In last night's victory over the St. Louis Blues, the Sharks surrendered two goals in the latter stages of the second period that cut their three-goal lead to a one-goal lead, forcing them to hang on for an eventual 4-2 win. That came a week after the team entered the third period of a home game against the Islanders with a 2-0 lead only to eventually fall in a shootout. Prior to that, the Sharks blew a 2-0 lead en route to a 5-3 loss in North Carolina and ceded third period leads in games against division rivals Anaheim and Los Angeles that led to the Ducks and Kings earning precious standings points via shootout losses. This recent spate of blown leads has understandably created some concern over whether the Sharks are sitting back too often, summed up by this comment made shortly after Jay Bouwmeester scored last night to pull the Blues to within a goal:
#SJSharks take their foot off the gas after 3-0 lead, and now it’s 3-2. Disturbing trend continues. Not the mark of an elite team.— Kevin Kurz (@KKurzCSN) December 18, 2013
I don't mean to pick on Kevin here as he's far from the only person to hold this opinion. But I think this general sentiment stems from some combination of unrealistic expectations and a lack of understanding about how the scoreboard usually influences the way teams perform. First of all, whether or not this is even a disturbing trend seems in and of itself somewhat debatable. Only once this season have the Sharks lost in regulation after establishing a multi-goal lead in a game, meaning their apparent propensity to cough up leads has yet to hurt them all that much in the standings. Despite the narrative that's been building, the Sharks have actually been a positive goal differential team this season even after taking leads, as they've outscored opponents 31-29 at five-on-five while leading. Even during the minutes for which they've held a lead of two goals or greater this season, San Jose has scored 15 five-on-five goals to its opponents' combined 14. So it's not as if they're repeatedly getting run over after opening up a lead, but there's also a clear drop-off in their results once they do so let's run with the premise anyway.
Do the Sharks sit on leads? Absolutely. With the score tied during five-on-five play this season, the Sharks have outshot teams (including missed shots for a stat commonly referred to as Fenwick) 402 to 302, earning a 57.1% share of all unblocked shot attempts in those situations. That percentage dips to 53.0 when the Sharks have a lead. The wrinkle here is that their 53% Fenwick percentage while leading is actually the best mark in the league. How is that possible when the Sharks' 57.1% Fenwick Tied is only second to the Blackhawks? The simple answer is score effects, or the overwhelming tendency for teams to get outshot, and often outscored, while holding a lead. It's the reason hockey analysts prefer looking at team performance when the score is tied or within a goal (or otherwise adjusting for the phenomenon) and its impact is pervasive: every team in the league in each of the past six seasons has posted a worse five-on-five shot differential when leading in a game compared to when the score is tied.
Why this effect exists is largely up for debate. Driving Play put together an insightful look at the marginal strategic value, particularly given the existence of the loser point in the NHL, of going into a defensive shell with a lead. To a large extent, though, I think score effects underscore the human element of sports and our natural tendency to lose some motivation after acquiring security. It's also almost certainly a function of parity in the league; it's rare that a team can reasonably expect to dominate another for a full 60 minutes and ebbs and flows are a natural part of the game. But those are just guesses. So while I suppose it's fair to characterize those ebbs as the result of a team's failure to keep their foot on the gas or put in a 60-minute effort or any of the other cliches usually employed to describe those game situations, I'm not sure it's reasonable to expect teams not to experience them given that literally every club in the NHL controls play to a markedly lesser extent with a lead as opposed to when the score is even.
I'm talking in absolutes when there are obviously shades of gray here; if a team is truly getting bludgeoned on the shot clock and scoreboard while holding a lead relative to how other teams are performing in that game state, there's reason for legitimate concern. But we're discussing the Sharks who, again, have the best five-on-five shot attempt differential in the NHL when leading this season. Considering they finished 15th in that category a season ago, they've certainly taken a huge stride in protecting leads even if the results don't entirely reflect it yet. But in a results-driven league, does their middling goal differential while leading take them out of the running for being considered an elite team as the tweet above (and many others) suggests? It's hard to come to a consensus on what teams constitute elite so I took the somewhat simplistic route of compiling data on the four conference finalists in each of the past five postseasons. Courtesy the indispensable stats.hockeyanalysis.com comes the percentage share of five-on-five goals (GF%) and unblocked shot attempts (FF%) the team earned while leading in games as well as those same numbers during score-tied situations. I've also included their overall average in each category as well as the Sharks' numbers through 33 games this season to compare.
|Team||Leading 5v5 GF%||Leading 5v5 FF%||Tied 5v5 GF%||Tied 5v5 FF%|
|08-09 Red Wings
|08-13 "Elite" Average||51.9||50.0||55.3||52.6|
There's something of a mixed bag here but, on average, these teams' five-on-five goals for percentage while leading looks nearly identical to what the Sharks are currently sporting and their shot attempts for percentage is worse. Even focusing solely on the ten Cup finalists leaves the leading 5v5 GF% unchanged. The takeaway here is that it's hard to make the argument the Sharks aren't elite on the grounds that their goal or shot differential while leading doesn't stack up well against other elite teams. Because it does.
As we'd expect, the average score-tied numbers are higher across the board and those are ultimately the ones worth paying attention to when assessing team performance. Particularly noteworthy is the case of the 2011-12 Los Angeles Kings who had the best score-tied 5v5 GF% in the league that season but stumbled into the playoffs as an eighth seed due to an awful 5v5 GF% while leading. Guess which number ended up meaning something.
The mark of an elite team isn't so much their ability to prevent opponents from making games competitive again after falling behind but in opening up that initial lead in the first place. The myth of a 60-minute effort is alluring but not really practical; every single team sits on leads to varying extents and as a result success is driven more by whether a club can consistently control play when the score is close and establish leads to sit on rather than how they perform when protecting a two- or three-goal lead. This isn't to suggest the Sharks should stop pressing all together once they have an advantage on the scoreboard but given that they possess the puck while leading more than any other team in the NHL, I just don't think this is an aspect of their game worth worrying about.