Diverging Realities Part One: The Great Divide From Sharks Fans To All Of Hockey
Hey, Shark Circle here. Just wanted to pad the length here by putting up a “note” explaining this next series of blogs, Diverging Realities. In contrast to the more specific, ne’er continued series Beware of Advanced Stats In The Hands Of Less-Advanced Statisticians, Diverging Realities is going to take a more general look at the consequences that false statistics, biased “studies,” and the advanced stats community’s inability to acknowledge anything they can’t calculate are having on the hockey world.
Part One will touch on the Sharks, but focus mostly on explaining what I mean by “Diverging Realities” and what caused the phenomenon, and how it has divided the hockey world. Then Part Two will really dig into how this phenomenon has attached itself to the debate around Joe Thornton and Patrick Marleau in SJ and made the Sharks into a polestar for this divide.
Not For nothing, it’s been the most difficult blog I’ve written here, even more involved than the Beware blog, or the blog advocating for more skill in the NHL-game, and bigger ice. Or even the ill-conceived Scout With Almost No Footage one. But as someone who likes to be part of the ongoing, evolving conversation that is NHL-hockey, I feel this is one of the most important subjects I’ve written about. Because the way we all experience, analyze, and understand the game is very important to the way we discuss it, and I’ve been disappointed the last few years to see hockey discourse become almost as partisan as political discourse. And that’s what I felt needed addressing, so I hope you enjoy the slog blog!
If you’ve listened to Sharks’ GM Doug Wilson speak frankly this offseason, then you know that’s probably a pretty good approximation of what he’s been thinking to himself about Joe Thornton and Patrick Marleau lately, as it now appears he’s finally come to realize what most NHL fans have believed for years: that there is something going on with those two as leaders of the Sharks that has hindered the team’s ability to succeed in the playoffs relative to what was expected of them (internally and externally).
While that something is too nuanced to quantify in the form of a single number, most Sharks fans who have watched the same stories unfold every playoff for the last decade have a pretty good idea by now of the recurring issues. And now, with the Sharks coming off a four-year stretch in which they have only won a single playoff series, it seems that, finally, Doug Wilson does too.
After all, it is his job to know what the team’s underlying issues are, and the conclusions he’s come to after analyzing “every level” of the organization this offseason confirm what most Sharks fans have suspected for years: that the Sharks have just as many problems off the ice, in the locker room and in the team’s psyche, as they do on it. From reporting how some players told him that they felt more like “co-workers” than teammates, to giving voice from inside the organization to a concern long-held by many Sharks fans about Joe Thornton and Patrick Marleau with the hint that some players just want to “live here,” Wilson painted a clear picture that the team suffers from problems related to culture and leadership.
And in the 29 other NHL-cities, shouts of “NO WAY!” could be heard reverberating from rooftop to rooftop.
If only someone had, like, said something! Or thought of that earlier!
Yes. Yes, indeed…. But when it comes to getting this team on the right track, better late than never.
Would it surprise you to hear, though, that not everyone can agree on what the right track even is? Because in case you haven’t been on the internet lately, it seems a very vocal group of people would disagree with Doug Wilson’s admissions about the Sharks’ culture-problems. The very same group of statistically inclined folks (#fancystats is the twitter hash-tag) who have so often disagreed with me in the past when I criticized Doug Wilson’s performance as Sharks’ GM. And so no one can be reasoned with. Not only can they not agree with me or the consensus on Thornton and Marleau’s playoff-history, they can’t even agree with themselves. Because those idiots used to praise Doug Wilson all the time. Those previous-selves. It’s all so complicated. All they seem to know is that they can’t praise Wilson anymore, or quite possibly ever again. Cue dramatic music. Not now that he’s come to the same conclusion about Thornton and Marleau as the rest of us who trust our eyes on the subject have. Not now that he looks like he’s finally ready to take the Sharks in a different direction, one that doesn’t include Thornton or Marleau.
If they’ll waive the no-trade clauses he just gave them a few months ago, that is.
Yes, the beloved Doug Wilson. He of the secret lab of #fancystats that the rest of us couldn’t even imagine. He of the brilliant use of corsi-stats to draft Tomas Hertl, a home-run. He of the brilliant use of whatever unimaginable #fancystats that directly contradict corsi to sign Mike Brown… to a pretty sizeable raise…
Unimaginable, indeed. With contradiction like that coming from within the Sharks organization, maybe it should come as no surprise that a small segment of Sharks-fandom appears to be losing its collective mind (and boy is it collective. Like a cult). Nevermind that they are known for contradiction as well. But I have to say I’m
disappointed frustrated disheartened, too, which is why I’m writing this long-overdue blog on the Sharks (mostly in Part Two) and much more. Disheartened primarily, of course, by the Sharks’… season, but also by what I see as the emergence an almost cultish, arrogant attitude among a certain segment of fans who believe, based on their excessive use of advanced stats (aka #fancystats), and the aggressive culture of group-reinforcement that always accompanies any group that believes they’ve found a magic formula, that only they have access to reality in the context of understanding hockey, while anyone who disagrees with them must just have the wool over their eyes or be stuck in their old ways (the latter sometimes being true), to describe the way they view their dissenters much more kindly than they do.
But this attitude has persisted to such an extreme that now even the most obvious happenings on the ice that occur before our very eyes are being ignored by this segment of fans if they cannot find empirical statistical data that also says it happened. And no, it doesn’t go both ways. Because who needs eyes, those outdated, unreliable old things, if you already have access to a perfect formula for quantifying everything that happens on the ice, and a couple hundred friends on the internet to keep reinforcing the false assumption that nothing of any import could possibly happen outside of what that formula quantifies, whether on the ice or off?
Living in San Jose during the midst of the Thornton and Marleau sagas, where notions of culture, leadership, “grit,” and so forth have always loomed, it makes sense that the introduction of a new lens to view hockey through would stir a divide within the Sharks fan base. Now there is one group of Sharks fans who watch the games and report on what they see happen with common sense and the aid of stats, within their limits, and there is another group of Sharks fans who may also watch plenty of hockey games, but who look first to stats for their analysis of those games, and who believe in nothing they see on the ice if it is not substantiated by the stats they have on their screens.
And while the divide between these groups is certainly at its worst when you combine it with an emotional debate surrounding two controversial players within a fan base where the always-difficult question of regular season vs. playoff-performance looms, lately I’ve started to see it come up everywhere, whenever anything significant happens in the hockey world. No signing or trade happens now, with any team(s), without the #fancystats community speaking up to correct everyone else’s take on it. No, he’s really not that good. Or No, he’s really way better than that. His zone-starts and corsi-rel show… So I believe what we’re seeing with Sharks fans is just an extreme example of a divide that’s happening with fans throughout the NHL. And that is a divide between the eye-test-and-common-sense fans who consider reality to be what they observe happen on the ice, and the #fancystats fans who essentially have their own hockey-reality, which is calculated and created in excel documents, rather than seen or experienced, and which only represents actual reality so long as the stats are correctly, and objectively, calculated.
Because you won’t see much of a debate among the Sharks fans who trust what they’ve witnessed, (like me), when it comes to Thornton and Marleau’s playoff-history in San Jose, to go back to that example. We feel like we know what we saw. After all, it’s been happening nearly the same most years for almost a decade now.
So there’s no divide or debate here. It’s only when you introduce a collection of “advanced stats” that claim to overrule your observed-experience that you start to doubt your experience and buy into this new reality. That’s how the divide begins inside many people. And since everyone knows the eye-test is imperfect, the instinct for many of these people is to let the latter, “statistically based” reality overrule the one they see for themselves. After all, everyone knows numbers are facts, and since statistics use numbers, everyone assumes statistics are facts, too. Nevermind that numbers aren’t always facts, either, not when they’re miscalculated. Is 5 + 6 = 17 a fact? Because 5, 6, and 17 are all numbers.
Sadly, it really can be that easy to trick people into believing that an incorrect opinion is an indisputable fact. And when you start to see the incorrect as being correct, and the subjective as being objective, then you really have been tricked into a believing a “new reality.” It might sound harsh to say some hockey fans have become disconnected from reality, or created their own, and by no means is that the equivalent of going crazy or hallucinating in real life, but speaking strictly in the context of hockey and “truth” represented by statistics, and what you type into your keyboard, the difference between actual reality and a false, new one can be as small as one miscalculated number on a screen, or as nuanced as knowing the difference between fact and opinion. And since facts, i.e. numbers, can be easily miscalculated, and since people can be easily manipulated into believing opinions are facts, it shouldn’t seem a bridge too far to believe that hockey fans in mass can, and have already, become divided into seeing two different versions of hockey-reality.
And falling for it can be the easiest thing to do, especially if you like the look of the new reality you’re being shown better than to the old one, and you’re being told it’s okay to believe in it because “it’s actually always been the true, statistically-supported one reality anyway.” And the further you fall down the rabbit hole, the more you like what it says about the team or players you wanted it to say those things about in the first place, and the more empowered you feel for being able to find, and even “calculate,” all these secret-truths, that most other people don’t know about. Because while you can’t control what the players do on the ice, or how well they perform, and that can be very frustrating, you can control numbers on a spread sheet, and often you can even control what they say.
It doesn’t even take a skilled statistician to skew the truth because people inherently believe any argument that claims to be stats-based, even if that means letting that argument, or the people presenting it, overrule what they see. And if you’re, say, a Sharks fan who hasn’t seen much to like in terms of playoff-results over the last decade, then, perhaps, especially if it means overruling what you’ve seen. Let’s face it; given a choice, who really wants to believe their favorite players, who they have rooted on and defended for years, have turned out to be guilty of the flaws everyone always said they were? If someone can offer you a plausible counter-argument to the disappointing history you’re trapped in, you’re probably going to take it without much investigation into its validity and skip all the way home along singing Thank You Mr. Fancy Stan. (I expect lyrics to this in my inbox within the week!)
And so you have an alternate reality that is not only all to easy to trust because it carries the label of statistics, but one that is also very enticing to fans who either want to feel superior because of the implication that statistics (i.e. the objective) overrule the eye-test (i.e. the subjective), or who are all too happy to hear that their favorite team or player isn’t as disappointing as the results would suggest.
It’s really the reason this group of fans is so unlike any other in its self-assuredness. They believe they have God on their side, God in this case being numbers or facts, which we all agree should be all-powerful in any analysis or discussion if that’s really what they were. But in reality, these people still maintain control over that analysis or discussion. Because they control the stats. They calculate them, and, often without knowing it, maybe when they really want to win an argument or have the stats come to a predetermined conclusion about their favorite player or team, manipulate them. So they get to simultaneously control and manipulate the most powerful tool in any discussion, “facts,” while at the same time carry with them the attitude of how could anyone disagree with me when all I do is cite facts, and everyone knows facts are by nature uncontrollable and un-manipulateable, otherwise they wouldn’t be facts?
Obviously, the people in this position are not accounting for their own biases or qualifications when calculating those “facts” and relaying them to the hockey-masses. Instead these people view themselves and their #fancystats as manifest destiny for hockey analysis, basically, which is a very problematic attitude to have when you’ve awarded yourself that lofty status without being vetted first, without the question being asked as to whether you’re actually qualified for that position on top of the world of hockey-analysis.
For instance, are your stats really as infallible as you say they are? Your “calculations” say they are, but isn’t there a conflict of interest involved when you’re testing your own product? And since “testing” in this instance involves (no pun) advanced statistical-calculations, forget the conflict-of-interest question for a moment–are you even qualified when it comes to the making such complicated (in some cases) mathematical calculations? Maybe the stats themselves really are the future of hockey analysis (when used with common sense), but what makes you qualified to take them there accurately? What is your resume? Do you have a degree in statistics? What work have you done in the field that qualifies you to do this work without making miscalculations that lead to incorrect conclusions that then get spread throughout the hockey world as fact and gospel?
These are the kind of questions that should be asked of anyone, in any field, before you elevate them to Supreme Ruler of their domain, before you trust them and allow them influence over your beliefs, as so many hockey fans have. You wouldn’t trust someone’s take on neuroscience if they didn’t have a high-level degree and experience in the field. If your life depended on solving an advanced calculus equation, you probably wouldn’t like your chances if the only person you had to help you was a blogger with no degree in mathematics. You’d probably much prefer someone with a Master’s. But in a field as complex as statistics, we’re just supposed to trust the studies done by various people and various blogs, by everyone, really, without first asking any questions as to their qualifications?
Unfortunately, yes. At least, that seems to be the answer within the #fancystats community. Because these questions haven’t been asked of this group very often. And when they have, like when I investigated about a year ago whether the calculations done in a few prominent #fancystats blogs were even accurate, the answer was very clear in each case, as my long-time readers know. Many of them are not qualified, otherwise I wouldn’t be so critical. And I don’t mean any offense to these bloggers who I don’t doubt love stats and love hockey (in whichever order), but the fact is that their calculations are wrought with bias and mathematical error, and I feel a responsibility to report that. If they were calculating their statistics accurately and objectively, and not infusing their conclusions with bias or over-extending their stats to support biases, then there would be nothing to be critical of. Because the stats themselves, when calculated correctly and used within their limits, instead of being touted as nearly limitless, or used to explain away their limits as meaningless, or used to make bold proclamations, are a wonderful tool.
But the pertinent question remains, if even a layman like me can point out scores of miscalculations in the work of these #fancystats blogs just by looking at a few of their posts, then how many more miscalculations are there that I missed? Or in blogs I have not even looked at? That’s why the independent parties who should be testing these statistics, and testing the qualifications of the people presenting them, should not even be someone like me (if the structure of this sentence hasn’t made that clear already). And the harmful spread of misinformation that has resulted from miscalculations like the ones I pointed out is why it’s usually just considered standard operating procedure to make sure a person or group is qualified and objective before you start taking their “facts” at face value. In any field.
But the hockey-world at large has not done that in this case, and the consequences to that are many. You have scores of hockey fans believing they’ve found a greater level of truth, that they’ve discovered all these secret, underlying facts about the game, when in reality it can be the exact opposite. When someone comes to a conclusion about hockey based on bias or statistical-miscalculations, the result can be not just the same level of “misleading” that simply watching the games with your eyes can be, it can be even more misleading, and even more wrong, than when you just watch the games. Watching the games, you might miss out on the fact that a certain player is benefitting from a high percentage of offensive zone-starts, or vice versa, but at least you’ll normally be able to see who’s having a good game, whatever the underlying circumstances. You can still miss seeing underlying things that are happening but you won’t think you’re seeing something that never actually happened at all. Huge difference.
Consider it the difference between Mikhail Grabovski and Tom Gilbert. Mikhail Grabovski was buried with poor zone starts and difficult matchups while playing with the Toronto Maple Leafs under Randy Carlyle, so his point-production dropped as a result, and he became undervalued among most hockey fans, while the #fancystats community at large maintained they knew what was really going on, and that he was a better player than his production with the Leafs suggested.
And they were right, on the whole, since simple one-player analysis is harder to mess up. If you asked the average #fancystats-pledge during that time what they thought of Grabovski, and you asked the average casual fan what they thought, the #fancystats-pledges would have had the higher, and more accurate, opinion of Grabovski. However, some of that is also down to injury-problems and inconsistency. Grabovski finally got the shot he really deserved with a talented offensive cast in Washington this year, and he still failed to put up more than 35 points. Additionally, there are plenty among the eye-tester-common-sensers who could plainly see Grabovski’s skills with the puck and who understood, obviously, that Grabovski’s 9 goals two seasons ago were not the same as, say, if Brandon Bollig were to hit 9 goals in a season.
In other words, if you have a good eye for the game, you don’t need #fancystats to tell you that a flashy player like Grabovski is talented, and that he was always less productive than his talent-level suggested he should be. I pegged Benoit Pouliot as an underrated signing with the Bruins a few seasons ago for the exact same reason, despite his poor point-production, without ever looking at his #fancystats, and if his postseason play with the Rangers recently is any indication, I was right. The same goes for Anton Stralman, who I also pegged around the same time, and who I’ve now heard is a #fancystats favorite for his excellent possession numbers. I knew nothing of those at the time–just that I liked his skill set and the way he was performing under the radar for the Blue Jackets at the time.
And I even liked Grabovski. Loved him, actually, when he came on the scene under Ron Wilson, with a low cap-hit. Wanted the Sharks to get him. But the last few seasons, he’s seemed to fade a bit in my eyes. He’s still talented, but he just hasn’t looked as explosive to me.
Could that just be me getting fooled by the poor zone-starts? Is he actually as good as he always was? It’s possible. That’s how it works: the eye-test isn’t perfect. Those are the types of things you can miss. That’s where the stats can and should be super useful, if the analysis of the #fancystats community at large wasn’t impossible to trust at this point because their reputation has been dragged through the mud by a lack of quality-control.
But what about Mike Richards? I’ve had an eye for these things in the past, to the point where I’ve been able to predict how players would start the season the last few years just by watching them play in a single pre-season game. And in Mike Richards’ case, more than just the start of his season. Ever since he got traded to the Los Angeles Kings, I maintained that he was out of shape (compared to his best days in Philadelphia). And I never looked at his advanced stats. He just looked slower, weaker, and less physical to me. And I’ve said that consistently each year he’s been in LA. And each year his production has gone down further.
But does that mean he’s out of shape, or could it just be his zone-starts and quality of competition? Outside of the eye-test, I would have been none the wiser. I literally did not look at his #fancystats. But finally, after years of declining production, what does LA Kings GM Dean Lombardi come out and say last week? To paraphrase: “No, we’re not going to buy him out because he’s just 29, so we know his declining production isn’t due to age or injury. He should be in his prime years. And he knows too. He knows he can be better. You just have to understand that when you’re 29, you can’t prepare the same way you did when you were 23. Your body doesn’t respond the same.”
And so on. Has to be more committed. Body doesn’t prepare the same as when you’re younger. It’s code for “out of shape.” I was right about that one and I saw it from day one. Three seasons later and they’ve finally made a point of noticing it in LA and asking Richards to do something about it (good luck). And some people say there’s no such thing as varying levels of effort or commitment…
And Mike Richards is also a great example of how knowledge about what’s been going on off the ice can be helpful and illuminating to understanding what’s transpiring on the ice, even for someone like me who primarily relies on analyzing on-ice play. But when I saw Richards arrive in L.A. looking slower and weaker than he had during his best years in Philadelphia, and then I also heard that Richards’ commitment and attitude had been questioned in Philadelphia leading to his trade, that gave me a hint that he probably wasn’t just looking worse because of bad zone-starts or any other external factor. Those rumors about his attitude helped me surmise that he alone was at fault for his declining form, and that his alleged party-attitude in Philadelphia had probably manifested itself in worse offseason work habits which were negatively affecting his physical capabilities. And I believe I was right, as Dean Lombardi just hinted at.
Well, Joe Thornton and Patrick Marleau have had a fair share of murmurs and rumors surface about them over the years, with regard to their attitudes and commitment, particularly during the playoffs, just like Mike Richards has had his own brand of murmurs. And it seems that in his case, the smoke led to some legitimate fire. At a time when Puck Daddy still had Mike Richards ranked above Anze Kopitar on their Top 100 NHL Centers list, I made the common-sense connection between what I saw of Mike Richards’ game on the ice, and what I heard of him off the ice, and it left me with a greater understanding of Richards’ game than most had at the time. And yet this is a connection #fancystats-pledges, particularly those who call themselves Sharks fans, refuse to make, or even investigate, with regards to Joe Thornton and Patrick Marleau, simply because they can’t quantify any of it in excel. But they can’t quantify offseason workouts, either, and I don’t think there’s anyone out there familiar with Mike Richards who would argue that they don’t affect on-ice performance.
But that doesn’t mean I’m always right, either. I thought Justin Williams was slowing down a bit the last two seasons. I still do, actually, when you look at how he skates. But he just won the Conne Smythe. Still I don’t think I’m wrong about him slowing a bit at the age of 32. I think he just made a lot of great plays regardless. And at least when you go by what you see, you’re going to have a hard time actually making something up that isn’t there at all. So I noticed that Williams wasn’t accelerating as well as he used to, something that usually affects future production in a negative way, but he found a way to do his twirls along the boards and make great clutch plays all playoffs anyway. I mean that’s what happened. Even when I was wrong in a sense, I was still able to see what happened instead, as it happened. It’s hard to go wrong there.
Now, in contrast to that observation-and-common-sense-based method of hockey analysis, there is the Tom Gilbert way. And the Tom Gilbert way refers to when Fear The Fin wrote a blog on who the best US-born defensemen in the world were about a year ago and put Tom Gilbert at the top of the list next to Ryan Suter (ahead of Keith Yandle, who didn’t make the list at all), citing a Copper & Blue article that used faulty, incorrectly calculated statistics to elevate Tom Gilbert to elite defenseman. Was it on purpose? I highly doubt it. But that was an example of people thinking they had a “secret fact,” in this case that Gilbert was better than everyone else thought he was, when in reality they were just citing the kind of completely random conclusion that typing random numbers into your calculator will give you.
Because when you miss a calculation in math, it’s not like spelling where if you get one letter wrong out of ten, you still have the other 90% of the word spelled correctly. No, in math, once you get one calculation wrong, it causes every other calculation you make using that data to be wrong, too. Increasingly and exponentially wrong. Copper & Blue, or Fear The Fin building off their work, weren’t partly right, or mostly right. They messed up the stats and nothing they said in those articles about Tom Gilbert has any value or truth as a result, even though those articles look to both still be up on theirs site giving Canadiens fans (seeing as Montreal just signed him) literally-false hope to this day. (Although Gilbert has always been talented with above average size, skating, and skill for a defenseman, and he’s improved to my eye since leaving the Oilers).
But Fear The Fin did what most hockey fans do, unfortunately. They just assumed that because Copper & Blue is a prominent #fancystats blog, they must know what they’re doing. That they must be qualified. So Fear The Fin didn’t fact-check Copper & Blue’s calculations, and that right there is precisely how completely random, untrue information gets passed from one blog to another in the hockey blogosphere, and then from those blogs to hundreds of hockey fans, and from those hockey fans to their friends, and pretty soon you have a community of hockey fans in the thousands who are completely misinformed while simultaneously believing themselves the most informed. They aren’t just watching the games and maybe seeing eighty or ninety-percent of reality, while missing ten or twenty-percent of it; they are actually making up their own one-hundred-percent-untrue reality based, in that case, on faulty statisticians, or in other cases, on bias or a combination of both.
So that’s the difference between the imperfect Mikhail Grabovski way and the imperfect Tom Gilbert way. It’s the difference between a more humble set of fans who see a good deal of what’s going on in a hockey game, who may be missing some of the more complex, underlying aspects of the game but who still have a good conception of reality, and an extremely arrogant set of fans who, when they’re wrong, don’t see ≤100% of what’s going on in a hockey game, they see ≥100% of what’s going on. They spread ideas about hockey players that have never even happened on the ice, but rather originated in an excel document due to confused math or manipulated statistics. And that is the definition of believing in, and spreading, something false.
Now, my instinct is always to be as kind in my criticism as possible without sugar-coating the truth, as long as it’s still enough to make them cry (wait, what?), which is why when I began writing this blog, my instinct was to characterize it as a false-belief rather than saying that people had actually lost base with reality. But the problem with calling it a false-belief, instead of calling it what it is, is that we’re dealing with numbers and statistics here, and people who think they’re spreading facts. And that’s the one thing they’ve consistently gotten right, sort of: there is no room for opinion or different beliefs when it comes to numbers. You’re either right or you’re wrong. They just don’t realize when they’re wrong. But you’ve either observed, or in their case calculated, reality correctly, or you haven’t. And I just detailed the differences between the two methods when you get something wrong with each.
But if you really think calling this divide “diverging realities” is hyperbolic or unfair, as some probably will, let me put it this way. You’ve seen this exact same phenomenon happen before. You’ve seen it happen with religion, when Christianity was introduced to the Jews and some converted, and some stayed with Judaism. And to this day, people believe completely different things when it comes to religion. You just don’t call it “diverging realities.” You call it different beliefs. But beliefs, not realities. Because it’s religion, and even religious leaders say, “you have to take it on faith.”
You’ve seen people diverge in many other arenas, like politics. Republicans and Democrats can’t even agree on the basic facts anymore, in case you’re wondering what the dangers are in terms of where this divide could lead for hockey fans. Do you want to see hockey covered in the same partisan way that politics are, with each side having its own TV channel and its own favored writers and…
It’s already happening, isn’t it? Well there goes my point. It’s already happening to a degree in the blogosphere and on twitter, yes. Because now hockey has its own alternative to the way the game has been analyzed for decades: advanced stats. This is hockey’s new party, or new religion, except unlike religion, the #fancystats community does claim to be not just factual, but provable, statistical. So while a Christian person might refer to Judaism as a “false belief” because it’s a belief that conflicts with theirs, you have to refer to people who believe in a false statistical conclusion as people who believe in a false reality. But since no religion is a proven fact (or a proven fallacy), you can’t really say any one religion is false, indisputably, even if you believe that. With statistics, it’s different. It’s not just like I’m a member of one side, so I believe the other side is false. No, each statistical study is either calculated correctly or it isn’t. It’s either objective or it isn’t. So unlike disagreements between beliefs, when I say a statistical study is false, it’s not because of sides, it’s because it’s false, period. Yes, people still believe in false statistics, because that’s the only word the dictionary has to describe what they’re doing, but it’s different because you can actually prove, indisputably, that what they believe in is false. And once something can be proven as fact, it’s no longer a belief, it’s reality. Or when you can do the opposite and prove something is false, it’s no longer a legitimate belief, it’s a false reality, or a fallacy.
Of course, I’m not saying that every person who enjoys statistics in hockey has fallen into a “false reality.” Not at all. There are some within the advanced stats community who seem to very much know what they’re doing, who have a good grasp of how to calculate the stats, and who do so objectively. You just have to be diligent in fact-checking these studies and approach them, and the people who claim they can sell you the hockey-world, with a healthy bit of skepticism. But most people don’t, and so you can see how the potential for misuse, abuse, and misinformation-spreading is there, often-times unknowingly, if the person publishing the stats makes any miscalculations with the factual side of statistics–the numbers, or has a bias that causes them to confuse his or her subjective opinions with the objective statistics. And as I’ve already been over, I’ve found in my experience that they do.
So to summarize my overall attitude towards “advanced stats” in hockey in the most simple way I can think of, there is nothing wrong with the statistics themselves, or even the existence, in principal, of a community of NHL fans dedicated to #fancystats, other than that I’ve noticed the last few years that the #fancystats community that exists in reality keeps telling hockey fans that 5 + 5 = 12, and 7 + 4 = 18, and all sorts of other conclusions derived from incorrect statistics, and people keep believing them for the various reasons I’ve spent this blog detailing. I’ve tried speaking up before, without much success, and it’s not my job, and not a job that’s taken very kindly to either, to be the full-time fact-checker to these people who are clearly making plenty of passionate followers out of selling often-untrue information.
But at times it gets to a fever pitch that disheartens me to a point where I feel I have to speak up, for history’s sake if nothing else! You know, when people look back hundreds of years ago at these end-times in hockey analysis, I want them to know that there were some people out there who tried to warn people about the
zombie fancie stats plague before it was too late and everyone’s mind was taken and all that.
But, to be serious, I do think it’s important that hockey fans can at least agree on a basic reality, like how we all still agree that there are two hockey-nets on the ice (progress!), and I feel like we’ve reached a point lately where even agreeing on many of the basics has become a tall task. One day we all watch the NHL Draft, as just happened a few days ago, and listen to scores of scouts comment on which prospects have the best “shots,” and for that one moment, when everyone is too directly faced with the obvious to remember anything different, common sense rules the day. And then when training camp starts, and all the Sharks young players talk about how they want Logan Couture to teach them his “wicked wrista,” common sense rules for a few days more. But then the season starts, the excel sheets get fired up, and because shooting skill doesn’t fit with the preferred corsi-based statatistics of the #fancystats community, their divergent-reality comes roaring back on that subject. “There is no such thing as sustainable shooting ability. Maybe Ilya Kovalchuk, or Stamkos, but that’s it. Besides that, it’s luck. There’s no way the Anaheim Ducks and Pittsburgh Penguins are posting good shooting percentages for any other reason but luck and chance, so expect it to fall soon. It’s unsustainable. Expect them to fall soon.”
It’s a problem. The Ducks, for instance, just competed for the President’s Trophy two years in a row, but these fans will never give them their due until they do it with a higher quantity of shots, and a lower shooting percentage. It’s madness. They just drafted Nick Ritchie, who scouts say has “the best shot in the draft.” So when he’s shooting at an “unsustainable clip” for the Ducks a few years from now, will anyone remember that? At what point exactly does the shooting skill we look for in draft prospects turn into shooting luck?
Reality, folks. Two of them. I actually think that explains it the best. And it really is a problem for fans, and bloggers such as myself, who want to engage in discussions about hockey with other people within a basic, agreed upon reality of common sense. Imagine if, during the draft, when Bob Mckenzie commented that “Nick Ritchie has the best shot in the draft,” Pierre Mcguire went, “Actually, no, Bob, there’s no such thing. Above-average shooting ability is unsustainable in the NHL.”
“But I saw it! We even measured it! It’s also the hardest in the draft!”
“Well our stats show that for every 1MPH increase in shot-velocity, a shot loses the exact equivalent in accuracy, so that’s how everybody’s shot actually turns out to be the same, Bob…”
I mean come on. It would be madness. It is madness. But that is exactly what’s happening to people these days when they try to have a sensible discussion about hockey online. (I know, “sensible” … “online,” a stupid thought in the first place). You just can’t discuss hockey that way, pretending something doesn’t exist when it’s right in front of you. Just because you don’t know how to quantify something doesn’t mean it doesn’t exist. Especially when it’s so fucking obvious TSN talked about it 500 times during the draft, and Sharks fans see it every game with Logan Couture.
So hopefully I’ve done a decent job articulating the issue here, and hopefully people from both sides of the great divide will be more cognizant of it in the future, and more aware of what kind of information they’re letting pass into their brains without any scrutiny just because it has the label of “fact” or “statistic” attached to it.
Thanks for reading!
Written by Shark Circle
Subscribe to Shark Circle by entering your email and clicking the SUBSCRIBE! button in the top right corner! Your readership is appreciated!