Before the internet, and before anyone could have a blog, the kings of sports journalism were newspaper columnists. These were the guys (and I say guys because a large majority of them were, and continue to be, men) who knew sports better than anyone else. They were eventually joined by television personalities, who were like the old columnists, except for the fact that many of them once professionally played the sport they covered. Then came the blogger movement. With the advent of the internet, anyone could have a voice in the debate, and a good portion of the ‘blogosphere’ was either familiar with advanced stats, or actively creating new algorithms to better understand the game. ‘Advanced’ or sabermetric stats have been around since the late 1970s in baseball, and since then have continued to grow, though the last 15 years or so have seen an explosion in new, computer-driven statistics across all sports.
Today, there are still some great columnists and analysts, and there are still some terrible ones, all amid the new blog-heavy sports media landscape. The worst are the ones who fall into simplified narratives, who mythologize intangible things – ‘heart,’ ‘grittiness,’ ‘toughness,’ ‘clutchness,’ and ‘hard work’ being the most popular – and value them above talent. The worst columnists’ favourite guy on the team was the undersized, usually white player who succeeded through hard work and perseverance, because that was the best story.
At the same time, new writers have found new avenues for looking at the game – and, increasingly, seen that the tropes of the traditional sports writer were wrong. That ‘clutchness’ is more random than an actual ability; that the hardest-working player can sometimes be the worst player on the team. So, inevitably, we’ve reached a point where a ludicrous dichotomy has sprung up: blogger, advanced statistics nerds versus the old guard sportswriters who just know the game better. And it is an insufferable battle.
It seems like every week, without fail, someone will start a debate about ‘advanced’ statistics. It usually comes in the form of a tweet or a column that decries those nerds who use newer statistics and ignore the beloved clichés of ‘heart’ or grittiness. I’ll use a recent example in the hockey world: Gord Miller, of The Sports Network, tweeted Tuesday morning that “[s]ome analyze NHL [National Hockey League] teams using ‘advanced’ stats like Corsi #’s. I prefer the Bill Parcells approach: you are what the standings say you are.”
A little background: Corsi is a stat that counts the number of shot attempts (including blocked shots or shots that miss the net) by either team, with the idea being that if you shoot the puck more than the other team, then you possess the puck more than the other team, and have a better chance of winning. Repeated studies have shown that positive Corsi numbers highly correlate with winning percentages, and that playoff performance is also highly correlated with good possession statistics. Bill Parcells, mentioned in the tweet, was a football coach in the 1990s and 2000s, so Miller’s choice of cross sport comparison is weird. Basically, he’s saying that good old wins and losses are a better indicator of team ability than those newfangled stats.
Miller’s tweet – and, basically, his ideology on what makes a good hockey team – completely ignores the fact that a team can be outplayed and still win through some lucky bounces. It’s a simplified way of looking at the game – team gets win, therefore team is good – and it’s the most reductive way to discuss sports. But these are the kind of people who are making a career off of opposing new statistics.They’ve become the meathead jocks of the journalism world, decrying the ‘nerd’ bloggers who live in their mothers’ basements looking at a spreadsheet instead of the game, or who deny their beloved traits of stick-to-it-ness, their easy narratives.
Another trend of stat-denying is by fans or writers focused on teams that are winning in spite of statistical trends that predict that they should be worse, or that they will eventually fall to Earth. If their team keeps winning in spite of the statistics, these people use that to dismiss stats entirely. It’s a logical fallacy – if one thing is wrong, even one time, then it can never be trusted.
Bill Barnwell, a football writer for Grantland, is one of the most widely read stats-focused writers. For the past two years, he has predicted that certain teams will do worse than the year before because their lucky streaks will end, and statistical benefits for the team will regress toward the average the next year. Barnwell basically becomes public enemy number one for these teams’ fans, who refuse to believe that their team will do worse. This year, Barnwell predicted that the Indianapolis Colts would do worse than last year. The Colts have started the season well, and I’ve seen one Colts fan on Twitter tell Barnwell that this start should make him reconsider the very idea of statistical regression. Again: one prediction is off (and, even then, there’s a whole season left to go, in which the Colts could still regress), so ‘advanced’ statistics are fundamentally wrong.
In hockey, there’s the continuing case of the Toronto Maple Leafs, who made the playoffs last year despite terrible possession numbers, and have started this season well with the same terrible possession numbers. There’s a whole cadre of Toronto media outlets and fans that have vehemently denounced ‘advanced’ statistics just because they say that their team is getting lucky, and will eventually fall to Earth. For instance, on Tuesday, as the Maple Leafs were getting hugely outshot but still holding a slim lead, Globe and Mail columnist David Shoalts tweeted that he could hear “geeks’ heads exploding all over their spreadsheets” and, after that, that the “Actuary Army had [him] in their crosshairs,” providing yet another example of the complete dismissal of advanced statistics based on one event – even before the whole season has played out. The only thing missing from that tweet was the sound of a bespectacled dork being shoved into a locker by your least favourite person from high school.
I can’t help but get the feeling that these are simply reactions to a threat; that statistics are getting better and better at predicting performance and showing us newer and more interesting ways of looking at the game than the old guard can. It’s their innate knowledge being pitted against stats, with the narratives and received knowledge of the traditional crowd under attack by new statistic-driven discoveries. That’s not to say that analysis of sports should only be based on ‘advanced’ statistics – there are certainly things that can’t be measured entirely by statistics – but to deny them outright is just plain dumb, an act of willful ignorance.
As these stats become more and more entrenched, they have become increasingly accepted by the sports fan and journalist community; we can only hope that in a decade or so, we won’t have dumb, influential people plugging their ears to things that can only improve their knowledge of sports.