Sab(re)metrics: What is GVT?

As the NHL Lockout continues, I would like to briefly discuss the definitions and derivations of various hockey advanced statistics that have been used in previous Sab(re)metric articles. Today, I'll start to explain GVT, or Goals Versus Threshold.

With this wonderful lockout, it is safe to say that hockey blogging has become rather difficult. The promising news of having a shortened season that I have heard in the last few days has seemed to dissipate once Bettman and Fehr re-entered the negotiations, and now I am not really sure when this all could be resolved. Either way, I shouldn't not talk hockey; the sport is way too important to me.

Therfore, going forward (and suggested by Zach), I want to make a series of articles that explain what some of the advanced statistics are in the NHL. Some will be ones I have used, and some will be ones that I want and hope to use in the future. Hopefully, these articles could morph into an overall definition page for advanced stats on the DBTB site; that way, you can always glance at a quick list of definitions instead of wondering what the heck I was blabbing on about in an article. For today, I will begin to re-introduce GVT, or Goals Versus Threshold.

Since GVT is a relatively complicated stat, I would like to explain it in various sections as listed below. For today, section 1 will be discussed.

1. A general overview of GVT(this may seem like a review to a few readers)

2. Calculation of the stat (How to derive OGVT, DGVT, and SGVT using equations)

3. Some GVT analysis of the Buffalo Sabres 2011-12 Season


Goals Versus Threshold (GVT) is a stat developed by Tom Awad of Hockey Prospectus in July of 2009. Overall, it is quite a complicated statistic; one that Tom Awad explained in three separate articles (1, 2, and 3) when it was first introduced. To be brief, the stat itself was meant to do two things:

1. A way to compare various players in the NHL, independent of position

2. A way to quantify the importance of defense

What this means is that that there were already ways of comparing goalies (GAA and SV%) and comparing offensive players (goals, assists), but there was no way to say how much better a goaltender was than another player on the ice. Also, we always knew that good defensive players existed (i.e. Jay McKee), but there was no good* information that could be used to compare those players to other defensemen and offensivemen. Out of those predicaments, the GVT stat was born.

To compare various players, Awad developed the stat to differentiate the amount of goals a certain player is worth when compared to a replacement level player (the threshold). In this definition, a replacement level player would be one that is a 13th forward or 7th defenseman on a NHL team; in other words, a fringe NHL player. This is quite similar to the VORP and WAR statistics in baseball, where they also find a player's value by comparing their "worth" against that of another replacement-level player.

It should also be important to note that the GVT stat a summation of three separate variables, the offensive GVT (OGVT), the defensive GVT (DGVT) and the shootout GVT (SGVT). The latter variable has only been used since 2005, since that is the first season a shootout was used and could be quantified. The calcuation of each of these variables will be discussed more in the next article.

To finish off today's overview, I added on Tom Awad's general explanation of GVT that he included in his original 2009 article. I will add his links, as well as other interesting uses of GVT to a definition article in the future. If you have any specific questions on the stat, just ask below.

  • GVT is measured in goals. This makes it a convenient unit that hockey fans are already comfortable with.
  • GVT compares hockey players of all positions and over any period of time.
  • GVT only uses statistics that lead directly to goals. You cannot incorporate goaltender wins into GVT, because they are not a measurement of goals prevented. However, if you can rationally explain what are the odds of a faceoff win (or loss) leading to a goal or goal against, it would be possible to incorporate faceoff wins and losses into GVT, though I have not done so.
  • GVT has built-in accounting. The sum of player GVTs on a team equals that team's GVT plus the replacement level. This is essential, as player statistics often come with caveats. Kovalchuk scored 43 goals, but he doesn't play defense and his team isn't good. This makes it much easier to measure "how good would this team be replacing player A with player B?" It is also essential in that player success is correlated with team success, which after all is the entire point of the sport.
  • GVT automatically normalizes for the strength of the league. When looking at player statistics from different eras or different leagues, it is often difficult to know if a player was good or not. For example, for the last few years in the Czech Extraliga, a save percentage of 0.920 has been average or below average, while in the NHL today a save percentage of 0.920 is pretty good, and in the NHL 20 years ago it was unheard of. Similarly, a 50-goal season in 1982 was less impressive than a 40-goal season today. GVT takes all of this into account, giving you a single number that doesn�t need any further interpretation.

* I place good in italics because one could argue that +/- is such a stat, but I would say that is a very poor statistic to use to compare the defensive capability of a player. This was a tangential discussion to the article....hence the asterisk. The "Cons of +/-" will be discussed in a future article.