Pick Me a Winner

by steve laidlaw on April 11, 2014


Taking a look back at some of the statistical qualities of successful playoff teams.


If you are looking for an edge in your playoff pool, I just might have one for you. This isn’t a list of dark horses, Dobber has you covered there. Nor is it some overarching draft list, again the Dobber has that covered. What this is, is a mining of the data, eight years worth to be exact (every round since the 2005-06 playoffs) to try and determine what qualities were prevalent among successful playoff teams.


Determining which qualities mattered the most can help you to fill out your playoff bracket, which will guide your decisions at the playoff draft table. Of course, this isn’t perfect by any means. No one trait is a guarantee of playoff success. In the era of parity luck prevails over all. A team could have a great season and then run into a hot goaltender. Or it could lose its best player to injury. Or the bounces could go against them. Nothing is fool proof.


But if you are a fan of numbers then this study of the qualities of the recent successful teams is for you. Let’s take a look at eight different factors that may or may not have impacted a teams’ ability to win playoff series.


playoff stats1

What Home Ice Advantage?


As seen above, the team with home ice advantage has won just 57.50% of all series played over the past eight years. Once you take into account the fact that most teams who have had home ice were probably the better teams winning such a paltry percentage doesn’t bode well for the existence of a home ice advantage.


Sure enough, hockey doesn’t rate very high among the major sports with regard to the presence of a home field advantage. Surely one exists but it’s not as prevalent as you’d think and especially not when spread out over a series of games where a team may or may not actually receive more games at home than their opponent.


I wouldn’t put much stock in a team just because they get to play the last game at home.

Defense Wins Championships?


We’ve all heard the mantra but is it really all that meaningful? It turns out, probably not. At least not without a good offense to pair with that defense. Teams won only 53.39% of series where they had allowed fewer goals in the regular season than their opponents. Over such a sample that’s well within the realm of a veritable coin flip.


It is worth mentioning that over the past three seasons the team with the best defense won 72.73%. This likely owes to the trend of powerhouse teams like Boston and Los Angeles building their teams around elite goaltending and stingy defense, however it may be that defense has become more important as scoring has trended downward the past eight years.


If Not Defense Then Offense, Right?


Not so fast. While offense has been more prevalent among successful playoff teams only 57.14% of teams that scored more goals in the regular season than their opponents advanced.


2011-12 was a very bad year for top offenses. Only two teams with a goal scoring advantage actually prevailed in the 15 playoff series that were played. Certainly the Stanley Cup champion Los Angeles Kings had something to do with this. Their regular season offense was anemic but found another gear in the playoffs thanks to some timely scoring and the deadline acquisition of Jeff Carter. But the Kings only account for four of the 13 offensive upsets so they can only accept responsibility for some of the randomness.

So What Makes a Good Playoff Team?


The most prevalent quality among successful playoff teams in this data set was goal differential. The team with the best goal differential prevailed in exactly two thirds of every playoff series. This makes sense since scoring more goals than your opponent is the way you win games. The literal goal of the game is to score more than your opponent. How you do it doesn’t much matter, as long as you do it better than your opponent, or at least, more frequently.

Is Five-on-Five Play Important?


We all know that when playoff time rolls around the referees start to put away their whistles. This is especially true in the late stages of any tie game. And don’t even ask what it takes to get a ref to call a penalty in overtime. This would seemingly reward teams for their excellent play at even strength since so much more of the game is played there.


As it turns out only 61.74% of teams that had a better goal differential at five-on-five came out on top in their playoff series. There is definitely the chance of some statistical noise in here as the average isn’t so far off of the success rate we see with the generic goal differential. I certainly wouldn’t deplore a team for having a better five-on-five scoring rate but I’d probably ignore it in favour of the generic goal differential because it has been more prevalent among successful teams.


The flip side of this is not to ignore special teams play. Just because the refs tend to swallow their whistles doesn’t mean that a team with a very good power play can’t take advantage of the few chances they get to help themselves to a series victory.

What about the Fancy Stats?


Take your pick – generic shot differential, Fenwick Close %, Corsi Close % – none were more prevalent among winners than goal differential was, though they were all pretty close. You’d be a fool to ignore the value of such ratios. Consider this infographic demonstrating how teams have fared based on their Fenwick Close%. It’s pretty clear that these stats matter, just not as much as goal differential.


It is worth mentioning that there is only data for the Fenwick and Corsi figures since the 2007-08 season, which is when those stats began to be publicly tracked. It’s possible that that’s why of the three measures of shot differential, generic shot differential figured as a meaningful stat in the most series winning teams at 63.56%.

Is Goal Differential Really the Best that We’ve Got?


With the goal of finding a better method I did a two-factor analysis of every playoff series using the two most prevalent statistics among series winners: goal differential and shot differential. The chart below shows how meaningful those stats were when paired together:


playoff stats2


As it turns out, teams that had a better goal differential AND shot differential than their opponents won 75% of all series. As you’d suspect, teams used to scoring more than their opponents and possessing the puck more frequently than their opponents won consistently. Go figure.


Where things got muddy for teams with an advantage in goal differential was when they did not have an advantage in shot differential over their opponent. These teams won only 54.76% of the time.


The opposite of that were teams with an advantage in shot differential but without one in goal differential. They won the remaining 45.24% of series. That’s pretty much a coin flip if you ask me.


This isn’t exactly rocket science. When two strong teams (or weak teams) faced one another there was a decent chance that there would be a split in the goal differential and shot differential stats. So it became much more difficult for one team to impose its will on the other.

What to do about the Coin Flip Scenario?


This is where the numbers leave us and we venture into “watch the games” territory. You did watch the games didn’t you? Therefore you should have a perfect feel for how this will all turn out.


I kid, of course. But really, my advice here is no different. You are going to have to follow your gut. Or maybe you have some other metric that you are a fan of. You should be following your gut even in series where a team holds a significant advantage in both goal differential and shot differential though because you definitely want to be mindful of teams that made upgrades (or downgrades) mid-season.


A team that made a coaching change, a trade acquisition or an upgrade to overall health is a good bet to play better than their regular season stats indicate. Conversely a team making downgrades in this area is likely to play worse. Oh, and beware the hot goaltender!

So how do we Apply This Data?


This post is going up on Friday, April 11, 2014, at which point every team has one or two games remaining in the regular season so the season totals are not complete, (though this article will be updated the following Monday to reflect the year end stats.) However, here are the standings of the potential playoff teams in our two most important stats:


Goal Dif./Game

Shot Dif./Game



San Jose






St. Louis


Los Angeles




NY Rangers


San Jose


St. Louis






Los Angeles








NY Rangers




Tampa Bay






Tampa Bay


























The Big Bad Bruins


The Bruins have run away with the Eastern Conference and are the likely President’s Trophy winners. Their league-leading goal differential gives them an advantage over every opponent they might face. Only the Rangers have a better shot differential in the Eastern Conference and the Bruins wouldn’t see them until the Conference Final. If you aren’t banging the Bruin drum to at least that point I want some of what you are smoking.

The Defending Champs


Chicago’s Cup win last season was a case study in statistical dominance. They led the league in every metric I measured last season, which makes it little surprise they led start to finish. They weren’t as dominant this time around but still figure to be at least a coin flip against any opponent they might face.


The Blackhawks playoff success is contingent on their ability to get healthy in time for the playoffs to start on Wednesday. They should have both Jonathan Toews and Patrick Kane back in the lineup but at what capacity we don’t yet know.

Singing the Blues


If their recent swoon hadn’t allowed the Colorado Avalanche back into the Central Division title conversation I’d be a lot higher on the St. Louis Blues. As of right now they sit in second in the Central and headed towards a first round death match with Chicago. Good luck handicapping that series.


The good news is that the Blues can still claim the Central so long as the Avalanche don’t win out. Avoiding that first round matchup with Chicago is imperative for St. Louis but maybe it’s better if they have to face them in round one to add a little mystery to things. After all, I think everyone is picking the Blackhawks if Colorado slides back into the second position.



Colorado’s goal differential suggests a team primed to contend however that shot differential ranks bottom four in the entire league and isn’t scaring anyone. Never mind the fact they could be without leading scorer Matt Duchene for the opening round.

The Problem Penguins

The Pittsburgh Penguins really blew up the shot differential scores, as well as those of the “fancy stats”. The Penguins were a dog shit possession team until Dan Bylsma arrived but Michel Therrien was the coach for much of the two seasons where they made back-to-back Finals so they really made the case for poor possession teams having success.


Of course, the Penguins have appeared to be a great possession team ever since and have subsequently flamed out early in the playoffs with the exception of last season when they were once again a poor possession team.

They have been decidedly average this season. I have no idea what that means but if their trend of going against the grain continues the Penguins should be in for a long playoff run.

The Big Apple


Like many of the Eastern playoff teams the New York Rangers had a really tough start to the regular season. However, starting goaltender Henrik Lundqvist has rediscovered his game and the team is starting to come into its own under Alain Vigneault. It could certainly help if deadline acquisition Martin St. Louis could start to gel in this lineup to help push some already impressive numbers (for the Eastern Conference) to a whole other level.


No Eastern team has a better shot differential, which gives them a shot against anyone.

California Screamin’


The Sharks-Kings series is set in stone and the numbers seem to indicate that the Sharks are favourites here. Emotionally I am on their side but there is this nervous tick in the back of my mind that worries about the goaltending matchup of Antti Niemi vs. Jonathan Quick.


If you are looking for playoff adjustments the Kings added speedy winger and potentially lethal scorer, Marian Gaborik at the deadline to lineup with Anze Kopitar. The two have started to gel of late, which is perfect timing.

Meanwhile Sharks super rookie Tomas Hertl could return in time for the playoffs. You’ll recall he was flying high before being clipped knee-on-knee by the Kings’ Dustin Brown. Add more fuel to the fire.


Hertl returning could push Joe Pavelski back to line three giving the Sharks enviable depth.

Don’t Duck the Ducks


Anaheim doesn’t rate as well in stats like Fenwick% or Corsi% but by simple shot differential they are doing pretty well for themselves and obviously that goal differential is what you like to see. This is a team that was upset early in the playoffs last year and has a coach in Bruce Boudreau who hasn’t had a ton of playoff success at this level.


Also, who is their goaltender?


The Ducks could start any one of John Gibson, Frederik Andersen or Jonas Hiller. It’s an embarassment of riches but the uncertainty is unnerving.


With the top seed in the Pacific wrapped up the Ducks are looking at one of Dallas, Minnesota or Phoenix in round one. As intriguing as the Stars are given their quality shot differential they’d still be no match for the Ducks.


All bets are off in round two, however.



The Montreal Canadiens are locked into a first round matchup with the Tampa Bay Lightning though home ice advantage is not yet determined (not that it really matters). This does not appear to be a favourable opponent for Montreal although if Ben Bishop’s injury holds him out or even lingers this could be a different story.


Also, the Canadiens may have found another gear with the deadline acquisition of Thomas Vanek clicking so well with Max Pacioretty and David Desharnais. The numbers really aren’t on their side though.

The Best of the Rest


The numbers don’t look favourable for any of the remaining teams. There are years – like 2012 with the Kings – where lower seeded teams have favourable underlying numbers. This is not one of those years.


The Flyers did make a coaching change this season but that was after three games. Their star player Claude Giroux also suffered through early season struggles and his torrid second half pace has led to the Flyers’ resurgence so they are probably a bit better than their numbers suggest. The only team I’d really favour them against is the Penguins and that’s mostly because they have shown a real psychological edge over their in-state rivals. An opening round against either Boston or New York is extremely unfavourable.


The Blue Jackets aren’t of much interest, although good for them for making it this far. The injury news about Nathan Horton and some of their other players has them looking like easy pickings.


The Minnesota Wild fans seem to think that the Ducks are ripe for the picking though the numbers would suggest otherwise. Plus, that series is growing increasingly less likely. I guess we will have to wait and see what comes of Colorado hanging onto their lead in the Central.


We’ve already mentioned how intriguing the Stars are but they still don’t stack up well against of the West’s juggernauts by these numbers.


The Red Wings are looking awful frisky now that they are rounding into health and are getting contributions from the next wave of talent in Gustav Nyquist, Tomas Tatar and Tomas Jurco. But they’ll still be without Captain Henrik Zetterberg to start the playoffs so the team is far from perfect.

Finally, the Coyotes are likely done but since there’s still a chance we’ve included them middling numbers.




I hope you found this informative, helpful or at the very least entertaining. Obviously this study is far from perfect. I’d do well to trace my data back as far as I could to gain a larger data set and more concrete results. I could also stand to differentiate between small differences in some of the figures versus larger ones, which was not the case here. Any advantage in the numbers was deemed relevant and included.


Still, I think that the results reflect what you would have suspected to be the case anyway. By no means does the confirmation of the hypothesis validate the study but I am encouraged all the same. I look forward to hearing your comments below.


Check back in on Monday or Tuesday for my playoff picks and those of the rest of the writers on our annual DobberHockey Experts Panel.


Comments are closed.