on the bottom side of the field.
• Louisville (+400): Odds are fair but payout isn’t spectacular. You’re taking the second best probability on the board with the second best payout. This isn’t a money machine but it’s pretty safe with a related return.

TAKE

• Florida State (+650): Fair odds, like Virginia and Louisville, but higher payout than what the Cavaliers and Cards are getting. There’s some value here — the path to the final isn’t neck-breaking and in the final against North Carolina or Louisville the Seminoles would be no worse than a four-point ‘dog based on the Sagarin (Predictor) ratings.

TAKEAWAY

Look: None of this imparts what will happen; it only discuss the likelihood of scenarios coming to fruition. A single-elimination tournament is both the best and the worst: Only the result matters, and that can burn a favorite as much as it benefits an underdog (there are no 100-game iteration); if you replayed the tournament right after it finished you could have drastically different results (nothing is static). But, there are still favorites and long shots in this thing, and it’s useful to see what a simulation looks like in order to set at least a manageable baseline of expectations.

* Note: Ken Pomeroy creates title probabilities each year using his ratings. His probabilities differ slightly from those utilized in the Sagarin (Predictor) model. Both, however, project similar results.