The (post)rat king: the rigorizing analytic vs the sensitive sharp

by RIPDCB

When analytics broke into professional sports, it brought with it a major ontological shift: the rigorizing pipeline. This is not to say that decision makers were operating purely on vibes, but the introduction of a data-centered approach to roster construction and game planning allowed GMs and head coaches to find value in underexplored areas. It gave them the ability to conceptualize and then formalize different ways of looking at production against the noisy, outcome-driven received wisdom of their respective fields.

The 2002 Oakland Athletics might be the most classic example of a team who let data take the reigns. Their analytics department made strides in approaching value from different angles, with their reassessment of the walk’s offensive potential. Baseball is all about scoring runs and preventing runs. Runs are, for the most part, scored through hits, and the more men you have on base when you get a hit means that you’ll drive in more runs on those hits. In the pre-Moneyball era, a hit (when a player puts the ball in play and makes it on base safely) was seen as definitively more valuable than walks (when a pitcher gives a batter a base on balls), but the data showed that whether you took a base on a single or a walk actually made no difference—what mattered was getting on base. Batters who were good at walking were significantly cheaper to acquire than were players who were good at hitting. The 2002 Oakland Athletics took this insight (among others) to build a team with the league’s lowest payroll that would go on to win an, at the time, record 22 straight games.

Given the roaring success of a data-centric approach in shaping professional sports, it was only natural that some would want to apply analytics to sports gambling. The stock market can be inefficient, so why wouldn’t the betting markets be? The explosion of mobile sports books and legalized gambling across the US only created more opportunities for people to make money, with different companies competing with each other to offer gamblers the best prices and promotions to rope them in.

The nerds had to slay the aging dragon of managerial conservatism in order to break into front offices, but they’ve had much stiffer competition in gambling spaces: the Sharp. Sharps are professional gamblers. They are winners, which few are, at least with any consistency. As such, they move markets. Their opinions are so valued that the biggest sportsbooks will often offer their sharpest and most privileged clients early access to lines (although with low limits), let them take a side, and then adjust the line accordingly before releasing it to the public. They are incredibly secretive in an era of gambling influencers flooding every corner of social media, shoving winning slips constantly down your throat. They also are incredibly exclusive at a time when there is no shortage of services offering access to picks and data (their exclusivity is also leant a certain gravity, at least for the older sharps, because of their past associations with mafiosi when the wise guys ran all the books and the casinos).

But sharps can also be traditional. They value truisms, sometimes at the expense of finer grained analyses; they trust their intuitions even if they’re not grounded in data; they’ve been known to bet historical trends that don’t have causal grounds. They are also painfully superstitious. But they keep winning, and gambling is nothing if not a results based business.

There’s a little bit of a rat vs. post-rat dynamic that exists between analytics heads and sharps. The former believe that the game can be effectively represented through numbers and models. Quantification is a matter of when, not if. They embrace the burden of proof and challenge others to match the thoroughness of their calculations, and if you can’t then you’re wrong. The intensity of their desire to conceptualize and quantify all aspects of the game is admirable, but it comes at the costs of certain intuitions that gamblers and sports viewers alike have, intuitions that while not quantifiable nonetheless feel true (momentum with respect to game flow is a good example of an intuition that can’t be formalized but can be felt, and felt tangibly enough that it shouldn’t be equated with superstition).

While an analytics approach to gambling has had some success—mostly with player prop bets—they haven’t been able to beat the sharps when it comes to the most public markets: point spreads, mainlines (i.e. the odds on an outright winner), and totals. For the sports betting marketplace, exposure and efficiency share a one-to-one correlation. The more exposure a given market gets, the more efficient it’ll be, and the most efficient markets are the ones that are not just bet the most but are also bet by the most people. American football side betting, for example, is notoriously the toughest market to beat, and somehow sharps still manage to find ways to be profitable.

How though? They’ll never tell, and that’s because sharps operate anti-inductively. They have no incentive to give their secrets away. While some sharps become public personalities and give rare interviews or write glamorous books in their older ages, the only significant way to tell what side a sharp is on is to read the market—watch the way it moves, the splits between tickets and money (a side with few tickets but lots of money tends to be the sharp side, since sharps make fewer bets for bigger totals while public bettors tend to make more bets for less money), look out for confusing lines (such as the perceived-to-be better team being an underdog), etc. Trace these tells across the season of a given sport, or even across different sports, and you’ll usually find the biggest discrepancies between sharp sides and public sides tend to come up as defense vs. offense, traditional vs. unconventional, boring team vs. the shiny new toy. While they might subscribe to certain truisms (e.g. ‘defense wins championships’), they’ll act as more of a guide than an instruction, a historical fact that should be thought through, neither accepted at fave value or dismissed entirely. They’re also often concerned with the phenomenology of playing the game itself, all the unquantifiable and unverifiable aspects of being a person that bleed into the on-field product (i.e. a team’s morale at particularly low or high point of the season). They’re sensitive to the market’s ability to correct: they know to buy low and sell high, fade the hot hand at the right moment and back the team that’s better than their record implies. Ultimately, sharps operate post-rattishly in that they’re intuitive and flexible, at least up until a point.

So a rough dialectic can be constructed between the rigorizing analytic and the sensitive sharp. At the end of the day the two sides present a nice complement to each other. In my experience, the analytic approach finds some neat edges in underexposed markets, but analytic discourse unnecessarily positions conventional gambling wisdom as its enemy, and in doing so shortens its potential to incorporate the sharp’s feel for both the game and the market. The sharp does ultimately hold an upper hand, given that he operates anti-inductively, hiding his moves while the analytic must hold himself to the highest standards of transparency. He must show that the world can be mapped by numbers more precisely than the betting markets. The sharp doesn’t have to prove anything, though. He can just keep counting his cash.