There’s always a lot going on this time of year and news & thoughts start to pile up. Before getting into the specific hockey stuff I thought I’d briefly go over something I see regularly (in hockey and other circumstances): confirmation bias. I think the term is broadly familiar, but not the application (or, perhaps, the concept). The dry-as-dust definition: “the tendency to search for, interpret, favor, and recall information in a way that confirms one’s preexisting beliefs or hypotheses.” So let’s take this (ignore who posted it, as I think David Johnson understands this concept–I’m interested in the reaction to it). The link is related to Corsi as a predictor (CF% teams have not done well in the playoffs this year). There are a lot of people (including Sens management) who are uncomfortable with analytics–it’s new, it’s math, it’s dominated by nerds, and it tends to dismiss old ways of thinking. This makes people uncomfortable, and who wants to be uncomfortable? A natural reaction is to reject the new information and look for any evidence that it’s wrong (cf classic Tortorella comments). When you don’t like something–when new information makes you uncomfortable (or in the case of business, costs you money or threatens your job)–there’s a natural tendency to want to find reasons to dismiss it.
Travis Yost picked the Sens to win in 7, saying the series is “brutally close; I have no reason to believe EDM/OTT are any better than ANA/NYR. But Best Skater Theory marches in” (this, I think, is partially in reference to the Alex Novet piece below). I certainly hope he’s right, but I lean the other way.
There was a lot of chatter about the attendance for game one against the Rangers. Nichols goes through all the nuance of why it’s not a sellout, but to me the main point is this one:
every effort should be made to exhaust ways to help the front office efficiently get the most out of its [limited] resources — which is where it feels that Melnyk’s playoff mandate does not work in concert with a long-term vision that brings a championship to the nation’s capital.
It does not, nor does his over reliance on old school thinkers like current management.
Speaking of management, Nichols’ most recent stenographic expedition looked at recent comments from Pierre Dorion and what stood out to me is this:
The numbers express confirmation bias in that the organization strictly relishes or emphasize the numbers that they want to know or hear about
This is exactly right. While Nichols tries to squeeze a positive in the Sens not just employing yes-men, having contrary opinions makes no difference if you are going to ignore them.
Ross A made me happy writing this about the Sens/Rangers stats:
some people might have wanted me to include such as blocked shots, faceoffs, and hits, but those haven’t shown any strong correlation to winning, so I found it a bit unnecessary to include them
Absolutely; including these stats, often thrown out by broadcasters, would be pandering. Ross knows these numbers, while fun, are irrelevant, and rather than including them as if they matter, he excluded them–bravo.
Ross also linked Alex Novet‘s article from March where he attempts to demonstrate that hockey is a strong link game (more akin to basketball than soccer). What is a strong link game? Novet breaks it down as follows:
-
the team with the best player usually wins
-
Therefore, teams should prioritize acquiring the very best elite talent, even at the cost of having weaker depth than opponents
-
This is important for roster construction now and has the potential to become even more important as teams get better at assessing talent and market inefficiencies become less common
My gut, which isn’t involved in thinking processes at all, rejects the second point. I like depth and I’m aware of teams that have failed due to a lack of it, but Novet is looking at and overall numbers and sometimes teams will buck a trend. The stats he sampled are from 08/09-15/16, so for old warhorses like me I have to throw out impressions from the old days when considering it. After going through these numbers, he concludes:
Getting the very best players is essential to success. Phrased this way, it sounds obvious. But the above shows that this is the case even at the cost of creating weaknesses elsewhere in the lineup. This has implications for many of the major decisions that general managers make.
Novet does admit a lot more work needs to be done to solidify his theory:
it would certainly benefit from attention from a real statistician. I have tried my best to be rigorous and transparent, but my statistics knowledge is limited and it is possible that this work is flawed. … more work on the effect size of roster imbalances would help define exactly how much the tradeoff for elite talent is worth. Second, Jack Han suggested splitting this work into offense and defense to see if the weak link / strong link distinction is clearer in one side of the game, and this sounds like a worthwhile investigation. In addition, it would be nice to find a compelling method for including goalies in this analysis. Finally, more work should be done to better understand the marginal cost of improving each part of the roster.
I agree wholeheartedly about the need for more work. On the surface one elite player is not enough for a Cup; adjusting for position (forward/defense), adjusting for goaltending, and finally, determining if there’s gradations of depth (just how bad can it be before it goes beyond the pale of elite players to solve). Regardless, it’s an interesting data set and not one I would have surmised (I always knew elite players were required, but assumed there was a baseline of depth needed).
I didn’t comment on this at the time, but let’s talk about the Sens signing Maxime Lajoie (5-133/16) and ask ourselves: why the rush? They’d just drafted the Swift Current Bronco, so they had two full years to make their decision, and because he’s from the CHL there’s no possibility of him jumping to Binghamton–so why? Is he a highly regarded prospect? No–while a universal pick from drafting sources, no one saw him as particularly remarkable (you can read my post-draft breakdown here). So why the rush? I can’t find any similar case outside of universally accepted top prospects for the Sens. It’s a puzzling decision–why not wait and see how he performs? Ottawa burns enough contracts on players who don’t pan out (like Vincent Dunn). It’s a head-scratcher.
The Belleville Sens made their first roster move, signing 24-year old CIS grad and defenseman Jordan Murray to a two-year AHL deal. Needless to say the CIS is not a common route for even AHL-players to be signed (and I have no idea why it’s a two-year deal). The former QMJHL player (three seasons, the last with Drummondville, 64-13-43-56), finished with a career year for the University of New Brunswick (30-14-26-40). He spent 5 games with Binghamton at the end of the season (5-1-1-2), which seems to have sealed the deal. I have no idea what to expect from Murray–I have no issue with the Sens signing players from unusual leagues, but the commitment to him is strange, particularly given how many blueliners are already signed for next season.
This article is written by Peter Levi (@eyeonthesens)
eyeonthesens 