What is the value of a game-tying goal?

There's a wealth of statistical data just a mouse-click aways these days, but I'm an old-fashioned guy who still likes his newspaper fix. Every Tuesday during hockey season the Edmonton Journal devotes an entire broadsheet page chock full of agate type providing (almost) up-to-date stats of all 800 or so NHL players. Following a practice of many years, I like to linger on this page for a few minutes every Tuesday, longer if I get some whizzy little project in mind.

There are no fewer than a dozen columns of data to consider. Besides traditional "boxcar" stats -- games played, goals, assists, points, plus/minus, penalty minutes -- there are columns for shots and shooting percentage, plus a breakdown of goals: powerplay goals, shorthanded goals, game-winning goals, game-tying goals.

What's that you say? Game-tying goals? Didn't they go the way of the unicorn when Gary Bettman waved his magic wand after the lockout and did away with ties once and for all?

The answer is yes and no. The "GTG" column has been a waste of space for lo these five seasons now, a long, long column of unbroken zeroes. Yet goals which tie games in regulation time are more valuable than ever before ...

The old definition of a GTG was the last goal scored in a tie game. It was always a huge goal, every single one of them worth a standings point. As a group they were considerably more important than game Winning goals, which were worth a standings point when they converted a tie to a victory (in essence, in a one-goal game) but often were just another goal in a blowout that the hero's team would have won anyway. There was at one time a school of thought that GTG were the best measure of a "clutch" player, whatever that is.

But in the modern game of the Bettman Point, especially Ver 2.0 which sees every game taken to a decision, a game-tying goal -- defined here as the last regulation goal scored in an overtime game, a.k.a. a Regulation Tie -- is more valuable than ever. It not only secures a standings point at the end of regulation, it provides the hero's team with a 50% chance of taking a second point by winning the game in overtime or, increasingly likely these days, the shootout. That 50% might vary from team to team but on a league-wide basis it's an immutable average that not even Gary Bettman can screw up.  

By this definition a GTG is worth 1.5 standings points. It is the ultimate Big goal in the regular season.

Meanwhile, in Gary Bettman's NHL a game winning goal has been effectively devalued. The decisive goal in a one-goal win translates to a standings point, sure, but without that GWG the game would remain tied and the team would continue to have a 50% chance of securing the win later. They would simply return to what I call the Schrödinger's Cat phase of the game, where all you know for sure is that eventually the box will open and the cat will be alive or dead, or in this case that third point will be awarded to the good guys or the bad guys. It's a coin flip which will come up heads for exactly half of the teams. Mathematically speaking, a GWG is the difference not between a win and a tie, but between a win and a tie-with-half-a-chance-to-win. Thus by my math a GWG is worth only 0.5 standings points.

By this logic, the difference between GTG and GWG is 1.5 - 0.5 = 1.0 points. Interestingly, that is exactly the difference between a game decided in regulation (2 points) and a game decided in overtime or shootout (3 points). The tying goal in effect creates the third point!

And make no mistake, while tie results are a thing of the past, there are more tie games than ever. As discussed in this piece late last season, NHL coaches are not dummies -- well, most of them aren't -- and have learned that it is much more important to not lose in regulation than it is to actually win. Get to overtime and a loss in the mini-game will be rewarded like a tie, but a win will be rewarded as a win. Thus a system that effectively awards 2 for a regulation win, 1.5 for a regulation tie, and 0 for a regulation loss. Note how this point distribution exactly mirrors the values of game tying vs. game winning goals as derived above.

Flip that logic around, and a game winner late in regulation is 3x more devastating (-1.5) to the team that allows it than it is rewarding (+0.5) to the team that scores it, so both squads govern themselves accordingly. The New NHL is supposedly about offence, but rewards in a very significant way teams who play conservative, close-to-the-vest hockey to preserve a tie; which not infrequently describes both teams down the stretch of a deadlocked affair. Post-lockout some 23% of all games have reached OT, an unprecedented level of 60-minute ties. So far in 2009-10, this percentage has surged further to 26%; over a quarter of all games are tied after 60 minutes, making the average game worth 2.26 standings points. The mean Points% has jumped anew from .557 to .566. The old standard of .500, once a mathematical constant of some significance, has faded into complete irrelevance -- .500 teams are no longer mediocre, they are downright bad.

This competitive imbalance could easily be resolved through the introduction of what I have long called the "Three Point Must" system. If every game were worth three points -- 3 and 0 for a regulation decision or 2 and 1 in extra time -- the risk and reward  would be exactly balanced. A late goal in regulation would be worth just as much (effectively 1.5 points) to the team that scored it as it cost the team that allowed it. A team playing it safe and guaranteeing its one point would be giving up its shot at taking all three points. The Three Point Must is an elegant, progressive solution which has been suggested but gruffly rejected by Gary Bettman and his antediluvian cronies like Brian Burke.

As a final indignity, not only does the NHL not recognize critical game-tying goals, it doesn't properly recognize game winners either! Games decided by shootout -- as fully two-thirds (52 of 77) of three-pointers have been so far in 2009-10 -- designate but do not credit an official game-winning goal scorer. Think about this: there's a game-winning goalie, but no game-winning goal. Thus the GWG column of all players does not add up to the number of wins of all teams, nor for that matter does the G column add up to the number of GF displayed in the standings. This could be easily and elegantly resolved by rewarding an official goal to exactly one player in each shootout, the guy whose shootout tally decided things (à la Jakub Voracek last night, pictured up top). Such a single player is always identified in the official game summary, but his goal is unnumbered and uncredited. It would be such a simple fix ...

Not in Gary Bettman's NHL, though, where fundamental concepts like balance, symmetry, zero-sum accounting,  respect for tradition, and simple logic have been laid to waste in the pursuit of "growing the game", artificial parity, and of course the almighty buck. As a result, the league's statistics, standings, and record book are utter shambles.

Not only is the historical record of the game beyond repair, I'm even more concerned about the immediate present. Every time two teams sit on a tie score to their mutual benefit, the very integrity of the game on the ice is compromised. Is it any wonder that far too many people don't take the sport seriously?

Important note: The above rant applies to regular season hockey only. Unfortunately, regular season hockey is all the author's beloved Oilers have played for the past several years.  

Praise the hockey gods, the playoffs remain pure: game-tying goals are appropriately huge, but game-winning goals are Everything.