Mishkin’s Musings: The Wide World Of Stats (Again – Three Years Later)
NHL game sheets contain a variety of statistics, all of which are accessible on www.nhl.com. Each game sheet includes more than just the goals, penalties, power play numbers and player ice time. You’ll also find individual and team hits, giveaways, takeaways, blocked shots, faceoff percentage and others.
Last week, after the Lightning’s 1-0 loss in Columbus, some of those stats surprised me. The Blue Jackets were credited with 42 hits and assessed only one giveaway. The Bolts were given only two giveaways. A team in a typical game finishes with 15 to 30 hits and perhaps 5 to 10 giveaways. As I broadcast the contest between the Lightning and Blue Jackets, it did not seem exceptional to me in terms of being especially physical for Columbus. Yes, the Blue Jackets finished their checks throughout the game. But no, it wasn’t appreciably more physical than any other NHL game. Nor was it essentially turnover-free. In fact, in the sequence leading to the Columbus goal, I thought the Lightning committed two giveaways in their own zone. There were many other Lightning giveaways in that second period, which was one of the sloppiest the Lightning have had this year. And while Columbus defended reasonably well in the game, I didn’t think the Blue Jackets coughed up the puck only once in 60 minutes.
With these stats, one has to take into account the ‘eye-of-the-beholder’ element. I don’t believe any off-ice official is intentionally trying to inflate – or deflate – these stats. But from building to building, the interpretation of a hit or giveaway may change. For example, the Toronto Maple Leafs have played 29 games this year, with 15 at home and 14 on the road. They’ve been credited with more than 200 more hits in those 15 contests at the Air Canada Centre than in the 14 away games. There may be some truth to the notion that a team plays more physically at home, but it seems unlikely that the Leafs’ game would change that much when they skate on the road.
This inconsistency is nothing new. In fact, I wrote about it three years ago, opining about the stats that I find useful and those that I don’t when I prep for a broadcast. I still feel the same way. So here is a look back at that column, written in December of 2010.
The Wide World of Stats
Sports statistics are a mixed bag. Some provide profound insight into the performance of a player or team, ranging from a very specific component to a more general assessment. Others are less helpful; they seem to exist only because they can exist. They don’t tell us much about how a team or player might excel or struggle.
When prepping for a game, I write down tidbits that I feel would be useful to mention during the broadcast. Conversely, I won’t bother with the ones that seem superfluous.
A couple of weeks ago, Erik Erlendsson of the Tampa Tribune wrote an excellent column questioning the accuracy of some of the NHL’s “real-time” stats, such as “hits”, “giveaways” and “takeaways”. He pointed out that, from building to building, there appears to be a lack of consistency in recording those stats.
The stats subject had been on my mind for a while, even before Erik’s piece appeared. I’m not referring to the biggies, like goals and points, but some of the less prominent ones.
I spoke with Benny Ercolani from the NHL office and he forwarded to me the official definitions of hits, giveaways and takeaways. In the course of our conversation, we agreed on several points. First, the off-ice officials have a very difficult job. Tracking these statistics over the course of a fast-paced game is challenging (I know I couldn’t do it). Second, there is a subjective element to recording those three aforementioned stats. One must factor in the ‘eye-of-the-beholder’ element. Third, he said that he tells the off-ice officials to “call it as they see it”. That’s all you can ask of officials, really, whether they are off-ice or on-ice.
But for my purposes, I can only use – or not use – the final numbers that result from this process. So here are my two cents on which of those statistical categories I think are valuable and which ones the league should put out to pasture.
Hits: What is a “hit” anyway? As Erik addressed in his column, trying to answer that question presents the biggest problem with this stat. Is it a player finishing his check? If so, we’d have many more hits credited than we do. Does it have to make you say, “Ooooh”? Then we’d have far fewer.
According to the league, a hit occurs when a player applies legal and intentional physical contact to impede the puck carrier such that his forward progress is contained or he loses possession of the puck, and/or, legal physical contact is applied to a player who has just relinquished possession of the puck temporarily impeding him from rejoining the play
Reading this definition – especially the last part of it – I would classify a hit closer to the ‘finishing his check’ standard, but that’s my interpretation. I can understand how others might view it a little differently.
To me, a helpful stat tells you something about a player or team. I guess a player with a high hit total would be a difficult one to face. And a team with many hits would play an “in-your-face” style. That’s valuable information. But there are plenty of sandpaper-type players who aren’t even close to the league leaders. Ryan Malone, Nate Thompson and Adam Hall, for example, regularly pressure opposing defensemen and create turnovers with their aggressive forecheck. And they’re not doing it by sneaking up behind the D and yelling “Boo!” As for the defense, no one on the Lightning hits harder or more fearsomely than Mattias Ohlund. But none of those players ranks in the top-60 in the league. There are eight New York Rangers, however, who have been credited with more hits than Malone, who is the Lightning leader. Eight! Unless the stat can accurately reflect the personality of a player and team, I have no use for it. Verdict: Pasture.
Blocked Shots: Unlike hits, there’s no grey area with blocked shots. Either a player blocked a shot or he didn’t. It takes courage to block a shot, so we get a glimpse into the character of willing shot blockers. It’s also a crucial team stat. Obviously, those teams that block a lot of shots make it harder on the opposition to get pucks to the net. It becomes part of a club’s identity. Incidentally, as Benny clarified for me, deflected shots also count as blocked shots. How many times is a dangerous scoring chance thwarted because a defending player gets his stick in the way? Instead of going on net, the shot misses. Often, it goes out of play, giving the defending team a chance to regroup with a faceoff. I review team and individual blocked shot numbers before every broadcast. Verdict: Valuable
Giveaways/Takeaways: Here are the league definitions: A giveaway occurs when a player’s own actions and decision making results in the loss of team possession of the puck.
A takeaway occurs when a defensive player causes a turnover and takes possession of the puck or when a defensive player makes a definitive effort to intercept a pass attempt and takes possession of the puck.
Suppose a player makes a terrible pass in his own zone that leads to a turnover and a scoring chance against. Or in a one-on-one rush, a defenseman uses a pokecheck to stop the attack and steal the puck. Those would be credible examples of a clear-cut giveaway and takeaway. But the giveaway/takeaway is not quite like a football quarterback throwing an interception. In hockey, the puck can change possession a hundred times during a single game in a variety of different ways. Yet only a fraction of those are marked as official giveaways or takeaways. So which ones make the cut? As with hits, that’s at the discretion of the official.
What can these stats tell us? You would figure a player with a high giveaway total is prone to making bad decisions while one with a lofty takeaway total is adept at reading plays. But the player’s “giveaway” total only highlights potential bad decisions. There’s no corresponding number for how many good decisions he makes with the puck. (At least in football, the quarterback offsets his interceptions with other figures, such as completion percentage, yardage and touchdowns). Detroit’s Pavel Datsyuk is usually among the league leaders in takeaways; not surprising, given how crafty and deft he is with his stick and how well he anticipates plays. But as a whole, I don’t get the sense that these stats fully reflect the decision-making or play-reading capabilities of a skater. Verdict: Pasture
Faceoff Percentage: It’s an individual skill, although some faceoff wins are “team” wins; in other words, the actual faceoff between the centermen is a tie, so it’s one of the wingers actually getting possession. But good faceoff men win plenty of draws all on their own and their contribution to their team is enormous. Faceoff wins give teams puck possession, which they covet. Washington’s David Steckel leads the league in faceoff percentage and head coach Bruce Boudreau typically puts Steckel on the ice to begin a penalty kill. The likelihood is that Steckel will win the faceoff and the Capitals will get an immediate clear, burning off the first 15 to 20 seconds of the penalty. This is a useful team stat as well; a team’s overall faceoff percentage, which includes the “team” wins, is a pretty good barometer for how it will fare during a game. In advance of a broadcast, I always look up the faceoff percentage of the Lightning’s opponent, as well as where the individual centermen rank. Verdict: Valuable
Shooting Percentage Leaders: In the daily league-wide stat package, a whole page is dedicated to the individual shooting percentage leaders. Ugh. The league should save the paper. There’s no mistaking the accuracy (no pun intended) of the numbers, since the math is so straightforward. A player with ten goals on 50 shots has a 20% shooting percentage. Here’s the issue. Usually, the leaders in this category have both modest goal and shot numbers. The reality is that most of these players wouldn’t maintain the high percentage if their shot total appreciably increased. Shooting percentage isn’t the same as baseball’s batting average; in baseball, all everyday players rack up lots of at-bats (the equivalent of shot attempts), so players’ batting averages become apple-to-apple comparisons.
This year, there have been two exceptions on the shooting percentage leader page: Steven Stamkos and Sidney Crosby. The league’s top two goal scorers rank in the top 10 in shooting percentage and in the top 15 in shots attempted. They are both accurate and prolific, which makes their percentage numbers very impressive. But as I mentioned, they are the exception, not the rule. The league doesn’t need to list the top 50 “leaders” in shooting percentage every day. Verdict: Pasture
Average Time On Ice: This is a great statistic. First of all, looking at a team’s TOI leaders will indicate which players on a team are the coach’s “go-to” guys. Second, it speaks to a player’s conditioning and versatility. The leaders in ice time are playing in all situations: even strength, power play and even penalty kill. Of course, the ice-time figure can indicate team depth, too. When the ice-times are well-balanced on a team, it means that the coach is rolling his forward lines and defense pairings regularly throughout the game. By the way, speaking again of the off-ice officials, I’m amazed that they are able to track this stat so accurately. In each game, they need to monitor 36 different skaters jumping on and off the ice. Plus, they can’t just rely on line combinations, which often change over the course of a game. Anyway, Time On Ice is another stat that I review before a broadcast. Verdict: Valuable
Zone Time: Wait a minute. This isn’t a league stat. Not anymore, anyway. The league used to track zone time, but discontinued it several years ago. (It was already gone when I joined the Lightning in 2002). Please bring it back! Puck possession, as much as any other stat, tells the story of how a game is playing out. If the puck is in one defensive zone for the majority of the game, then the attacking team dominated play. All teams want to have the puck, particularly in the offensive zone. The best defense is a good offense, after all. The other team can’t score if it doesn’t have the puck. Verdict: Valuable – Take It Off The Pasture
I suppose that how people view statistics is a bit like the officials that track them – depending on the stat, it can be in the eye of the beholder. Maybe others see value in some of the categories I’ve dismissed. Some might not care about the ones I like to look at. Perhaps a wide spectrum of opinion explains why the league deems it important to make them all available.