The Sabres Did the Tank Perfectly. The Lottery Did Not Care.

Part I of this series argued that the National Football League’s draft is a perfect Pyrrhic-victory machine because the rules are deterministic: the worst team picks first. The 2020 Jets won two meaningless games and surrendered Trevor Lawrence because the math is that simple — one win, one slot, no luck involved. The NHL has the same incentive at the bottom of its standings, but it has built a second machine on top of it. It is called a draft lottery, and what it does, formally, is randomize the prize. What it does practically is reintroduce fortuna into a sport that had spent six months pretending to be about virtus.

This is, the Romans would have said, the natural condition of human striving. Sallust’s line was that fortuna in omni re dominatur — Fortune is mistress of all things — and he meant it as a warning to anyone planning a campaign. Plutarch would later record the dictator Sulla’s habit of attributing his victories to luck rather than his own ability and adopting Felix, the Lucky, as a public surname. The ancient world had a fully developed theology of how a sound plan could be overturned by something outside it. The Buffalo Sabres rebuilt that theology in modern dress, between the months of October and April, and lost the lottery anyway.

The Race to the Bottom, and the Team That Won It

The 2014–2015 NHL season was, from its earliest months, a race that everyone watching could name. Two prospects were available the following spring: Connor McDavid of the Erie Otters, the consensus best junior in twenty years, and Jack Eichel of Boston University, who would have been the number-one overall pick of any other draft year in recent memory. Three teams — the Sabres, the Arizona Coyotes, and the Edmonton Oilers — spent the season conducting what the league’s reporters described, charitably, as a deliberate rebuild and, less charitably, as the worst pretense of competition that the modern NHL had seen.

Buffalo won that race. They finished 30th and last with 23 wins, 51 losses, and 54 standings points. They traded their veteran center Cody Hodgson and moved their captain. They started a backup goaltender for stretches that no team had a sporting reason to. Arizona finished 29th. Edmonton finished 28th, which was an improvement on its prior four seasons in the basement and which, by any reasonable expected-value calculation, should have removed it from the McDavid sweepstakes. None of this turned out to matter.

April 18, 2015: Fortuna’s Day

The 2015 NHL Draft Lottery was held on the evening of April 18, 2015. Under the pre-2016 rules, only the first overall pick was lotteried, and the odds — weighted in favor of the worst team but capped at one-in-five — ran from 20.0 percent at the bottom to 1.0 percent at the top of the eligible pool. The chart below shows the full distribution.

Edmonton, the 28th-place team, won. The Oilers fans who appeared in the lottery broadcast wore the same look of giddy disbelief that won-the-lottery fans always wear. The Buffalo fans wore something else. Both reactions were correct. The lottery had done what lotteries do, which is randomize, and Edmonton had landed Connor McDavid — the team’s fourth first overall pick in six years, following Taylor Hall in 2010, Ryan Nugent-Hopkins in 2011, and Nail Yakupov in 2012. The Oilers had, by then, become a kind of statistical anomaly: a franchise that managed to be terrible enough to pick first, year after year, without ever ascending out of the basement. Their previous three top picks had produced a 28-place finish in 2015. None of that history, of course, had any bearing on the next bounce of the lottery balls. That is what makes it a lottery.

The Consolation, and What Happened to It

Buffalo, holding the second pick, was guaranteed Jack Eichel. By any reasonable standard, that should have been a Pyrrhic victory of the gentlest kind — you tanked, you missed the lottery, and you still came home with a franchise center. Eichel was the second-best player in his draft class by a margin that would have made him the easy first pick in any other year, and the Sabres reasonably expected, in 2015, to be building around him for a decade.

He played for them for six seasons. He never won a playoff game in Buffalo, because Buffalo never reached the playoffs. He was named captain, posted multiple seasons of point-per-game offense, and then, in 2021, after the team and the player disagreed about how to treat a herniated cervical disc — Buffalo preferred a fusion surgery, Eichel preferred an artificial-disc replacement that no NHL player had had before — he forced his way out. On November 4, 2021, the Sabres traded him to the Vegas Golden Knights for Alex Tuch, Peyton Krebs, and two early-round picks. Eichel had his preferred surgery, recovered, made his Vegas debut on February 16, 2022, and in the spring of 2023 hoisted the Stanley Cup as the Knights’ first-line center.

What the Romans Knew

The deeper statistical point underneath this story is the one the ancient world was good at and the modern sports column is mostly bad at: tanking is not a transaction. It is a bet. Buffalo executed the tank perfectly — the highest-skill move available to a non-competitive roster — and was rewarded with a probability distribution, not a player. Twenty percent at the first pick, eighty percent at the second. The Sabres landed at the eighty. The Oilers landed at one of the eleven and a half. Neither result was a verdict on either front office’s preparation; the bounce of the balls did not care.

This is where the smooth-curve fallacy from Part I changes shape. In the NFL, the cost of a meaningless win is deterministic — one slot, no chance. In the NHL (and as Part III will show, the NBA), the cost is probabilistic, which means the bottom of the standings is a game of expected value rather than guaranteed return. A team can play that game well and still lose the hand. The Romans, who put a goddess in the middle of every army’s prayer, would have recognized this immediately. The modern question is whether a franchise should organize itself around an expected outcome it knows it cannot count on. Buffalo organized itself that way for a decade. The hand it drew is, statistically, the hand the math always said was most likely.

The March Ahead

Part III takes the lottery model to its most theatrical form: the National Basketball Association, where weighted lotteries have been used since 1985 and where the gap between “best odds” and “the actual pick” has produced some of the cruelest near-misses in modern sports — including, in 2012, the Charlotte Bobcats finishing the season at 7–59 (the worst single-season winning percentage in the league’s history) and watching the first overall pick go elsewhere. Part IV turns to Major League Baseball, where draft picks have always been worth less, and where the league’s 2023 lottery rule change was a quiet acknowledgment that the Pyrrhic-victory incentive had been doing damage. Part V leaves the draft behind for college football, where the same shortsightedness wears different clothes. And the capstone returns, briefly, to the deterministic NFL: 2011, the Colts, the cleanest successful tank of the modern era. When the system is rigged in your favor, sometimes the right move really is to lose.

An aside the editor cannot resist: the Sallust line about Fortune ruling all things has been quoted in every century since the first one BC to explain why the wrong side won, and the next century has always lined up to roll the dice again anyway. The Sabres will be in the next one. So will Edmonton. So, by the math, will the team you root for. The goddess is patient. The cap is a hard ceiling. The next lottery is in June.