Plus / Minus -- the Stat that Inherently Punishes Skilled Forwards

[The Panther]

Even-strength on-ice shot attempt differentials (Corsi) for:

1984 Stanley Cup Finals Game 5

1987 Stanley Cup Finals Game 7

It's fascinating to look at the vastly different results for these two games. One shows us a future dynasty team having to try to earn its place against a battle-hardened dynasty with 19 straight playoff series wins. Another shows us a two-time Stanley Cup champion with something to prove against another great (but ultimately underpowered) team.

In 84, Gretzky was the king of generating offense. As such, he had the most chances for (on-ice 13 shot attempts for), and the best shot attempt differential when he was on ice (+8, tied with linemate... Semenko). As for defense, Gretzky's 5 on-ice shot attempts against is surprisingly the second best on the team (behind only linemate... Semenko with 2). Messier's line, seeing the tough assignments with tougher zone starts, is basically mauled. Messier himself had an on-ice -15 shot attempt differential, worst on the team. The Islanders, at evens, outplayed the Oilers (32 vs 45 attempts) but lost, because everyone's time eventually has to end.

In 87, how things have changed: the best line was Messier's line (though that's probably thanks to Gretzky's concussion). Messier himself had the best on-ice shot attempt differential (+21!) followed by linemate Anderson (+20!). The Oilers outplayed the Flyers, with an even-strength shot attempt differential of 65-25. I kind of get why Hextall got the Conn Smythe.

This is very interesting, and thanks for posting it. The only thing I would say is, I think you're drawing a lot of conclusions based on one game! It would be more instructive to do this kind of thing based on an entire series, of course.

The Oilers, like all great teams, could shift roles and adapt play to beat opponents. In certain games, the Messier or the Gretzky line (or occasionally another line) would have more space to do more offensive damage. One game won't show this in balance. Oh, and about Hextall -- yes, he's a deserving winner of the '87 Smythe, but again focusing on game 7 will exaggerate the dominance of Edmonton, as they played their best game of the series in game 7. They completely ran Philly into the ground from the 2nd period on (I think Philly had something like 4 shots on goal in the last 20 or 25 minutes). Game one was similar in the third period, but otherwise the rest of the series would be much more 'even'.

But, thanks, interesting way to study it.

 

[Black Gold Extractor]

This is very interesting, and thanks for posting it. The only thing I would say is, I think you're drawing a lot of conclusions based on one game! It would be more instructive to do this kind of thing based on an entire series, of course.

The Oilers, like all great teams, could shift roles and adapt play to beat opponents. In certain games, the Messier or the Gretzky line (or occasionally another line) would have more space to do more offensive damage. One game won't show this in balance. Oh, and about Hextall -- yes, he's a deserving winner of the '87 Smythe, but again focusing on game 7 will exaggerate the dominance of Edmonton, as they played their best game of the series in game 7. They completely ran Philly into the ground from the 2nd period on (I think Philly had something like 4 shots on goal in the last 20 or 25 minutes). Game one was similar in the third period, but otherwise the rest of the series would be much more 'even'.

But, thanks, interesting way to study it.

Yeah, one-game sample sizes are not the way to go! (Otherwise, Sam Gagner is probably the best Oiler since Gretzky...) Additionally, watching only Oiler wins (the easily available media, of course) probably biases it in a certain direction.

As for other Conn Smythe "worthy" performances, I'd say Gretzky's 2 goals and 10 assists (8 primary) in 7 games is pretty impressive. The Oilers scored 22 goals, so Gretzky was involved in 54.5% of the team's offense. He was on the scoresheet every game and probably would have won it if he hadn't put up "only" one primary assist in each of his last three games (again, small sample sizes).

This article seems to indicate that the NHL might be looking into putting up a lot of older data publicly for the first time over the next several years. Of course, this kind of lets on the fact that the NHL has been tracking advanced stats in secret long before fans "invented" things like "Corsi" and "Fenwick". There's even a hint of some other stats going back to 1917-18 (though not as detailed as later stuff, obviously).

Hopefully, it's not just empty PR hype...

 

[Any Colour You Like]

David Rundblad was a +17 last year for the Hawks, one of the highest on the team. It is literally meaningless.

 

[Sens Rule]

David Rundblad was a +17 last year for the Hawks, one of the highest on the team. It is literally meaningless.

And on the hockey reference site in Defensive point shares on the Hawks he is 6th in DPS on the Hawks last year.... In only 49 games.

Somehow Rundblad contributed more defensively to the Hawks in 12:48 per game over 49 games then Hossa did in 82 games playing 18:33 minutes per game.

Also Hossa played 1:33 per game on the PK... Playing every game. Rundblad zero minutes period on the PK all season.

Marian Hossa was 7th in Selke voting. He played 18:33 minutes a game and played every game. He played a minute and a half a game killing penalties. He faced the best opposition lines most shifts.

Rundblad played as the sheltered 6th D for 55% of the season and faced the easiest match ups possible. He did not spend 1 second killing penalties.

Both were plus 17.

Rundblad had a 2.9 defensive point share. Hossa a 2.6 DPS.

How is that even remotely possible?!?!

How are the Hockey Reference point shares even calculated? Especially defensive point shares?

This shows the lie of plus/minus. But also the lie of most of hockey references supposed higher level stats. Adjusted points. Defensive and offensive points shares. All their special stats are totally garbage and useless. As much or more useless then Plus/Minus.

I use hockey reference all the time. Love the site. For raw stats only though. Any stats they are calculating themselves are totally irrelevant.

I guess I am ranting in this thread... But the use of flawed stats on the History board should not be acceptable really as an argument.

Every time I see adjusted stats used in arguments between different era players I think that all the 4 Art Ross trophies by Mikita don't even seem to register as great in adjusted stats. So much fail there.

I hope the History board can see beyond just obviously flawed stats and not take these calculated stats from hockey reference as bible.

And of course plus/minus in general.

 

[VanIslander]

A stat that punishes skilled forwards?

Oh,...

Ah,...



The vast majority of stats OVERVALUES point producers.

There are a lot of skills that don't translate directly into goals and assists.

This ain't news to anyone who's played the game.

 

[The Panther]

A stat that punishes skilled forwards?

Oh,...

Ah,...



The vast majority of stats OVERVALUES point producers.

There are a lot of skills that don't translate directly into goals and assists.

This ain't news to anyone who's played the game.

No, no. What you're talking about is that there are three stats (goals, assists, points) that are calculated based on scoring.

That's true, but what's the problem? If the stats say a guy scored 10 goals, then the guy put the puck into the net 10 times. No problem. If another guy scored 40 goals, we can generally conclude that the 40-goal guy is a better goal scorer than the 10-goal guy, right?

My point is that plus/minus can't even do this simple thing, for comparison purposes. One player might be +15, and another -1, and the -1 guy might be a much better defensive player and offensive player.

You seem to be complaining that there are a lot of stats for scoring. That's true. But I'm talking about legitimacy of a popular, officially-kept stat. Different matter.

 

[Dennis Bonvie]

No, no. What you're talking about is that there are three stats (goals, assists, points) that are calculated based on scoring.

That's true, but what's the problem? If the stats say a guy scored 10 goals, then the guy put the puck into the net 10 times. No problem. If another guy scored 40 goals, we can generally conclude that the 40-goal guy is a better goal scorer than the 10-goal guy, right?

My point is that plus/minus can't even do this simple thing, for comparison purposes. One player might be +15, and another -1, and the -1 guy might be a much better defensive player and offensive player.

You seem to be complaining that there are a lot of stats for scoring. That's true. But I'm talking about legitimacy of a popular, officially-kept stat. Different matter.

Sure. But it doesn't necessarily tell you who is the better player. (Pierre Larouche or Bob Gainey?)

Say we compare a 20 goal scorer to a 10 goal scorer on the same team. Their plus/minus numbers could be helpful in determining contribution to the team.

 

[Sens Rule]

Sure. But it doesn't necessarily tell you who is the better player. (Pierre Larouche or Bob Gainey?)

Say we compare a 20 goal scorer to a 10 goal scorer on the same team. Their plus/minus numbers could be helpful in determining contribution to the team.

Disagree entirely.

Gainey mostly had among the weakest plus/minus numbers in the dynasty Habs. Only +11 the last two years of the dynasty.

Doug Jarvis fares worse. Only +5 the last year of the dynasty and amazingly -5 in 79/80.

Pierre Larouche has mediocre plus minus numbers on the Penguins. (Flirting with even). Not bad considering how weak the team was. Then Larouche has very good plus minus on the Habs (better then Gainey and Jarvis). Then he goes to a bad Hartford team and has a horrible plus/minus.

So what does plus/minus tell us? That we didn't already know? About any of these players? I don't think anything at all.

With actual knowledge of the players we can explain all their plus/minus numbers. But without that knowledge you can't infer anything from plus/minus. In fact it just actually feeds you misinformation.

With knowledge of the players and situations you can infer Larouche was decently effective on a weak Penguins team 5 on 5 as their top scorer. He likely faced the best other teams had to offer in either playing the top line or top checking line and came out either out scoring them or not getting shut down.

In the Habs he was a sheltered offensive player... Given the easier minutes of the scoring forwards and he managed to successfully out score the opposition.

In Hartford Larouche was not as good offensively and on a bad team got badly out scored 5-5. Probably could not be sheltered on a bad team.

What can we infer defensively about Larouche over his entire career looking at his plus minus? Not really much at all. Was he merely scoring a lot to get the plus/minuses he got or was he decent defensively? You can't tell crap.

And Jarvis and Gainey have very mediocre plus minus scores and we know they were the elite of elite defensive forwards. Why? They faced the most challenging defensive situations versus the scoring lines of other teams.

But you can't "tell" that from plus/minus. They could be grinders facing other 3rd and 4th lines and out scoring them a bit. Or they could be an elite checking line and out scoring the Esposito's and Dionne's and Perreault's and Clarke's and Sittler's by a little bit and keeping 100 point players to 40 point outputs against them. We know it is the later. But plus/minus doesn't tell us this whatsoever.

 

[VanIslander]

So what does plus/minus tell us?

The top career +/- numbers:

1. Larry Robinson 730
2. Bobby Orr 597
3. Ray Bourque 528
4. Wayne Gretzky 518
5. Bobby Clarke 506
6. Serge Savard 460
Denis Potvin 460
8. Guy Lafleur 453
9. Bryan Trottier 452
10. Nicklas Lidstrom 450

The bottom career +/- numbers:

1 Bob Stewart -260
2 Don Lever -240
3 Gary Croteau -227
4 Dave Babych -223
5 Guy Charron -208
6 Willie Huber -203
7 Joe Cirella -201
8 Jack Lynch -197
9 Mike Sillinger -191
10 Ivan Boldirev -190

 

[Sens Rule]

The top career +/- numbers:

1. Larry Robinson 730
2. Bobby Orr 597
3. Ray Bourque 528
4. Wayne Gretzky 518
5. Bobby Clarke 506
6. Serge Savard 460
Denis Potvin 460
8. Guy Lafleur 453
9. Bryan Trottier 452
10. Nicklas Lidstrom 450

The bottom career +/- numbers:

1 Bob Stewart -260
2 Don Lever -240
3 Gary Croteau -227
4 Dave Babych -223
5 Guy Charron -208
6 Willie Huber -203
7 Joe Cirella -201
8 Jack Lynch -197
9 Mike Sillinger -191
10 Ivan Boldirev -190

While obviously the best players of all-time on great teams are going to have the best plus minus. Those with the worst are not near the worst players with decently long careers. And not the worst players defensively either. They are just the players with decently long careers playing on HORRIBLY BAD teams.

The worst 10 in plus minus are mostly no where near bad players. Or necessarily defensively terrible.

 

[Sens Rule]

The thread title should be: Plus/Minus. The stat that inherently punishes truly elite defensive checking line forwards that play versus other teams top lines.

Or Plus/Minus. How to tell who is playing in an elite team versus a bottom feeder.

 

[Chili]

The top career +/- numbers:

1. Larry Robinson 730
2. Bobby Orr 597
3. Ray Bourque 528
4. Wayne Gretzky 518
5. Bobby Clarke 506
6. Serge Savard 460
Denis Potvin 460
8. Guy Lafleur 453
9. Bryan Trottier 452
10. Nicklas Lidstrom 450

The bottom career +/- numbers:

1 Bob Stewart -260
2 Don Lever -240
3 Gary Croteau -227
4 Dave Babych -223
5 Guy Charron -208
6 Willie Huber -203
7 Joe Cirella -201
8 Jack Lynch -197
9 Mike Sillinger -191
10 Ivan Boldirev -190

There are some good hockey players on that bottom list, they just played on several bad teams. Charron never got to play a playoff game.

Those lists remind me of Paul Ysebaert who had the best plus minus in the NHL one year and the worst another year...with different teams.

 

[Doctor No]

David Rundblad was a +17 last year for the Hawks, one of the highest on the team. It is literally meaningless.

You're either misusing the word "literally" or the word "meaningless".

Or both, I guess.

David Rundblad was on the ice for seventeen more non-power play goals for than against last year. That has meaning.