Odds and Ends: Week 2

Photo: Imagn Images

A good sports betting column should be backed by a profitable gambler with a proven track record. It should offer picks generated by a sophisticated and conceptually sound model. Most importantly, it should treat the subject with the seriousness it warrants.

This is not that column.

Instead, this will be an off-beat look at the sports betting industry-- why Vegas keeps winning, why gambling advice is almost certainly not worth the money, and the structural reasons why even if a bettor were profitable, anything they wrote would be unlikely to make their readers net profitable, too.

While we're at it, we'll discuss ways to minimize Vegas' edge and make recreational betting more fun, explain how to gain an advantage in your office pick pools, preview games through an offbeat lens (with picks guaranteed to be no worse than chance), and tackle various other Odds and Ends along the way.

Are We Over the Unders?

Returning readers will know that this column mostly focuses on picks against the spread, but there's another column in our table of lines-- the Over/Under (or O/U). For this bet, the sports book lists a point total, and you bet whether the teams will combine to go over or under that total in the game (regardless of which team scores the points). 

Teams can hit the over if one team absolutely steamrolls the other (like with New Orleans' 47-10 victory over Carolina last week). They can hit the over if the game turns into an unexpected shootout (like with the Eagles' 34-29 victory over the Packers).

Likewise, the under can hit because one team shut down another far more than expected (as in Minnesota's 28-6 throttling of the Giants) or because the game devolves into a defensive slugfest (New England's 16-10 win over the Bengals).

But by and large, an over/under bet is a bet on whether the game is more or less likely to become a shootout than expected.

While most bets are 50/50 propositions, bets on the under historically win ever-so-slightly more often than overs. Not often enough to be profitable (bettors need to win about 53% of the time to overcome the vigorish), but enough to be slightly less unprofitable.

If Vegas is so good, why have they left such a glaring inefficiency? Because people love betting overs. Scoring is exciting, and rooting for more scoring is more fun than rooting for less scoring. Plus, "over" bets tend to remain "live" (or winnable) for longer. You might have bet "over 44.5" and the game might only have 31 points late in the fourth quarter, but teams could always just score touchdowns on the next two drives, and there's usually at least a remote possibility of overtime to juice the total.

(My favorite example: in 2013, the Vikings and Ravens played a game with an o/u of 41.5. As they approached the 2-minute warning in the 4th quarter, the teams had combined for just 19 points. They scored 36 points in the final 2:05 of the game and beat the over by nearly two touchdowns.)

By contrast, since points never come off the board once scored, "under" bets can be lost in the first half, leaving nothing to root for in the second. If part of the fun of gambling is buying the fantasy of winning, your money goes further on overs.

Given that Vegas exists to maximize its profit, it makes sense that it'd shade things a hair so that the more popular bet loses slightly more often. Not quite enough for the "sharp money" to come in and take advantage, but enough to skim a bit extra off of recreational bettors.

In Week 7 of the last two seasons, though, I've noted that Unders were winning at a rate much higher than the expected 51-52%. In 2022, they opened the year 57-36-1, a 60.6% win rate. In 2023 they started even better, going 58-34-1, a 62.9% rate-- if you bet an equal amount on every game, that would have been enough to see a 20% profit through six weeks even after accounting for the typical vig.

In both seasons I asked: given the structural advantage for "under" bets and the scorching start, could we turn a profit by just mass-betting the under in every game? My hypothesis was no: the toughest feature of Vegas is that it is self-correcting. In the short term, it can be surprised by shifts in the game of football (such as the rise of Cover 2 defenses designed to take away deep passes and inhibit scoring), but it quickly learns and adapts, lowering the over/under lines it offers to the new equilibrium.

In 2022, unders went 84-74-3 after the column, a 53.1% win rate that was technically profitable! (If you'd wagered $10 on every under for the rest of the season, you'd have walked away with an extra $23.63.) In 2023, they went 80-79-4 down the stretch-- a near-perfect 50% win rate with Vegas skimming the vig off the top. Someone who bet the under on every game would have lost $62.73.

In total, mass-betting unders after both predictions would have cost you $39.10 over two seasons. Which, as far as Vegas is concerned, isn't bad at all! 

Unders are not opening the 2024 season with the same kind of heat; they went 7-8-1 in Week 1 (at least according to the lines I posted in my last column). But maybe if we jump on the train a little bit early we can get some of that sweet early-season bank.

I mean, probably not-- believing Vegas is good at its job means I suspect every bet is going to be a money-loser in the long run. But I'm open to being proven wrong, and even if I'm right, this bet is probably a little less of a money-loser than most others. So let's give it a shot.

Last year we tracked how we'd have done using two betting strategies. The first was an "equal bet" strategy where we just put $10 on the under in every game. The second was a "snowball" strategy-- we started with a $160 bankroll and rolled it over every week, assigning an equal share of our remaining bankroll to every game that week. If we doubled our money to $320, we'd bet $20 on all 16 games. If our bankroll fell to $80, we'd only bet $5 per game.

The "snowball" strategy is good for one thing: it caps your maximum losses. If you start with $160, the most you can possibly lose is $160. It also gives you much higher possible gains-- if you get on a real hot streak you can find yourself betting $30 or $40 per game, which triples or quadruples your potential payout for the week.

But in terms of maximizing expected profit, it's not necessarily optimal (we'll get to why later in the season), so it's fun to compare it to a simple "fixed bet" strategy in real time to see how the results differ.

Lines I'm Seeing