Leafs Projections: Playoffs?

Playoffs? This team? 

Yes, this team. They’re in the position right now and some people don’t seem to believe it or even understand how. The Leafs are not as bad as most people make them out to be after every loss and is better than last year’s version.

In the summer I developed a pretty simple algorithm for playoff success that combines possession, special teams and goals (methodology here) which I talked about in a post that looked at why the Leafs are nowhere close to contending for the Cup. 

Basically, they’ve been really bad in the past, but we already knew that right? This year is their best mark ever, and there’s reason to believe they can continue at that level of play.

The method itself correlates pretty well with points, and the better team has won 77 percent of playoff series over the past five years. I’ve been doing some weekly power rankings and projections at my own blog every Wednesday (here’s this weeks) based on my playoff output projection (POP) and Justin has let me shameless self-promote my work here.

Screen Shot 2014-11-26 at 2.18.16 PM

After Florida’s game last night, every team has officially passed the 20 game mark so I figured it would be as good a time as any to show where the Leafs are at this year. After the 20 game mark teams start to vary less and usually stay consistent for the rest of the season. At the 20 game mark last season, points achieved and points projected by POP had a correlation of 0.55. 

The Leafs played their 20th game last Thursday against the Lightning, a huge 5-2 win that brought them to an even 10-10 record with 22 points thanks to shootout losses. It’s a 90 point pace which is pretty close to their projection for the season which is 93 points.

For the projections I use an adjusted version of the algorithm because it’s still early in the season and teams are more spread apart than they would be at the end of the season. So I looks at how far teams are away from average considering the distribution and compare that to past seasons.


It’s still early and a lot of things can change, but the reason for the Leafs’ playoff projection is the same reason most people thought they could squeak into a wildcard spot in the summer: improved, but still not good, puck possession that’s offset by good goaltending, and good special teams. 

Bad puck possession buoyed by goaltending and special teams was the Leafs recipe for success in the lockout shortened season, but this time around the possession is much better so there’s hope that what we’re seeing is real and not a mirage.

Puck possession is the most important part of the algorithm and it’s where the Leafs have made the biggest strides and why there’s reason for optimism with this team. The Leafs puck possession, while still mediocre, has improved a lot year-over-year.

There’s still much to be desired, but they’re about break-even in close situations which is a big step up from last year’s woeful mark. You can probably attribute that to the analytics department as well as the forward depth added over the offseason which has done wonders for the bottom of the lineup.

The power play has been very good so far at generating shot attempts, but the results haven’t been there recently and they’ve run a bit cold. That’ll correct itself over time. Their PK is top ten, but it’ll probably regress a bit because it’s mostly due to a high save percentage. Those two should cancel each other out, but overall the special teams units have been decent.

On top of that, the goalies aren’t stealing as many games as they’ve had to in seasons past, and have actually not been their usual spectacular selves. That should bounce back up as well with the talent Bernier and Reimer have. 

Those three skills are all a part of POP and can still be volatile this early in the season, but in the Leafs case there doesn’t seem to be a big concern for regression.

And really, those three things should be enough for a playoff berth. With the Leafs calling the Leastern Conference home, it’s not at all far-fetched to think that could mean playoffs this season. They’ve improved where they needed to and still have huge strengths in other key areas that’ll definitely put them in the mix. After all, there’s different ways to skin a cat.

But despite the positive changes, the way the Leafs are playing still makes them an average team. It may be good enough to make the playoffs, but it won’t be good enough to compete with actual contenders, and that should be the goal. 

They’ve made some positive gains, sure, but the Leafs still have a long way to go. Getting to the playoffs is baby steps to the end goal, but this team still isn’t good enough.

  • Benjamin

    Let me get the premises out of the way, so they’re in the open and up for debate.
    1) We can’t contend with Bozak, Kadri and Holland as the top 3Cs.
    2) The pipeline of alternatives at the position consists of Nylander and…
    3) Good luck acquiring a long term fix at C through trade or free-agency.
    4) The Leafs are a cap team now and for the rest of time, by that fact alone they will not be in line to get a franchise C in the draft.

    Now, on to the crack-pot scheme: What if the Leafs really committed to converting JVR to centre?

    I know, I know, its been tried. But I’m not talking a 10 game trial because others are hurt. I’m talking about a systematic effort to teach him over the course of the year and ease him into it.
    1) He’s a solid, two-way player already seeing time on the PP and PK.
    2) His FO% is 44% from 2010 to present. Yes that’s not great but given his size, awareness and the hands he’s shown in the front of the net, I’d bet money he could improve pretty readily in that department.
    3) He’s not completely without experience at C, with the short trial the other year and 2 years at the college level.


  • CMpuck

    Playoffs are the ceiling? Hooray the Leafs might be adequate.

    It’d be nice after the post Quinn Era if this team had a coherent plan, blueprint and/or identity.

    It’s great that advance stats proves what we already know about this team though.

  • CMpuck

    If Dubas was smart he would realize that this team is not going to amount more then playoff fodder and immediately commence an accelerated teardown and rebuild.

  • Jeremy Ian

    Dom, nice work.

    First, a technicality. As I read your model, you are trying to simulate projections that don’t make puck possession the sole story, thus allowing for special teams, goaltending variables etc.

    But isn’t this a more complex story that’s still basically anchored on possession? (Yes, you can have top goaltending or a vicious PP, but they cannot compensate, over the long run, for poor possession).

    So your POP correlates better with your model over simple possession, but it doesn’t fundamentally change the story: without strong possession and driving play, a team won’t see sustained success.

    (BTW: I think it’s ok to have better models than the mono-causal obsession with puck possession, so this is not a criticism; just wanting clarification on what you are aiming to achieve)

    I think your bottom line is correct. It’s likely they will make the playoffs and win some of those games, and we’ll have some fun. But they have ground to cover to be a contender. (As opposed to other disgruntled fans, I actually think this is a good thing; strong sustainable success does not leap out of nowhere).

    Your model does beg basic questions: if you are a squeaker for the post-season, do you resign Franson, Santo, and Winnik or cash them in for decent picks? These three players are reasons why the possession curve is finally bending upwards. But they are also valuable chips.

    What does a developing bubble team do, managerially speaking, when it can credibly get into the playoffs but does not yet have the pieces to push deep?

    • Yeah that’s basically what I’m doing and trying to figure how important all three factors are in relation to each other and in relation to team success. I do think possession is the most important part and that’s why it’s weighted the most.

      But if there’s two teams and one is slightly better possession wise, and the other has a better goalie and much better special teams, then I would think they’re the better team even though they aren’t as good at possessing the puck.

      So it doesn’t change the story that possession is king, I just think it adds missing parts to the equation of what it takes to win.

      Or in the Leafs cases, how much do the other strengths hide their glaring weakness. I think it’s enough to be average and to make the playoffs, but without that possession game they’re still far and away from being a contender and are probably easy round one fodder for a good team.

      As for what to do from a managerial perspective, I would’ve signed Franson long term years ago. Still do it now. Winnik and Santo probably keep to make the playoff push unless the offer is outstanding. Players will be more attracted to coming over in the off-season for a playoff team I think. Then I’d probably cut them loose, unless they take the same or similar money next season. Players like Winnik and Santo (and Booth) are available every season on super cheap deals. Maybe the market corrects itself this summer, but that’s been the inefficiency for a few years now.

  • CMpuck

    It would be interesting not only to see a team’s overall POP score, but the relative contributions of each part of the model to the overall score against the mean. For example, which teams POP score is inflated by their special teams? Which teams have a deflated POP by a bad score-close goal differential?

    Excellent work, nonetheless! Keep it coming.