“Score effects” has become the blanket term for a concept we’re all familiar with in sports: when one team jumps out to a big lead, they often “sit back” while the other team takes it to them, and the momentum appears to shift.
It’s not uncommon to see teams like the Toronto Maple Leafs push back late after a rough start, get close to making a comeback, than say things after the game like “We just need to play like we did at the end of the game for 60 minutes,” because they’re somehow oblivious to the fact that they didn’t get better, the game got easier for them. (You’ll hear the same from teams that run out to a great start then falter – gotta play the full 60.)
Take last night’s Winnipeg Jets/Anaheim Ducks game last night. Here are the shots from a game in which the Jets led 4-0 and managed to lose 5-4 in overtime:
What we think we see is hockey’s version of the “prevent” defense, where you let a team bite off huge chunks of yardage to avoid the one big play. But it’s different than that.
What’s strange is, score effects don’t really come from a change in tactical play as it does in soccer. I’d be willing to bet that Paul Maurice didn’t walk into the Winnipeg Jets dressing room before the third period and say “Okay, this period we’re switching from 2-1-2 to a more passive 1-2-2 and I’d like you to stop pressing offensively.”
Instead, score effects seem to be derived from a number of minor influences.
The first is sheer luck. There’s more parity in hockey than just about any professional sport, and when you’re using a puck that can be tipped, goalies that can be screened, and glass with stanchions, anything can happen. So, it becomes like flipping a coin – do it enough with all things equal, and as time goes on, the breaks should even out. If a team gets a couple breaks early, they may not get them late.
The second is human nature. There’s a very small percentage of the world who earn everything they want and continue to work as hard as they can. It’s way too easy when you’re up a couple goals to stop doing the little things that I’ve defined as “hockey tough,” like winning puck races and taking a hit to make a play when you can just pull up, be second on it, and cream an opponent. Hell, that even looks better to old school coaches. “Up a couple goals and Bourney’s still bangin’ boys, follow that lead.”
A third cause, and the only thing that I believe coaches affect, stems from something like the rat food-reward experiments we used to do in my psychology classes.
With a lead, coaches are less prone to use guys who “take risks,” which seems to be a trait that goes hand-in-hand with “having talent.” The last thing a coach wants is to see Nazem Kadri make a dangle inside his own blueline, turn the puck over, and give up a goal.
What they do want, is Jay McClement to chip the puck out of the zone because, like fans, they’re less stressed out when the puck isn’t in their zone. So, it gets out, coach feels relief, sees who made the clear, and the rat has been rewarded. He wants more of that.
Players know this (even if they don’t relate it to psychology), so they try to appease coaches late in games to continue getting more shifts. What you’re doing now is basically using your less talented players more while asking your more talented players to play like less talented players, because you’re terrified of the odd moment where it’s glaringly obvious who messed up. You’ve created a plan where you’re actively giving the puck to your opponent at every opportunity.
But, for the players, it's get the puck out, get more ice. The rat—this time the player—gets rewarded.
In the D-zone? Chip it out. Here you go, have it back, now we’ll sag and try to stop you. In the neutral zone? Chip it in. Here you go, you never have to defend against us, just take it. This way when your opponent goes in and scores, there’s not a single player to point at to say “Here’s why they scored on us and came back.” Instead you get to say “They just made some plays and took it to us in the third.”
Combine all these things with the fact that the losing team is full of players that the coach is irate at for under-performing (because “losing” all-too-often means “playing bad” to coaches in my experience), so they’re turning up the pace and effort as high as they can to keep getting ice, and powerplay time, and good linemates going forward.
Score effects are why advanced stat advocates use Corsi and Fenwick “close” (tied or within one) metric instead of the overall number – it gives us a better idea of who can drive play when neither team is in a shell and both are competing for goals.
Again: I’ve never seen a coach legitimately change the system because his team had the lead. It’s all the little factors that almost guarantee a team that plays great in the early part of a game will have trouble sustaining that dominance.