Well, about a third of the 2015-16 NHL season is in the books. Our proprietary metric, the Advanced Bracton Score, seems to be doing well so far in predicting conference rank and results. For those of you who are following us, you are aware that we have been updating a Spearman’s Rho statistical rank calculation to compare our predictions versus actual results as the season has unfolded.
Through 12/5/2015, we have posted a 0.60 in the eastern conference, and, after lagging for much of the season, our western conference Rho is now a respectable 0.44. Both of these numbers indicate, at least to us, that there is a reasonably good correlation between our statistic and the results demonstrated so far this season. We will explain below some of the surprises so far which frankly, we did not see coming from our analysis.
Before we probe that, we wish to remark that we have also performed a Rho calculation for work performed by sportingcharts.com prior to the 2014-2015 season. In their study, the authors attempted to determine final conference rank for last season using Corsi and Fenwick possession stats as the precursor for team success. What we found was that their Spearman’s Rho, which we have never seen in print before, was 0.46 in the east and 0.27 in the west. We bring attention to this because the Advanced Bracton score has handsomely outperformed this popular view of how hockey teams win games in 2015-16. While it is not our intent to dispel the value of possession statistics, we will be quite interested if the Advanced Bracton metric continues to outpace Corsi and Fenwick (ie., posting higher correlation to final conference rank determination than the sporting charts authors found).
While crystal balls are mostly universally foggy, our metric has been dead on in many cases where much of the hockey punditry has swung and missed….badly. We present Exhibit A, the Anaheim Ducks as the poster children of a hockey enigma to the mainstream…..but not to us. Prior to the season, we kindly asked readers of Anaheim’s SBnation.com message board if they could tell us why so many in the hockey community had Anaheim as an odds on favorite to win the cup this year (over a variety of teams in the east, excluding Washington and Montreal, interestingly). We mentioned that the team had only 10 more goals scored than allowed last year, goalies that ranked in the lower half of the league, and an Advanced Bracton score that was mediocre at best (and should have resulted in a first round playoff loss). Not only did Anaheim look like an also-ran last season, but their 11th place result so far this year is in line with what the AB score would suggest. However, since ANH Corsi and Fenwick scores were excellent in 2014-2015, the conclusion behind their first place finish last season must have been due to possession. We recognize that the west is tight in the middle teams and that there is a lot of hockey remaining, but the slow start of the Ducks is in no way a surprise to us.
What is however an enigma to hockeyfreeforall.com is the lackluster play of the Calgary Flames and Tampa Bay Lightning. More than any two teams in the league, these two are killing us (although Calgary has experienced an overdue resurgence, 8-2 in last 10). Consider this, we predicted Calgary would finish second in the conference. If Calgary were not presently 11th, and were swapped with Colorado, who is seventh, our western Rho would jump from 0.44 to around 0.63……excellent…..and the Flames would still be nowhere near our preseason rank. The three main reasons for this dampening are the net penalty margins of Jooris and Hamilton (which actually is NOT a surprise), the decline in scoring from Monahan and Giordano (who signed a big contract in the offseason), and inconsistent goaltending. If these things turn around (and maybe they have), and the players mentioned have demonstrated exceptional talent in the past, our western Rho should improve.
As for Tampa Bay, suddenly the team who was amazing 5×5 last year and has a goalie this year with a GAA around 2.0, is actually having trouble scoring despite all their fire power (24th in the NHL at 2.37, while they were first in the league in 2014 at 3.16 goals/game). TB’s 9th place showing, which we do not think is going to persist (5-4-1 in last 10 games), is even more startling to us than Calgary’s poor start. The injury to Palat has been impactful, but the scoring drought can be placed squarely at the hands of Stamkos (16 goals, 28 points), Kucherov (13 goals, 27 points), Johnson (who was on FIRE in the 2014 playoffs, 4 goals, 11 points) and Callahan (5 goals, 13 points). Over their careers, all of these players have performed substantially better than they have so far this season. Maybe they are lagged from their deep Stanley Cup run last year? While TB has impacted us negatively, we do not believe that the team is a 9th place entity (although we had them 1st in the conference preseason). If they were to swap places with Washington for example, who we did not see getting off to the best start in franchise history, our eastern Rho would be over 0.80!
To summarize, we think this Advanced Bracton may have a decent degree of merit and usefulness as a predictor of team season outcomes. Our results so far have been encouraging, and could be wildly so if our outliers begin to perform in a manner approximating our analysis. The rest of the season is upon us, and we will keep you updated as it progresses. We hope to provide a midseason AB league-wide or at least some commentary on what the keys to successes and failures have been to this point for each team.