Interpreting How The 2021 Sabres Fare in Projection Models Archive by Eddy Tabone - January 13, 2021January 12, 20210 With the 2021 Buffalo Sabres, there’s optimism wrapped up in the uncertainty of how the offseason forward acquisitions improve the Sabres in the standings in a strong Eastern Division. A 56 game season opens the door for higher performance variance affecting the final standings. For reference, while the 2018-19 Sabres were 10th in the conference following their 56th game (with games in hand from surrounding teams acknowledged, of course), their 63 points had them on a 92 point pace that would’ve exceeded any point total since their last playoff season (96 points) in 2011. So while a smaller sample fogs analysis, as that year’s Sabres and their 76 point pace finished looking like nowhere close to a 92 point team, when the franchise’s short and long term future could depend on playoff contention this season, the fog is the price to pay. Unfortunately, public projection models have the Sabres hovering around 56 points, which is in line with the previous season’s point per game pace pre-shutdown. How do we interpret this? Optimistic Interpretation: The Sabres have an objectively better roster than last year solely by adding Taylor Hall to their top line. Jack Eichel received Hart Trophy votes last season and is still only 24 on the ascending side of aging curves. While not an MVP candidate, Sam Reinhart in his age 25 season will get his first full time opportunity to drive a line. By the same ascension logic, while the defense remains flawed and returns the same 6 primary defenders as the season before, Colin Miller is the oldest defender at 28, so there is still an implication that personal improvement is expected for the whole group. The same goes for Linus Ullmark at age 27 and Carter Hutton following eye surgery. For Rasmus Dahlin and Henri Jokiharju, this jump will be much more substantial given their age and early career jumps, so even if the rest of Ullmark, Miller, Jake McCabe, Brandon Montour, and Rasmus Ristolainen remain stagnant as they near their peak on the curve, there’s reason for optimism combined with the improved penalty killers for less goals allowed than last year. Pessimistic Interpretation: With that, the optimism is out of my system, and we return to acknowledging that the East Division has the offensive star power to overcome a still bad, even if better, defense, and Steve Smith remains in charge of last year’s 30th ranked penalty killing team that has lost Johan Larsson (UFA) and Zemgus Girgensons (IR) before opening puck drop, and with Jeff Skinner away from the top 6 in the more recent training camp practices, the coaching staff continues to double down on defensive-first hockey that misfits the personnel of the roster. All of this brings us back centered to the expectation of an evenly balanced point per game pace in the ironic shadow of last season’s quest for evenly balanced production out of their lines and pairings. Welcome to the world of the normal distribution: a range of possibilities centered around a most frequent expected value. You’re not getting out of math that easily. Understanding the distribution of outcomes While there would be value in creating an Expected Buffalo projection model for the NHL season, I’m more interested in what the public models that have already been created tell us about the range of possibilities for the Sabres this year. The table below shows the mean and standard deviation for the projection simulations from Charting Hockey, Evolving Wild, Dom Luszczyszyn of The Athletic, and HockeyViz. Those parameters were used to sample 100,000 points from a normal distribution with the given means and standard deviations, and those points were plotted on the graph of the probability density functions below the table. Model SourceApproximate Mean Point TotalApproximate Standard DeviationCharting Hockey55.377.48Evolving Wild536.78HockeyViz56.97.1Dom Luszczyszyn557.25Distribution Parameters from NHL Projection Models In simplest terms, probability density functions determine the probability that a random variable takes a given value – so in the context of this exercise, the probability that the Sabres finish with a specific number of points. With each of the four projection simulations, we can calculate the probability that the Sabres reach or exceed a given point total. Points in 56 GamesCharting HockeyDom LuszczyszynEvolving WildHockeyViz56 (82 Point Pace)46.65%44.51%32.91%55.04%62 (90.8 Point Pace)18.79%16.7%9.22%23.63%65 (95.2 Point Pace)9.91%8.38%3.84%12.7%68 (99.6 Point Pace)4.57%3.64%1.35%3.64%Expected Probability of Reaching or Exceeding A Given Point Total Since the threshold to make the playoffs usually sits between 95 and 100 points, the probability that the Sabres make the playoffs this season sits quite low, likely not exceeding 13%. Of these four models, only Micah has a greater than 50% chance the Sabres have a better record than last season. While that’s not great, three of the four models had the projected fourth place finisher with less than a 95 point pace, so let’s look at the expected probabilities of reaching or exceeding the point totals that the fourth place teams in the Eastern Division received in each model. ModelProjected Point Total For 4th Place In The Eastern DivisionCharting Hockey Model ProbabilityDom Luszczyszyn Model ProbabilityEvolving Wild Model ProbabilityHockeyViz Model ProbabilityCharting Hockey62.3 (91.2 Point Pace)17.72%15.69%8.51%22.35%Dom Luszczyszyn63.7 (93.3 Point Pace)13.28%11.49%5.73%16.91%Evolving Wild66 (96.6 Point Pace)7.78%6.45%2.76%10.00%HockeyViz61.1 (89.5 Point Pace)22.19%19.99%11.61%27.71%Expected Probability of Reaching or Exceeding A Given Point Total Conclusion: On par with the theme, there’s a lot of ways the East could finish, and the while Sabres may be closer to the Rangers and Devils than the Capitals and Islanders, the gap is more narrow than it would be in an 82 game season. Am I confident enough to put my fist on the table and tell you I know where the Sabres will finish this season? No. Am I confident enough to give an educated guess? Also no, but if projection models are centering expectations around the same as last season with a higher ceiling offense, it’s enough to maintain interest and get to know the teams around the Eastern Division to see if the Sabres will have enough to finish ahead of them. What’s Next? Later this week, I will investigate further into the 2021 projections on Charting Hockey and HockeyViz for more hints on what to expect for the Sabres this year. Thanks for reading! Dom’s Preview Micah’s Preview Sean’s Projections Evolving Wild’s Twitter, since their projections have been updated a few times this week Photo Credit: Bill Wippert/NHLI via Getty Images