Has a player actually registered a point per game in a season?

[Granlund2Pulkkinen*]

Has a player ever registered a point per game in a season... in other words.. did he player end up on the score sheet every game?

Click to read more...

 

[Clint]

Has a player ever registered a point per game in a season... in other words.. did he player end up on the score sheet every game?

An 82 game scoring streak?

I don't know for a fact, but I severely doubt it.

 

[Other Little Mikey]

No. Gretzky has the record for most consecutive games with a point: 51 in 83-84.

 

[Novak Djokovic]

In future, I doubt this will EVER happen. The Goalies are TOO good. (Their equipment and their well conditioned)

 

[Granlund2Pulkkinen*]

I'd think that someone in the 20's may have...

 

[Sens Rule]

No. Gretzky has the record for most consecutive games with a point: 51 in 83-84.

And of note - he did it from the start of the season. I think that he was only held off the scoresheet in 3 or 4 games that whole season. I am sure someone here knows exactly how many games he got points in that year.

 

[Wisent]

I'd think that someone in the 20's may have...

Dunno for sure but that is very possible. 20s or earlier.

 

[superpie]

Dunno for sure but that is very possible. 20s or earlier.

because those Fat Goalies back in the 20's sucked

 

[Hockey Outsider]

I'd think that someone in the 20's may have...

I looked into this.

In 1918, only the Ottawa Senators weren't shut out at least once. (If a team gets shut, then every player on it has at least one scoreless game). Ottawa's top player was Cy Denneny with 46 points in 22 games. He played in every game that year. So he might be a possibility. I estimate there's about a 6% chance of him scoring once in every game given his stats.

In 1919, again, only the Ottawa Senators weren't shut out at least once. Nighbor & Denneny both played in all 18 games... but they only scored 22 points each. It's extremely unlikely that they scored one point in every single game (about a 0.2% chance).

In 1920, for the third year in a row, only the Ottawa Senators weren't shut out at least once. Nighbor led the team in scoring but missed a game (so he's ineligible). Darragh was second in scoring but missed two games (he's also ineligible). Every other Senator scored less than 1 ppg. So we can say with certainty that nobody scored 1 point in each game in 1920.

In 1921, Ottawa and Hamilton weren't shut out at least once. Ottawa had one player over 1 ppg: Cy Denneny, who played in all 24 games. With 39 pts in 24 games he has about a 0.5% chance of scoring 1 point in each game. Hamilton had 2 players over 1 ppg: Joe Malone (who missed 4 games) and Goldie Prodgers (who missed 1 game), thus making both of them ineligible.

In 1922, Toronto and Ottawa weren't shut out at least once. Ottawa had 2 players with at least 1 ppg. Denneny missed two games and is ineligible, but Punch Broadbent played in all 24 games and won the Art Ross with 46 pts. I estimate that he has a 2% chance of scoring >1 point in each game.

Toronto has 4 players with at least 1 ppg. All four of them played in all 24 games. Dye, Cameron, Corbett Denneny and Noble scored 37, 27, 26 and 25 pts, respectively. I estimate that Dye has around a 0.3% of scoring >1 point in each game; the others have virtually no chance.

In 1923, 1924 and 1925, every team was shut out at least once. Therefore nobody scored in every game (unless they were traded midseason, perhaps).

Conclusions
I might look into a few more years. Based on what I've done so far (1918-1925), the most likely possibility is Cy Denneny in 1918. But even then, he only had about a 6% chance of scoring >1 point in each game. I strongly suspect that nobody has ever done it.

The probability of scoring at least 1 point in every game is phenomenally remote in an 82 game season. If a player scored 300 points (!) in a season, he'd still only have a 12% chance of scoring at least one point in every game.

 

[Big Phil]

Yeah Gretz's 51 will be hard to touch. Only one person in the NHl could do it IMO (you know). Mario had a 44 game streak but then left a game injured in 1990. Sundin actually had a 30 game point streak in '92-93

The only thing close that I have heard of someone scoring in every game is when Guy Lafleur scored one season vs. every team in the league. I believe it was '77-78. Other than that these types of streaks just dont happen.

 

[wedge]

I looked into this.

In 1918, only the Ottawa Senators weren't shut out at least once. (If a team gets shut, then every player on it has at least one scoreless game). Ottawa's top player was Cy Denneny with 46 points in 22 games. He played in every game that year. So he might be a possibility. I estimate there's about a 6% chance of him scoring once in every game given his stats.

In 1919, again, only the Ottawa Senators weren't shut out at least once. Nighbor & Denneny both played in all 18 games... but they only scored 22 points each. It's extremely unlikely that they scored one point in every single game (about a 0.2% chance).

In 1920, for the third year in a row, only the Ottawa Senators weren't shut out at least once. Nighbor led the team in scoring but missed a game (so he's ineligible). Darragh was second in scoring but missed two games (he's also ineligible). Every other Senator scored less than 1 ppg. So we can say with certainty that nobody scored 1 point in each game in 1920.

In 1921, Ottawa and Hamilton weren't shut out at least once. Ottawa had one player over 1 ppg: Cy Denneny, who played in all 24 games. With 39 pts in 24 games he has about a 0.5% chance of scoring 1 point in each game. Hamilton had 2 players over 1 ppg: Joe Malone (who missed 4 games) and Goldie Prodgers (who missed 1 game), thus making both of them ineligible.

In 1922, Toronto and Ottawa weren't shut out at least once. Ottawa had 2 players with at least 1 ppg. Denneny missed two games and is ineligible, but Punch Broadbent played in all 24 games and won the Art Ross with 46 pts. I estimate that he has a 2% chance of scoring >1 point in each game.

Toronto has 4 players with at least 1 ppg. All four of them played in all 24 games. Dye, Cameron, Corbett Denneny and Noble scored 37, 27, 26 and 25 pts, respectively. I estimate that Dye has around a 0.3% of scoring >1 point in each game; the others have virtually no chance.

In 1923, 1924 and 1925, every team was shut out at least once. Therefore nobody scored in every game (unless they were traded midseason, perhaps).

Conclusions
I might look into a few more years. Based on what I've done so far (1918-1925), the most likely possibility is Cy Denneny in 1918. But even then, he only had about a 6% chance of scoring >1 point in each game. I strongly suspect that nobody has ever done it.

The probability of scoring at least 1 point in every game is phenomenally remote in an 82 game season. If a player scored 300 points (!) in a season, he'd still only have a 12% chance of scoring at least one point in every game.

wow... great research

Ottawa spend 5 years without being shut out.. interesting

 

[Nalyd Psycho]

The only other possible times are the 29-30 season and during WW2.

 

[reckoning]

The only other possible times are the 29-30 season and during WW2.

No. Before Gretzky set the record at 51 games, Lafleur held it. Not sure how far his streak went, but it was definitely under 40. So any season with more games than that would be impossible.

 

[colonel_korn]

The probability of scoring at least 1 point in every game is phenomenally remote in an 82 game season. If a player scored 300 points (!) in a season, he'd still only have a 12% chance of scoring at least one point in every game.

Would you mind going a bit deeper into the math on that figure? It seems pretty surprising but I don't doubt it... just wondering how you came up with it.

 

[Complaining Customer]

(If a team gets shut, then every player on it has at least one scoreless game)

Unless your PPG player was hurt or scratched for whatever reason for that very game.

Edit : ok nevermind, if he has to go through all the season with a PPG, he can't miss a game... silly me.

 

[ck26]

ConclusionsThe probability of scoring at least 1 point in every game is phenomenally remote in an 82 game season. If a player scored 300 points (!) in a season, he'd still only have a 12% chance of scoring at least one point in every game.

I understand the statistical basis of this, but you're talking about randomly distributed points, and I don't think you can accurately and fairly apply this methodology to sports. If a player really did score 300 points in a season, he'd be averaging almost 4 per game. Getting shut out once in a 4-points-per-game season, would require he put up 8 in another game to balance it out. I can't find a link, but 8 a game has only happened a couple times in league history. While you probably can't apply logic to a 300-point season, CERTAINLY, we aren't talking about putting up 8 and 9 points a dozen-or-so times.

If somebody did manage 300 points somehow, the odds would be much higher than 1/8 that they also scored at least once every game for this.

Edit: somebody load up NHL 07 and conduct some trials.

 

[GSK*]

Did Joe Malone did it in 1917-18 ? 44 goals in 20 games, there was no assist back then.

 

[Donofan10]

I understand the statistical basis of this, but you're talking about randomly distributed points, and I don't think you can accurately and fairly apply this methodology to sports. If a player really did score 300 points in a season, he'd be averaging almost 4 per game. Getting shut out once in a 4-points-per-game season, would require he put up 8 in another game to balance it out. I can't find a link, but 8 a game has only happened a couple times in league history. While you probably can't apply logic to a 300-point season, CERTAINLY, we aren't talking about putting up 8 and 9 points a dozen-or-so times.

If somebody did manage 300 points somehow, the odds would be much higher than 1/8 that they also scored at least once every game for this.

Edit: somebody load up NHL 07 and conduct some trials.

Somone could score 5 points in 4 games to balence it out.

 

[reckoning]

I understand the statistical basis of this, but you're talking about randomly distributed points, and I don't think you can accurately and fairly apply this methodology to sports. If a player really did score 300 points in a season, he'd be averaging almost 4 per game. Getting shut out once in a 4-points-per-game season, would require he put up 8 in another game to balance it out. I can't find a link, but 8 a game has only happened a couple times in league history. While you probably can't apply logic to a 300-point season, CERTAINLY, we aren't talking about putting up 8 and 9 points a dozen-or-so times.

If somebody did manage 300 points somehow, the odds would be much higher than 1/8 that they also scored at least once every game for this.

It doesn't have to be an 8 point game to balance out the shutout. It could be two 6 point games or four 5 point games. These totals may seem unrealistic, but remember that we're talking about an imaginary 300 point season which is pretty unrealistic in itself.

Just curious about if the same logic could be applied to a team that scored 300 goals in a season and their chances of getting shutout. Just to pick a past year at random and see how teams over 300 goals fared:

1995-96

Pittsburgh 362 goals - shutout 2 times
Detroit 325 goals - shutout 0 times
Colorado 326 goals - shutout 3 times

The Oilers were shutout once in their record `83-`84 season when they scored 446 goals.

 

[Hockey Outsider]

Would you mind going a bit deeper into the math on that figure? It seems pretty surprising but I don't doubt it... just wondering how you came up with it.

Sure. This is very stat-heavy, though.

Scoring in hockey basically follows the Poisson distribution (a discrete probability distribution that looks at the number of events occuring in a certain period of time). It's not perfect but it models real-life hockey data very well.

To use the Poisson distribution, you need to know two variables: x, the number of events occuring, and lambda, the frequency of the event in a certain time period. You plug the numbers into the standard formula and it tells you the probability of X events occuring. The formula is: [(e ^ [-1 *lambda]) * (lambda ^ x)] / (factorial x). It's very nasty to use, I much prefer Excel.

For example, when Cy Denneny scored 46 points in 22 games, he is expected to score 2.09 points per game (lamba). Plugging this into the formula we see the probability of him scoring 0, 1, 2, etc., points in a single game.

0 - 12.4%
1 - 25.8%
2 - 27.0%
3 - 18.8%
4 - 9.8%
5 - 4.1%
6 - 1.4%
7 - 0.4%
8- 0.1%

From there, you switch to the binomial distribution (actually, we can use a simplified version in this case). Denneny has an 87.6% chance of scoring a point in any game he plays. The probability of him scoring in every game is simply: P ^ G, where P is the probability of him scoring at least 1 point in a game (this is what we just calculated) and G is the number of games player.

If Denneny had an 87.6% chance of scoring a point in every game, and he played in all 22 games, he has a 0.876 ^ 22 = 5.5% chance of scoring in all 22 games. (In real life, of course, this might need to be modified if he was playing through injuries, started recieving more ice time, etc.)

I don't take this method too seriously because of the things I mentioned above. However it is useful in showing how incredibly difficult it would be for somebody to score a point in every single game, even in a 20 game schedule. If somebody actually scored in every game, it would be one of the most unlikely and difficult accomplishments in history.

Cottonking, you're definitely right that there are a lot of factors that this doesn't take into account. A few (like injuries and linemates) are very important. A 300-point season is so hard to imagine that it's hard to say what would happen. You're right that there have only been a half dozen (approx.) 8-point games in all of NHL history. Then again, there hasn't been a single player to play at a 300-point pace for any sustained length of time. I'm not sure how good my own estimate is, in this case.

 

[Hockey Outsider]

Did Joe Malone did it in 1917-18 ? 44 goals in 20 games, there was no assist back then.

Joe Malone played for the Canadiens in 1917-18. They were shut-out twice (0-0 vs the Wanderers and 8-0 vs the Senators) (link), so it's impossible for Malone to have scored in those two games.

 

[BM67]

Just like to point out that the 0-0 vs the Wanderers was not played, but is the defaulted game credited to Montreal after the Wanderers arena burnt down.

Also Joe Malone had 3 five goal games that year.

 

[colonel_korn]

explanation

Thanks! I figured you had to make some assumption about how points were distributed per game...thought you might be using a normal distribution, but Poisson does make more sense.

 

[Hockey Outsider]

Just curious about if the same logic could be applied to a team that scored 300 goals in a season and their chances of getting shutout. Just to pick a past year at random and see how teams over 300 goals fared:

1995-96

Pittsburgh 362 goals - shutout 2 times
Detroit 325 goals - shutout 0 times
Colorado 326 goals - shutout 3 times

The Oilers were shutout once in their record `83-`84 season when they scored 446 goals.

Yes, in theory it should work just as well. (In real life you'd need some adjustments based on injuries--if Gretzky is injured, the Oilers aren't expected to score 5 goals per game anymore). Here's the probability of those events occuring:

1996 Pittsburgh Penguins: 18.4% chance of getting shutout 2 times
1996 Detroit Red Wings: 20.7% chance of getting shutout 0 times
1996 Colorado Avalanche: 13.1% chance of getting shutout 3 times
1984 Edmonton Oilers: 22.9% chance of getting shutout 1 time

All of the things perfectly reasonable, statistically. I'm sure there are some bizarre occurences (poor teams not getting shutout much or dominant teams getting shutout quite often) but they should be very rare.

Just like to point out that the 0-0 vs the Wanderers was not played, but is the defaulted game credited to Montreal after the Wanderers arena burnt down.

Also Joe Malone had 3 five goal games that year.

Good to know, I forgot about that. Well, the feat is still impossible for Malone due to the other 8-0 loss.

The probability of him getting 3 five-goal games... is 6%. Uncommon, but not unlikely.

Thanks! I figured you had to make some assumption about how points were distributed per game...thought you might be using a normal distribution, but Poisson does make more sense.

No problem. Conceptually, a Poisson distribution makes more sense in this case. It's a discrete distribution whereas the normal distribution is continuous. Scoring in hockey is discrete... you can't score 3.2 goals or get 1.4 assists.

 

[ck26]

I looked into this.The probability of scoring at least 1 point in every game is phenomenally remote in an 82 game season. If a player scored 300 points (!) in a season, he'd still only have a 12% chance of scoring at least one point in every game.

Talk me through your math, yeah? I calculate the odds at less than 2%, not 12.

You mentioned certain statistical flaws which can't easily be dealt with, but there's another one too ... any more than 9 or 10 points in a single game is basically impossible. While the odds that 300 randomly distributed points would end up clustered in 8 games is extremely low, the chances of it are present in my formula, and I don't know how to deal with it.