When Peko Hosoi competes against her 12-year-old nieces in fantasy football, she abides by a firm rule: never play for money. In a family of self-proclaimed "excellent smack talkers," bragging rights alone constitute a worthy prize. And she’d rather not veer too close to fraught ethical territory; as the founder of the MIT Sports Lab in Boston, she has firsthand access to a trove of data about the results of games played through the daily fantasy provider FanDuel.
Based on the insights Hosoi has gleaned from that data, even the staunchest fantasy football players among us - the diehards who mine draft rankings to find potential steals and monitor the waiver wire all week to optimize their shot at dominating Sunday - will realize they can't prepare for every twist of fate.
In U.S. and Canadian daily fantasy sports, football is the toughest game in which to make an easy buck, given how influential random chance can be in determining the outcomes of its matchups. Deep knowledge of NFL depth charts, or proficient navigation of the weekly transactional churn, is often not sufficient to guarantee victory, at least compared to the experience of expert players in fantasy baseball, basketball, and hockey.
This quirk is a major takeaway from a 2018 research paper co-authored by Hosoi. She and her colleagues devised a metric to quantify the extent to which luck and skill decide winners and losers in different daily fantasy sports - a mathematical answer to the initial policy question of whether these games were skill-based enough to convince state legislatures that they should be legalized.
Hosoi's findings seem particularly relevant in October, the only time of year that the MLB, NBA, NFL, and NHL calendars - and, with them, each league's parallel fantasy universe - all intersect. For the millions of fans across those sports who manage a team of their own, one glance at the above spectrum should be enough to process another lesson.
"If you're somebody who's trying to make a bunch of money out of this," Hosoi said, "play fantasy basketball."
In order to situate each fantasy sport on the spectrum, Hosoi and her fellow researchers set out a few years ago to study the results of salary-capped baseball, basketball, football, and hockey games played on FanDuel during the 2013-14 and 2014-15 seasons. If skill tended to trump luck in those competitions, a straightforward comparison - a player's average performance in the first half of a season versus their average performance in the second half - would illuminate its impact.
"One of the hallmarks of skill is persistence," Hosoi said. If a game is mostly based on skill, a player who wins more often than not remains likely to keep winning all season. If you lose in a blowout every week, it would be reasonable to expect that dispiriting trend to continue.
"Whereas if I'm flipping coins," Hosoi said, referencing an activity that appears on the spectrum for context, "and I happen to do well (at) flipping coins in February, that's in no way predictive of what I'm going to do in March."
By delving into win splits, Hosoi and her team found that victory in all four fantasy sports - and in their real-world versions - depends mainly on skill, though to varying degrees. The actual NBA, where teams play 82 games and hoist nearly a hundred shots per night, rewards aptitude more than any other competition. The actual NHL, where teams play 82 games but generate far fewer quality scoring chances, hews closer to the midpoint of the scale - closer to flipping a coin.
Devoted hockey fans already know that a single fortunate bounce can mean a lot on any given night. The spectrum's innovation is showcasing the rapport between the balance of luck and skill in a real sport and the balance of luck and skill in most corresponding fantasy variants. In both types of basketball, talent is overwhelmingly likely to prevail, since it's a more predictable game. Hockey, in whichever form, is comparatively chaotic.
Two points initially struck Hosoi as odd, she said: "One of them I can explain and one of them I can't."
The first of those surprises is that fantasy baseball scored higher on the skill spectrum than the sport itself. Her theory? The proliferation of advanced stats in baseball has created a gulf between the best fantasy players and the rest of the field that exceeds the distance between MLB championship contenders and cellar-dwellers. Equipped with their detailed spreadsheets, these fantasy managers are better positioned to exploit a skill gap than they would be in other games.
It's harder to explain the NFL's close proximity to the pure-skill end of the spectrum. The gap between luck's hold on fantasy football and on actual football is greater than in other sports, even though logic suggests that random chance should be consequential in a league whose teams play only 16 games and don't score all that much.
"I'm going to speculate wildly," Hosoi said, putting forth two ideas as to why skill carries the day on the field. Maybe, like in fantasy baseball, there is a vast talent imbalance between NFL rosters, and it doesn't take many games for the proper pecking order to take hold. Maybe, as one of her students has hypothesized, it doesn't matter that scoring plays are infrequent because every down is a scoring opportunity, increasing the likelihood that the best teams will win out over time.
Individual skill doesn't carry the same sway in fantasy football, which many players who compete in leagues against their friends may intuitively understand. Despite that, football is the favorite fantasy sport of two-thirds of the 59.3 million people who play all manner of such games in the U.S. and Canada, according to 2017 data from the Fantasy Sports Trade Association.
To Hosoi, fantasy football's popularity goes hand in hand with its accessibility. When even diehard players need a little intangible help to win, odds are better that parity will reign.
"It's a game where a lot of people can participate regardless of their ability," she said. "You don't want to walk into a game where you're just going to get trounced every time. The way you make that fun for that population is you have to add an element of chance."
Nick Faris is a features writer at theScore.
Copyright © 2019 Score Media Ventures Inc. All rights reserved. Certain content reproduced under license.