THE FARCE THAT IS PLUS/MINUS
The most thankless stat in hockey
by Chris Botta
We called it “stealing a plus.”
After a game, a player – always a defenseman – makes a case to someone on the PR staff that he was screwed by the stat crew. The player argues he was not on the ice for a particular goal against and he should be even and not -1 for the game. Sometimes the offended player will even single out one of his teammates that he’s certain was on the ice, proving that for a few and the pride-less, hockey is only a team game until the final horn.
This is one of the many nightmares of the plus/minus statistic, one of the most useless in all of sports. Ever notice how NHL teams only bring up the stat when it conveniently spins a signing or trade in a positive light? Yes, I’m guilty as charged. The following is typical of many conversations I had over the years with Islanders general managers.
“Hey GM. Anything you want to put in your statement about how underrated our new defenseman is, since his stats don’t say much”?
“Geez, I don’t know. What was his plus-minus”?
“Last year he was plus-6 and the really good news is he was second-best on his team.”
“That’s great. Go for it.”
“For his career he’s a minus-57.”
“Hmm, okay, just play up last year and don’t bring up the rest.”
In his three seasons with Columbus, Ron Hainsey was plus-13 in ’05-’06, minus-19 the next year and minus-7 in his pay-or-I-walk year. Do you think this had any bearing at all when a half-dozen teams offered him 4 mill a year for four years? (He signed with the Thrashers for 5 years, $22.5 million).
Mark Streit was a combined minus-17 over his three seasons with the Canadiens. Do you think this came up even once when Garth Snow and Ken Morrow talked about him? The reason it didn’t matter for Hainsey and Streit: everyone knows they are good NHL players, simple as that.
The only time the plus-minus stat seems useful is when a defenseman has an off-the-charts high number. You’d be hard-pressed to say a dman with seasons of plus-40 was mediocre at his job. Nicklas Lidstrom (+348 for his career) Zdeno Chara – ignoring that minus-14 first season in Boston – Scott Niedermayer…these are really good defensemen. You don’t need an unscientific statistic to tell you that. On the other hand, there are plenty of dmen with shameful seasons of minus-20 in their past that any team would love to have on their blueline.
When I was with the Islanders we had a defenseman about a decade ago who was around plus-10 one year and minus-10 the next. He told me if I asked every coach or scout in the league they would declare he was an infinitely better all-around player the season he was a minus.
(The plus-minus stat for forwards? Goal scorers, checking centers, doesn’t matter. A complete waste of time to even look at it).
As for those dmen appealing for a plus-minus correction, so many times play would be reviewed and the off-ice official would come back with a bewildered look on his face: “I watched the play three times. It looks like the team may have been caught in a bad line change, but your guy clearly is on the ice and the player he claims should get the minus is nowhere to be found.”
There’s an easy way to identify the best defensemen on your favorite NHL team. Look at the team stats and find the column for Time On Ice. As a general rule, head coaches do not give more than 20 minutes a game to players they don’t think the world of. If a defenseman is playing less than six minutes a period, his coach doesn’t think he’s reliable in all zones.
That’s why when the GMs have their laptops open for the July 1 meat market, they are primarily looking at injury reports and TOI. Last season Jeff Finger was a plus-12, which led all Colorado defensemen but gets him nowhere except it referenced in the press release heralding his signing. Finger would not have received many offers if he played 14 minutes a game. He played 20 minutes a game – more than John-Michael Liles and Jordan Leopold – and to Toronto and the other teams making strong offers, this made all the difference in the world.
If the NHL ever decides to eliminate the plus/minus statistic I doubt you’ll hear a single complaint, even from the agents. For final evidence we present Brendan Witt, throwing himself in front of everyone and everything last season, making life hell for Sidney Crosby and other top forwards, playing in pain until he was practically shoveled off the ice by the training staff. What did Witt get for his efforts?
A permanent column in his career statistics that says he was a minus-8.
Comments.
31 Responses to “THE FARCE THAT IS PLUS/MINUS
The most thankless stat in hockey”
Please Wait
I think you could have added the fact that your best defensive defenseman is going to play against the opposing teams best forwards. And like in the days when Paul Coffey was essentially a 4th forward yet won the Norris Trophy as the best defenseman.
And what about Goals Against for a goalie?
Witt is a beast. I hope we can get good value for him before the deadline. We should be able to get 2 #2s for him hopefully. Also lets trade guerin, weight and comrie!!
could not agree more Mr. Botta. ’tis a bogus stat.
That was actually a very interesting and informative article. Thank you.
Great article CB. Brentdan Witt is a tough D-man and his +/- does not do him justice. He adds a lto of value to this team. All the more reason not to trade him…
exactly what i always thought. the +/- # is meaningless. thank you for confirming my suspicions.
I agree generally with the article. But as you highlighted with the example of Lidstrom, if a defenseman is at an extreme in the minus column, that will invariably effect him. I dont care if the D man plays 30 minutes a game, if he is -40 that is a glaring stat.
just like election polls that there is a the percentage of accuracy. if a d man is -10 there is always the a cushion of whether or not he was or was not on the ice. however with a stat like -40 even give or take the potential inaccuracies you cant get around that number.
+/- IS meaningless. I rember way back when Eric Weicnrich used DOMINATE that stat. IF a stat isn’t used by GM’s & Agents when negoitating a contract it has no value IMHO.
Witt’s like a minus 30 this season!!! Trade his statistically challenged butt IMMEDIATELY! All joking aside unless Witter is bringing back bigtime value in trade I’d much prefer him staying put here.
In Witt’s defense he doesn’t score its not his job he is there to shut down the other team’s top line. An offensive D-man should have a better +/- because he is shooting more and looking to score…
My vote is to let Witter decide his future, if he wants to stay, lethim stay… I am sure everyone agrees that he desrve that. He is adjusting to Go-Go system slowly but surely and he’s giving up his body on defence as much as the next guy. Witt is arguably aour best stay home defencemen and on offence he is not the last guy as well. Out of this whole team Comrie is #1 to get the boot. The rest can stay if they choose to stay.
I never follow stats or plus/minus stuff, I watch the guys on the ice, and I like who I like from that. Witt happens to be one of my Fav players on the team, true he’s not big with numbers/goals..etc) but every player I believe contributes in his own ways, Witt brings that bump an grind to the table, he works it, he’s a tough dude.
It takes everyone to make a team whole, and every man has his way of doin so, some score big, some play that in your face, grinder, just to help prevent goals scored on us, others skate with skill and speed an go to the net well. You need a mix to make a team solid in everyway out there, it can not be a team filled with youth, (imo) Guys like Weight, Witt..etc, they may be reaching the end of their careers, but that doesn’t mean they have nothing left to offer.
I agree that +/- is a useless stat, and even moreso in minor leagues. Look at the stats for an OHL game and I’ll pretty much assure you you’ll find a glaring mistake in the +/- stats. Sometimes they count it on PP/SH goals, sometimes, they don’t count it on ES goals, sometimes they count it for one team (ie. the +s), but not the other (-s). It is brutal. And then NHL teams will brag that their 7th round draft pick had 30 points and was a +20, meanwhile he could very well be a +5 if they kept the stat properly. I had a friend in the league a couple years ago and I tried to keep track of his actual +/- (assuming they were correct in saying who was on the ice that is) and they were off by 6 when I gave up in November
+/- is the most useless stat in hockey … CB outlined the many reasons why … I tend to use the +/- in a different way … I divide the +/- by the minutes played … and then compare those numbers with those of players on the same team … it is pointless to compare one player’s +/- with another player from a different team.
I even brought this up to a producer at Fox Sports in the ’90s who knew nothing about hockey … he was a Football / Basketball guy … he wanted to emphasize +/- in the pre-game meeting … and me being the Stat guy had to convince him that +/- was a bogus stat … I always thought blocked shots / take aways & hits were the big defenseman stats … any offensive catagory stat for a defender is gravy.
Also depends on the team … if a team scores goals by the bucket … you don’t even have to touch the puck to have a halfway decent +/- …. bogus stat … I say get rid of it … also … PP % and PK % are two other stats I have a problem with … but we’ll save that for another time.
I use the +/- only on the Sportacular app on the iPhone. The app doesn’t mention if a goal was scored on a PP or EV so I use the +/- to figure it out.
As far as the +/- goes. Any statistic that is completely subjective and dependent on things outside your control is absurd.
the only relevance it has is in this context. Eklund is -130 this season on information that has any truth to it.
MSA on Feb 6th, 2009 11:38:25 said:
Bystander…when I said I am no expert (on Wang paying for radio coverage), if you bothered to read the rest of the post, you would have seen I made the exact same point you did.
It seems I misunderstood your point. Sorry.
And Witter was a -8 on a pretty crappy team, to boot.
The -30 this season, methinks, is mostly to do with he and Gordon figuring out how to come to a compromise.
As far as I can tell it came down to the forwards helping out their D with the high forward haulin’ ass back to his blue line. A perfect example last night when Campoli lost the puck in the offensive zone the Panthers brought it back up, and there was a forward (’course I can’t remember who it as) helping out, which gave Campoli the chance to get back in the play and break up the rush.
At the start of the season that would have been a two on one quality scoring chance before Campoli got close. I know Witt’s a lot slower than Campoli, but sometimes a step’s all you need.
34 19 ISLE in NJ 22 on Feb 6th, 2009 11:57:08 said: Thanks for agreeing Bystander … examples of Stud defenseman … Lidstrom, Chara, Pronger, Rob Blake, S. Niedermayer, Phaneuf and Chris Chelios and B. McCabe in their primes .. and even S. Souray … or some retired players such as Denis Potvin, Ray Bourque & Kevin Hatcher … Those are STUD DEFENSEMEN
Of all the guys you mention I most agree about #5, Chelios, Bourque and Pronger. I would also add Scott Stevens.
And of course Lidstrom
i have to disagree here on this one. Although i dont think it always tells the tale, there are a lot of things that need to be taken into consideration, but if a players plus minus is extremely high or extremely low, its not a coincidence. This is actually a stat i always judge a player by. (atleast take it into consideration). Would you argue that Streits and Witt’s plus minus arent a tell tale story this year? I love Witt, but hes been terrible this year, although i think his play has improved a lot the past 5 games
Then whay does every team in the league keep the stat?…….SW
I’ve always liked the concept of the plus/minus stat; the problem is that it hasn’t been applied properly. Currently, it’s just too simplistic.
You could be a great defensive center and have an overall minus rating, but that could largely be attributable to the fact that you’re constantly being used to play against the other team’s top line. You’ve been on the ice for more goals against, but the net result is still better than if a less effective defensive center had been out there.
There should be some sort of a weighting system involved, taking into account the relative stats of the opposing players on the ice at the time the goal is scored. This wouldn’t be as hard to develop - particularly given what’s now available in terms of computer strength - as one might think.
It’s an iffy stat, but it’s not bogus. The important part of the stat is that it tells with reasonable accuracy how many goals for and against have been scored when the player has been on the ice even strength. On its own the stat doesn’t reveal much, but when you combine it with TOI and the respective numbers of other players on the team, it becomes more interesting. To study it in more depth you need to combine the plus/minus and TOI with analysis of the quality of competition and the quality of teammates. In NYI’s case the highest quality of competition is faced by Brendan Witt whereas the lowest quality of competition is faced by Campoli and Tambellini.
There is also the Corsi number that accounts for shots for and against when the player is on the ice. Not a tell-all stat by any means, it’s flawed like the plus/minus number, but the worst Corsi numbers on the Islanders are those of Jackman, Witt, and Sim. All in all, the plus/minus on its own doesn’t tell much, but combined with more advanced analysis it starts to tell something.
Witt is not a good defenseman this year. He is too slow and generally not a good skater. His -29 stat is a good indicator of whether he should be considered a shut down D. This year he is shutting down no one. Zednick made him look foolish last night.
The plus/minus stat is a good way to judge the defensive responsibility of the forwards. The stat is telling over the long haul. If a forward is consistantly a plus player that means he is backchecking and worry about his man the his defensive zone. If the same player is always a minus then it shows that he is not putting enough emphisis on taking care of his own end.
The stat is a mouch more valuable tool for forwards than it is for d-men….SW
Personally, I don’t mind +/-, but when you look at it, I always look at the goal differential of the team the player is on.
So the Islanders as a team are -45. Mark Streit is +2.
Without the team +/-, the stat looks meh, but in comparison you get the true picture which is that Streit is really good and the team is far better with him on the ice, than off.
Keep in mind … the +/- only represents when a player is at even strength or you get a plus if your on the ice when your team scores short handed … or a minus when a short handed goal is scored against … Some players get a huge portion of their ice time during a power play or short handed.
If player ‘A’ is on the ice for 10 mins a game at even strength … and player ‘B’ is on the ice for an avg. of 6 mins a game at even strength .. then player ‘A’s -/+ will fluctuate more wildly game in game out.
It’s a very subjective stat … I used it in conjunction with TOI and what lines they face … I don’t look at it as a seasonal stat.
Kind of like when a goalie has a .919 save % … but a 3.65 GAA b/c the team in front of him stinks defensively … which stat says the most about the goalies skills? It’s subjective … right?
Potvin (Denis) was something like +360 for his career……
Bogus?
I agree to a point. But think there are caveats. A strong player with a weak team may incur a - season true. Brian Leetch at the end of his career springs to mind But what of the player who stands out with a -20 rating on a good or excellent team? Daniel Briere may be excellent on the PP, but is essentially a defensive liability at even strength. there is no doubting his talent -but only in certain situations.