Probable, ‘Cause is a new feature where the Hockey Graphs staff give their predictions about the big unknowns in the hockey universe today. We’ll offer a question and give the reasons why it will and won’t happen, and then our estimate of the probability that the answer will wind up being yes. At the end of each question you’ll have the chance to submit your predictions as well, and we’ll review the group’s answers in the next edition of Probable, ‘Cause. This week, we’ll look at Braden Holtby’s odds of entering the NHL record books, whether Erik Karlsson can keep up his hot hand, and how nervous broadcasters should be at the prospect of an all-American playoffs this year.
Will Braden Holtby surpass Martin Brodeur’s single-season wins record?
Why it’s gonna happen: The Washington Capitals are currently sitting atop the NHL standings with 80 points and a 78 PTS%. It won’t come as a shock to tell anyone that the Capitals are a good team and that Braden Holtby is a good goalie. So it shouldn’t come as too big of a shock that Braden Holtby has a legitimate shot to crack the 48 win mark. The Capitals have won about 3 out of every 4 games so far this season and if they keep up that pace they would be on track to amass a team total of 61 wins.
Why it’s not gonna happen: 48 wins is a lot, not just for a goalie. Only 7 NHL teams got 48 wins total last season. A lot of factors need to keep going right for Holtby to break this mark, but the two most important ones are health and performance (both individually and team). If either one of those factors falter it could be a quick end to any record breaking talk.
Actual Chance: 35.9%
Does that seem high or low? I’m not really sure. I built a simulation to help with this prediction. I won’t bore you with all the details but here is a brief summary. Determine how many games Holtby will play by randomly selecting a number of games from a normal distribution with a mean of 26 games (if he continues the same pace of games played so far this season). The theoretical max was 31 games played and theoretical min was 21 games played. Then we simulate the remaining schedule using xG power rankings. However, since the Capitals are 51 games through their season and are still outperforming their xG ranking we can make some crude adjustments to account for this. The first adjustment was simply an extra win per season simulation and the second adjustment was to account for Braden Holtby’s performance (based on his year so far). We can then run this simulation 1000 times and see that Holtby falls short 49.4% of the time, finishes with 48 wins exactly 14.7% of the time and beats the record 35.9% of the time.
Will Erik Karlsson finish the season in the top 5 in scoring?
Why it’s gonna happen: Karlsson is currently tied for second in league scoring with the guy who won the Art Ross Trophy last season which is probably all the perspective you need on how insane his current pace is. He’s also seven points up on the guy in 6th (with three games in hand). Those two facts alone are probably reason enough for why it’s gonna happen as he’s scoring like an elite forward and has created a big cushion for himself. What he lacks in per minute efficiency he makes up for in volume of minutes played, both thanks to being a defensemen. He’s currently on pace for 91 points, but as I tweeted on Sunday he likely finishes closer to 86 points based on a regressed-weighted average of his points per 60 during his last three seasons (similar to Dom Galamini’s method here). That puts him in fourth behind Patrick Kane, Jamie Benn, and Tyler Seguin and five points up on Sidney Crosby for fifth.
Why it’s not gonna happen: Karlsson is currently scoring at 2.34 points per 60 which is slightly higher than his career norm (although he did put up similar numbers in 2011-12). Based on his last three seasons, I’d expect that to be closer to 1.94 which would put him at a 76 point pace for the rest of the season. Still great, but if he scores at an even lower pace or he goes on an extended cold spell, the door is wide open for any number of the talented forwards behind him to swoop in and pass him. I mean, he can’t keep scoring like this forever, can he?
Actual chance: 80%
It’s a difficult question because of how many players fates are intertwined, so I decided to figure out a point cutoff for the top five, or how many points will sixth place likely get, no matter who it is. I looked at the probability of getting 75 to 85 points for each member (excluding Karlsson) of the projected top 15 based on their projected scoring pace for the rest of the season. I added them together to figure out how many players would be expected to get a certain amount of points. Where that intersects with 4.5 players is the expected point cutoff: 81 points. With 29 games left, Karlsson will likely play around 829 minutes and if he’s expected to score in 3.2 percent of those minutes, the actual chance is how often he’ll get 81 points or more.
Why it’s gonna happen: Being tied for second in league scoring at the start of February has worked out pretty well for those who have found themselves there over the past 5 years. Since 2009-2010, 59% of players in the top 5 on February 1st have finished the year in the top 5, with the odds jumping to 73% if you were sitting in the top 3. Even at the individual level, Karlsson also has both key ingredients for putting up big point totals: boatloads of talent, and a ton of ice-time. While the dynamic defender has always played significant minutes for the Sens, 2015-2016 has seen him push this to levels he’s never seen before, with his average ice time up more than a minute per game since last year. With the absence of other offensive options on the Sens blueline, we should expect this trend to continue going forward, with Karlsson maintaining or increasing the number of minutes he plays as the season goes on and the Sens get more desperate for offensive contributors. And although his scoring is above his already inflated totals from years past, Karlsson’s rate of scoring (in Pts/60) is more or less in line with his historical averages, with his 5v5 Pts/60 of 1.8 only sitting slightly above his 2011-2012 total of 1.6, and his powerplay rate actually below his career-to-date average (he only has 1 powerplay goal all season, believe it or not). With the return of Marc Methot likely to free up Karlsson to take even more chances (his 5v5 Pts/60 with Methot has been 0.3 points higher than without him over the past 2 years), there seems to be little indication that EK65 will slow down his pace in the near-term, making a top 5 cutoff in the low 80s look mighty achievable.
Why it’s not gonna happen: Defencemen do not finish in the top 5 in scoring in the modern day NHL. This isn’t just a statement I’m making up to be dramatic, it hasn’t happened since the league was playing with 21 teams, with Paul Coffey being the last blueliner to accomplish the feat all the way back in 1985/86. A generous estimate puts the overall odds of any defencemen finishing in the top 5 in scoring at around 5%, which isn’t much to start with. Perhaps more troubling though is that more than a third of his points to date have come from secondary assists. While secondary assists obviously count towards his overall total, they historically have not been a good predictor of future scoring, particularly for defencemen. If we focus only on primary points (goals plus first assists), Karlsson drops down to a tie for 12th in league scoring. It’s not guaranteed that Karlsson’s secondary helpers will vanish overnight as they’re also subject to a fair amount of scorer bias, but it is a cause for concern heading down the stretch.
Actual chance: 60%
5% is obviously too low, and even if you expect his secondary assist totals to tail off, he’s just too integral to the Sens offense to believe that he won’t remain heavily involved in almost all of their goal scoring. If we move slightly back from the 73% of top 3 skaters that have ended in the top 5 we can hedge for the fact that his pace might slow down and that blueliners usually show more erratic scoring patterns. 60% puts us right around the level that a top 5 scorer on Feb. 1 would maintain their position and feels right in this case.
Will a Canadian team make the playoffs this year?
Why it’s gonna happen: If a Canadian team’s gonna make it, who is it gonna be? The Oilers are as bad as ever and McDavid’s return will hurt their draft position but he can’t drag them single-handedly to put up the 55% point pace it’s going to take to take them from the bottom of the league’s worst division. Vancouver are playing Derek Dorsett above the Sedins when trailing so I think they’re comfortably in the “we mean to draft well” camp this year. Ottawa are dressing at least and sometimes at most one top-four defender nightly and noted offence-generators Zack Smith and Alex Chiasson routinely get top-six minutes. Too many of the punches thrown land on their own face to have any to spare to land on their playoff-race opponents. Toronto appear determined to bury many of their best players in the Marlies and have more-or-less devoted this season, openly and from the outset, to trying to showcase virtually an entire roster of rentals. If they made the playoffs it would be entirely in the teeth of their forced obliviousness. Winnipeg have only ever threatened for a playoff spot when they can overcome their own goaltending woes, and Hutchinson seems like the new Markstrom-style bust, at least this year. Can’t make the playoffs from the penalty box, either, though the Jets can’t seem to help themselves. Montreal are imploding spectacularly and while they can’t possibly keep losing forever, betting on the Habs these days is like trying to catch a falling knife. The best hope for the country’s teams, at this moment, is Calgary, who are ruthlessly outshot nearly nightly, dress Deryk Engellend most nights, and deploy a notorious goaltending tandem of Jonas Hiller (88% save-percentage) and Karri Ramo (91%), with Joni Ortio (87%) for depth at the position. Long story short: None of the teams has the talent, so there’s no reason to think any of them will make it.
Why it’s not gonna happen: The easiest way to make a sure bet not a sure bet anymore is to try to parlay seven of them together. That’s giving yourself an awful lot of rope to get tangled in and every year one team (or several) goes on the kind of half-season tear that would propel any one of the Canadian teams into a wild-card spot. That sort of luck happens to Canadian teams, too — just ask Ottawa about out-of-nowhere goaltending or Calgary about how often you can pull points out of the fire with the goalie pulled.
Actual chance: 39%
Once you’ve got a fancy-pants simulation harness set up around a fancy-pants prediction model, you can ask it whatever questions you like–flip all of the weighted coins for all of the games yet to come a million times and measure the number of times any particular happenstance happens to stance, and that is just what I’ve got and that is just what I did — 39 per cent of the time, no Canadian teams made it through.
Why it’s gonna happen: As I write this, Montreal is only three points out in the East but with three more games played than the Penguins, who hold the second Wild Card spot. That’s a big gap to close even with your starting goaltender, and there’s no indication Price is coming back any time soon. The Sens are even farther back and would need to leapfrog three teams (not including the Habs) to claim a spot and that’s near impossible in the era of the lower point. The Leafs? Well, let’s just say that if they come anywhere even close to making the playoffs, something went very, very wrong with Shanahan’s plan.
Meanwhile in the West, it looks like the California teams have finally got their act together and have locked down the three playoff spots in the not-so-hapless-now Pacific. That leaves the four Canadian teams now vying for a Wild Card spot. Sure, the Avs probably won’t hold on, but Minnesota won’t keep slumping either and Arizona has shown that they can outplay their lousy goaltending.
Why it’s not gonna happen: The only way it doesn’t happen is if somebody goes on an insane run in the next month. I can’t see that happening for any of the Western teams. And we know it won’t be the Leafs. That leaves either Montreal or Ottawa. I’m not sure Erik Karlsson can, for all everything else he’s done this year, can carry the Sens there on his own, but it might be too soon to count the Canadiens yet. Even without Carey Price they still have a higher goal differential than all the teams they would have to get past, including the Red Wings, who currently hold the other Wild Card spot. So if they can just get some steady goaltending for a stretch, they just might be able to close that gap and sneak in. Especially if they can maintain contact until Price comes back. As I started with, it will have to be an insane run, but as they say, “doing the same thing over and over and expecting a different result is the definition of insanity.” And that’s exactly how you model predictions, so…
Actual chance: 43%
That’s the probability as simulated by DTMAboutHeart using his xG model. If it was up to me, I would have just gone with The Price Is Right strategy and pegged it at 40% just to make sure I beat Micah. But we’ll go with the math this time.
Hockey Graphs 