UGG Baratas online. Australia UGGs Outlet España.

Your #1 Source for Nashville Predators News, Updates, and Opinions – A Nashville Predators Blog

Advanced Statistics Breakdown: Predators vs Jets – October 20th, 2013


Oct 22, 2013


Need to understand any of the terms in the article below? Check out our Introductory Guide to Advanced Hockey Analysis

At first glance, the raw numbers from Sunday’s Predators/Jets matchup would suggest that Nashville presented one of those exceptions-to-the-rule scenarios. However, today we’ll introduce a couple of new concepts that will help to explain the disparity: score effect and close (as it relates to advanced statistics)

In an attempt to dissect what occurred in this game, it’s easiest to segment it into chunks defined by the overall score. I’ve touched on the concept of “score effect” before, but with the Predators playing in a series of tight games, it’s not a concept we have had to really account for. Essentially, the idea is that when a team takes a lead of more than two goals, the on-ice philosophy changes for both the leading and trailing team. The team with the lead is less apt to take risks for the sake of offense. They might resort to sending only one man deep in the offensive zone. They’re going to focus on clearing the puck at any cost, as opposed to creating a clean, counterattack-based breakout in the defensive end. On the other side, a team that is playing from behind is wont to gamble more. They will send their defenders deep to assist on the forecheck. Their forwards become more likely to play higher up in the defensive end, ie: “cherry pick.” All of this phenomena adds up to an increase in shots/Corsi events for the trailing team, and an associated decrease in these same categories for the leading team.

Of the 48 minutes played at 5 on 5 in Sunday’s game, only 28 were played in a “close” situation, or the opposite of the criteria for score effect: teams are within two goals of each other. So with the knowledge that the score will have a bearing on the advanced statistics outcome, how do we account for that and make sense of the numbers? Fortunately, most sites that keep up with advanced stats will break the game into situational categories. We will start with the tried-and-true 5-on-5.

As always, 5-on-5 is our largest sample size. If you were to look only at these numbers, you might suspect that the Predators had a healthy dose of luck on their side. The Jets out-Corsi’d the Predators 55 to 30, or 65% to 35%.  The Predators’ Corsi events were comprised of 19 shots, 4 blocks, and 7 misses. The Jets’ 55 events were made up of 31 shots,14 blocks, 10 misses. Rough, right? Not when we apply our newfound understanding of score effect.

28 of the 48 minutes played 5-on-5 could be classified as “close.” When we look at those totals, we see that Winnipeg maintained a much smaller advantage in Corsi-For percentage – 58% to 42%, or 29 events to 21. If you remove the blocked shots from the equation (which Matt Fenwick implores you to do, believing that shot blocking is a defensive skill unto itself), the difference becomes an almost negligible 21 to 18 in favor of the Jets. That’s quite a difference from the seeming domination represented by the big picture numbers.

The final subset of numbers that we can examine are those that fell in the 18 minutes the teams were tied. This will typically represent the most level playing field, without even the minute impact that any sort of lead could have. In this scenario, the Jets had 54% of the Corsi events, whereas the Predators actually lead in Fenwick-For, 13 to 12. Again, the water is less murky and we see that the Predators actually played a pretty solid game in getting the victory. Now that we’ve looked at three different phases of the game, we see that 26 of the Jets’ total 55 Corsi events came in a situation where score effect was at play. That’s 47%, or nearly half of their total, which again validates the Predators’ effort.


Paul Gaustad: As Barry Trotz recently acknowledged, judging Gaustad by his advanced statistics is an unreliable pursuit. While true that his 16.7 CF% and -24 C-Rel% were the worst totals on the team, his 44% defensive zone start excuses these results somewhat. For the record, his linemates Matt Hendricks and Victor Stalberg weren’t far behind. For argument’s sake, Gaustad was slightly better at 5-on-5-close, bumping up to 29% Corsi-For% and and -16% Corsi-Relative.  More along the lines of what we’re used to seeing in a game where Gaustad has a heavy concentration of defensive zone assignment.

Matt Cullen: File this under “statistical oddities:”  Cullen posted a CF% of 56 at 5-on-5, inclusive of all score scenarios. This is interesting not only because he and his linemates were the only positive Corsi players in this situation, but also because his CF% in close minutes (53) was actually marginally worse. What that means is that Cullen did not reasonably attempt to score more goals and generate more offensive chances when the Predators took a substantial lead.

Kevin Klein/Mattias Ekholm: I singled Klein and Ekholm out in the last breakdown, promising to keep an eye on their numbers going forward. They responded on Sunday with Klein having a team-best 67% Corsi-For in score-close situations. Ekholm ranked third with 59%. They did see a lighter-than-average defensive deployment, which may account for some of that improvement.

Matt Halischuk: Knowing that Halischuk doesn’t play for Nashville anymore, I was curious to see if his advanced numbers were in line with what we saw while he was with the Predators. There’s been some speculation that sometimes it’s not the player, but the team/system, so Halischuk makes a good case study. His 45% Corsi-For on the year is nearly identical to what he produced while in Nashville. For this particular game, Halischuk’s 46% is among the worst for forwards on his team. In fact, he accompanies only Bryan Little and Blake Wheeler as Jets forwards with a negative Corsi percentage.

If you watched the game and thought, “I’m afraid to see what the advanced stats breakdown has to say about this one,” you likely soothed yourself with the fact that the Predators pulled out a rare 2-goal victory. As it happens, the advanced stats, when taken within context, were actually pretty solid. With the concepts of “score effect” and “close” in our pocket, we can get a clearer picture of what actually happened in the course of the game.

Statistics Courtesy of The Extra Skater