The best goalscorers in the NHL-3

[Namba 17]

The previous one was here The best goalscorers in the NHL - 2

But this time I came with bigger data (from 46/47) and different criterias.

I still have no ideas how to make good tables inside this forum. So, enjoy fotos.
In this post I include table with the best goalscorers 47 - 2024 ranked by ratio to average goals scored by players, which scored at least one goal in the NHL.
In the nexst post ill attach tables differentiated by amount of teams, played in the different time periods.

Youll also find insde % of goals, scored by the best NHL goal-scoreres and his team and all NHL goals and ratio between goals by the best goal-scorer and average goals per player in the corresponding season.

Click to read more...

 

[Namba 17]

by time periods

 

[Namba 17]

By ratio between % of player/ team and team/league

 

[Namba 17]

And no reactions at all
Some kind of a record

 

[The Panther]

I think we're all interested, but it's just a little reader-unfriendly (at least, on my Surface Pro, it's kind of hard to see the data). I'm not sure about other devices.

It's great that you expanded it to way back in time -- always interesting.

Do I understand that each item is based on individual seasons in isolation? (i.e., you're not considering eras or short-periods of time; just 1 season by itself? just checking.)

(By the way, there's an error where you're listing Gretzky 1983-84 as a "22 team" League, when it was 21.)

 

[Namba 17]

I think we're all interested, but it's just a little reader-unfriendly (at least, on my Surface Pro, it's kind of hard to see the data). I'm not sure about other devices.

Yes, I understand and sorry about that. If smone teach me how to insert tables here in better way - I'll do it. Im too dumb to learn it myself.

Do I understand that each item is based on individual seasons in isolation? (i.e., you're not considering eras or short-periods of time; just 1 season by itself? just checking.)

Yes, each row is one season.

(By the way, there's an error where you're listing Gretzky 1983-84 as a "22 team" League, when it was 21.)

Yes, you right, thank you!
It didn't change anything in calculation though.

 

[Michael Farkas]

Yeah, sorry, namba...I don't totally understand what it's trying to say here. I know how to use Tableau, if I can figure out the end goal, I might be able to help with the data viz haha

 

[Bear of Bad News]

Yeah, if I saw the intended math behind it, I could probably figure out how to help here.

 

[Namba 17]

Yeah, sorry, namba...I don't totally understand what it's trying to say here. I know how to use Tableau, if I can figure out the end goal, I might be able to help with the data viz haha

Yeah, if I saw the intended math behind it, I could probably figure out how to help here.

Ok, thanks.
So what did I do.
Im still running around with idea to estimate goalscorers of different epochs of the NHL.
To do this I made a table with:
1, Names of the #1 goalscorer of the NHL in each season since 46/47 and their results in corresponding season.
2. Amount of goals their teams scored in this season
3. % of goals of the team which #1 goalscorer scored in corresponding season
4. Amount of goals scored by all players of the NHL in correspondin season
5. % of goals of the whole NHL scored by #1 goalcorer
6. Avarage amount of goals scored by each player, who stepped on ice in corresponding season
7. Ratio between amount of goals by #1 goalscorer and average amount of goals of every player in corresponding season
8. Avarage amount of goals scored by each player, who scored at least 1 goal in corresponding season
9. Ratio between amount of goals by #1 goalscorer and average amount of goals of players, who scored at least 1 goal
10. % of goals, scored by team of #1 goascorer, of total amount of goals, scored by league in the correspondin season.
11. Ratio between share of goals of #1 goalscorer of his team's goals and share of goals of his team in total amount of goals, scored in the league.

Doing all this I found out that its almost impossible to compare players who played in different epochs of the NHL -epochs with different amount of teams.

Thats why I did one table, where I rated player by ratio between their goals scored and average amount of goals scored by players, who scored at least 1 goal (per one player) - 1st post.
Then I divided NHL by periods by amount of teams, played in each period (6 teams - one period, 12-14 teams - another etc) and rated players inside those periods - 2nd post
Then I rated players inside those periods by ratio between share of goals of #1 goalscorer of his team's goals and share of goals of his team in total amount of goals, scored in the league -3d post.

Thats what I want to share with community
If its sounds complicated - please, ask questions

 

[MadLuke]

My general first thought would be that give advice to do exactly like what I did to compare scoring accross era which would lead to boring similar result (but could catch my mistake)

i.e. focus more on what the top goal scorer did than league wide scoring, if someone played in a era that defenseman did not score that many goals versus someone else that played when it was popular to do so, should it say anything about how good they were at scoring goals ?

In an era where there was 12 skaters getting ice time versus an era where they rolled 18 ?


Going with league wide scoring seem to introduce all those issues and looking what the average goal scorer did among all skaters that scored goals is just a complicated way to get league wide scoring numbers + introducing a bigger issues when the amount of skaters by team change over time.

 

[Bear of Bad News]

Gotcha - so if I can try to boil this down with an example:

In 2025-26, Mike Farkas is the leading scorer in the NHL and he plays for the HFBoards Historians. He scored 30% of the Historians' goals.

Overall for 2025-26 across all 32 teams, the team's leading scorer accounted for 24% of their teams' goals (this is the sum of goals by each team's leading scorer divided by the total goals in the league).

2025-26 would receive a "score" of 30%/24% or 1.25?

(And would it matter if the second-leading scorer in the NHL, Mad Luke, scored 32% of his team's goals?)

 

[Namba 17]

Gotcha - so if I can try to boil this down with an example:

In 2025-26, Mike Farkas is the leading scorer in the NHL and he plays for the HFBoards Historians. He scored 30% of the Historians' goals.

Overall for 2025-26 across all 32 teams, the team's leading scorer accounted for 24% of their teams' goals (this is the sum of goals by each team's leading scorer divided by the total goals in the league).

2025-26 would receive a "score" of 30%/24% or 1.25?

(And would it matter if the second-leading scorer in the NHL, Mad Luke, scored 32% of his team's goals?)

I didn compare each team best scorers. I took only #1 results.

In 2025-26, Mike Farkas is the leading scorer in the NHL and he plays for the HFBoards Historians. He scored 30% of the Historians' goals.
Then we'll see how many goals his team scored and find % of total amount of all goals scored in the season (by all teams), say, its 15%.
Then we divide 30 on 15. The result itself doesnt mean much. But we compare it to the same result of the best goalscorers of the NHL, who played in the NHL in different seasons but as long, as there are 32 teams in the NHL.

My general first thought would be that give advice to do exactly like what I did to compare scoring accross era which would lead to boring similar result (but could catch my mistake)

i.e. focus more on what the top goal scorer did than league wide scoring, if someone played in a era that defenseman did not score that many goals versus someone else that played when it was popular to do so, should it say anything about how good they were at scoring goals ?

In an era where there was 12 skaters getting ice time versus an era where they rolled 18 ?


Going with league wide scoring seem to introduce all those issues and looking what the average goal scorer did among all skaters that scored goals is just a complicated way to get league wide scoring numbers + introducing a bigger issues when the amount of skaters by team change over time.

Im not sure I got your point

 

[MadLuke]

Im not sure I got your point

As roster get bigger, it will be natural for top scorer share of goals to get down (but it will not be linear) considering this what does their share of league goals tell us ?

And because it will not be linear, we cannot adjust (like hockey reference do by doing a simple that year roster size / 18 ratio type).

You risk really strongly to end up doing exactly what hockey reference do, average team scoring, some roster size ratio, which is not bad per say but already available.

 

[Namba 17]

As roster get bigger, it will be natural for top scorer share of goals to get down (but it will not be linear) considering this what does their share of league goals tell us ?

Thats why I compare result only inside time periods with the same amout of teams in the League.

You risk really strongly to end up doing exactly what hockey reference do, average team scoring, some roster size ratio, which is not bad per say but already available.

I havent see anything like what i did but if anybody already did this i wont complain

But to say if its the same or not - people should see what i did. and it seems that thats the main problem

If you know how to put tables here I can send my results to you, if you wish and you will post them here in better shape

 

[MadLuke]

Thats why I compare result only inside time periods with the same amout of teams in the League.

I was more talking about the number of players that play for a team, nhl roster size changed constantly.

The history of roster sizes

I've searched a decent amount on this, but haven't been able to find any concrete answers. Since many of you guys have access to incredible resources, and many of you guys are incredible resources yourselves, I figured I'd ask here and give it a shot. Obviously, in the NHL today the roster...

forums.hfboards.com

• 1925-26 - Teams to dress a maximum of 12 players for each game from a roster of no more than 14 players.
• 1928-29 - Exclusive of goaltenders, team to dress at least 8 and no more than 12 skaters.
• 1938-39 - Maximum number of players in uniform increased from 14 to 15.
• 1942-43 - Player limit reduced from 15 to 14. Minimum of 12 men in uniform abolished.
• 1949-50 - Clubs allowed to dress 17 players exclusive of goaltenders.
• 1951-52 - Number of players in uniform reduced to 15 plus goaltenders.
• 1952-53 - Teams permitted to dress 15 skaters on the road and 16 at home.
• 1952-53 - Teams permitted to dress 15 skaters on the road and 16 at home.
• 1954-55 - Number of players in uniform set at 18 plus goaltenders up to December 1 and 16 plus goaltenders
• 1960-61 - Number of players in uniform set at 16 plus goaltenders.
• 1965-66 - Teams required to dress two goaltenders for each regular-season game.
• 1971-72 - Number of players in uniform set at 17 plus 2 Goaltenders.
• 1982-83 - Number of players in uniform set at 18 plus 2 goaltenders.

Even in the 06 era, roster shift quite a bit.

In 52-53 if you are not Red Kelly and play defense, you will not score many goals, all the Ds in the nhl scored a combined 95 goals (19 of those from Kelly)

There was 1006 goals scored that year, a bit less than 9.5% of them by Ds, by 66-67 they scored 155 goals in a league that scored 1252, they share increased to 12.4%.

That would naturally decrease the percentage of goals of forward scorer, should this impact how good they are considered at scoring ?

 

[Namba 17]

I was more talking about the number of players that play for a team, nhl roster size changed constantly.

The history of roster sizes

I've searched a decent amount on this, but haven't been able to find any concrete answers. Since many of you guys have access to incredible resources, and many of you guys are incredible resources yourselves, I figured I'd ask here and give it a shot. Obviously, in the NHL today the roster...

forums.hfboards.com

• 1925-26 - Teams to dress a maximum of 12 players for each game from a roster of no more than 14 players.
• 1928-29 - Exclusive of goaltenders, team to dress at least 8 and no more than 12 skaters.
• 1938-39 - Maximum number of players in uniform increased from 14 to 15.
• 1942-43 - Player limit reduced from 15 to 14. Minimum of 12 men in uniform abolished.
• 1949-50 - Clubs allowed to dress 17 players exclusive of goaltenders.
• 1951-52 - Number of players in uniform reduced to 15 plus goaltenders.
• 1952-53 - Teams permitted to dress 15 skaters on the road and 16 at home.
• 1952-53 - Teams permitted to dress 15 skaters on the road and 16 at home.
• 1954-55 - Number of players in uniform set at 18 plus goaltenders up to December 1 and 16 plus goaltenders
• 1960-61 - Number of players in uniform set at 16 plus goaltenders.
• 1965-66 - Teams required to dress two goaltenders for each regular-season game.
• 1971-72 - Number of players in uniform set at 17 plus 2 Goaltenders.
• 1982-83 - Number of players in uniform set at 18 plus 2 goaltenders.

Even in the 06 era, roster shift quite a bit.

In 52-53 if you are not Red Kelly and play defense, you will not score many goals, all the Ds in the nhl scored combined 95 goals (19 of those from Kelly)

There was 1006 goals scored that year, a bit less than 9.5% of them by Ds, by 66-67 they scored 155 goals in a league that scored 1252, they share increased to 12.4%.

That would naturally decrease the percentage of goals of forward scorer, should this impact how good they are considered at scoring ?

Interesting point, but it didnt affect numbers (or almost didn't), while changing of amount of teams definately did. And you can see it in my tables also

 

[MadLuke]

Interesting point, but it didnt affect numbers (or almost didn't), while changing of amount of teams definately did. And you can see it in my tables also

Yes I believe you 100% of that, that the roster size change will not affect the top scorer a lot (except say if we compare the 20s vs today level of roster, once you have 14 skater or so, how many less minutes Howe get if you rise the amount of skater, are more fresh the skater he face ?, etc...), that one major issue with hockey reference that they assume it not only will, but that it will be linear instead of a log (at least for the player who interest us, the superstar, it could be perfectly true for third liner).

Would you have went back far enough to the Morenz day (or even more so Joe Malone), I think you would have seen the effect clearly.