Measuring offensive production

[matnor]

A common way of measuring how good players are offensively around here is to count top-5, top-10 and top-20 finishes in goals, assists and points (see, for instance http://hfboards.com/showthread.php?t=614595 and http://hfboards.com/showthread.php?t=633070). While I think it's a relatively good way of comparing players I feel it punishes players who misses 10-15 games in a season disproportionally. For instance, let's say Crosby finishes outside the top-10 in scoring this season. Just comparing top-10 finishes would then ignore his amazing play the during the first half of the season. An alternative way is then to use top-10 finishes in ppg. That, on the other hand gives too much credit for players who are injured often.

I'll propose a different way of measuring elite offensive production. The idea is that players scoring at a top-10 pace should be rewarded for that but only in relation too how many games they actually played. Specifically, I consider a player scoring at a top-10 pace if he has at least as high ppg as the player with the 10th highest ppg of those playing at least half of the games in a season. I then summarize the amount of games each player has played at such a pace. Additionally, I adjust the season-length to a 82-game schedule since modern players otherwise would have an unfair advantage due to playing more games each season. I only consider games played in the NHL.

Let me exemplify this method taking Maurice Richard's season in 57/58. In this season he scored 34 points in 28 games. Of all the players who played at least half the season (35 games) the player with the 10th highest ppg was Alex Delvecchio who scored 59 points in 70 games. Clearly Richard scored at a higher pace than that so he gets credited for having scored at a top-10 pace for 28*(82/70)=33 games.

This is the top-50 in games played scoring at a top-5 pace:

Rank|Player|GP at top-5 pace
1. | Gordie Howe | 1589
2. | Wayne Gretzky | 1103
3. | Maurice Richard | 882
4. | Mario Lemieux | 815
5. | Stan Mikita | 785
6. | Bobby Hull | 777
7. | Jean Beliveau | 700
8. | Jaromir Jagr | 680
9. | Phil Esposito | 651
10. | Cy Denneny | 632
11. | Syl Apps | 607
12. | Andy Bathgate | 569
13. | Marcel Dionne | 492
14. | Bobby Orr | 480
15. | Bernie Geoffrion | 477
15. | Ted Lindsay | 477
15. | Howie Morenz | 477
18. | Guy Lafleur | 474
19. | Babe Dye | 460
20. | Doug Bentley | 455
21. | Bill Cowley | 437
22. | Sidney Crosby | 415
23. | Charlie Conacher | 404
24. | Elmer Lach | 399
25. | Alex Ovechkin | 396
26. | Mike Bossy | 392
27. | Busher Jackson | 379
28. | Joe Sakic | 377
29. | Peter Forsberg | 346
30. | Joe Malone | 343
31. | Eric Lindros | 338
32. | Marty Barry | 326
33. | Nels Stewart | 322
34. | Max Bentley | 315
35. | Steve Yzerman | 311
36. | Bryan Trottier | 309
37. | Jari Kurri | 301
38. | Newsy Lalonde | 290
39. | Sweeney Schriner | 289
40. | Dickie Moore | 269
41. | Bill Cook | 246
41. | Martin St. Louis | 246
43. | Toe Blake | 244
44. | Ken Hodge | 243
45. | Frank Boucher | 242
46. | Markus Naslund | 241
47. | Joe Thornton | 240
48. | Evgeni Malkin | 239
49. | Peter Stastny | 237
50. | Milt Schmidt | 235

Next is the top-50 in games played scoring at a top-10 pace:

Rank|Player|GP at top-10 pace
1. | Gordie Howe | 1671
2. | Wayne Gretzky | 1267
3. | Jean Beliveau | 1096
4. | Maurice Richard | 1013
5. | Bobby Hull | 935
6. | Mario Lemieux | 890
7. | Stan Mikita | 785
8. | Nels Stewart | 782
9. | Ted Lindsay | 774
10. | Andy Bathgate | 733
11. | Phil Esposito | 732
12. | Marcel Dionne | 722
13. | Aurel Joliat | 720
14. | Howie Morenz | 712
15. | Alex Delvecchio | 706
16. | Joe Sakic | 693
17. | Mark Messier | 692
18. | Syl Apps | 681
19. | Jaromir Jagr | 680
20. | Bill Cowley | 645
21. | Elmer Lach | 636
22. | Cy Denneny | 632
23. | Mike Bossy | 629
24. | Bernie Geoffrion | 624
25. | Frank Boucher | 570
26. | Reg Noble | 562
27. | John Bucyk | 553
27. | Peter Forsberg | 553
29. | Teemu Selanne | 539
30. | Doug Bentley | 530
30. | Babe Dye | 530
30. | Steve Yzerman | 530
33. | Henri Richard | 529
34. | Eric Lindros | 528
34. | Paul Coffey | 528
36. | Guy Lafleur | 526
37. | Denis Savard | 519
38. | Bobby Orr | 501
39. | Sidney Crosby | 496
40. | Marty Barry | 489
41. | Bill Cook | 488
42. | Norm Ullman | 481
43. | Frank Mahovlich | 479
44. | Bobby Clarke | 478
45. | Jean Ratelle | 477
46. | Peter Stastny | 476
47. | Alex Ovechkin | 475
48. | Charlie Conacher | 472
48. | Rod Gilbert | 472
50. | Darryl Sittler | 466

Finally, here is the top-50 in games played scoring at a top-20 pace:

Rank|Player|GP at top-10 pace
1. | Gordie Howe | 1835
2. | Wayne Gretzky | 1391
3. | John Bucyk | 1244
4. | Jean Beliveau | 1226
5. | Norm Ullman | 1197
6. | Alex Delvecchio | 1193
7. | Frank Mahovlich | 1166
8. | Stan Mikita | 1151
9. | Phil Esposito | 1133
10. | Joe Sakic | 1120
11. | Marcel Dionne | 1116
12. | Nels Stewart | 1104
13. | Maurice Richard | 1089
14. | Bobby Hull | 1014
15. | Jaromir Jagr | 992
16. | Bernie Geoffrion | 901
17. | Mario Lemieux | 890
17. | Mark Messier | 890
19. | Jean Ratelle | 881
20. | Andy Bathgate | 880
21. | Ted Lindsay | 856
22. | Steve Yzerman | 851
23. | Howie Morenz | 842
24. | Henri Richard | 820
25. | Frank Boucher | 813
26. | Aurel Joliat | 800
27. | Gilbert Perreault | 796
28. | Cy Denneny | 792
29. | Elmer Lach | 787
30. | Teemu Selanne | 778
31. | Rod Gilbert | 777
32. | Mike Bossy | 771
33. | Bill Cowley | 758
34. | Toe Blake | 754
35. | Paul Coffey | 748
36. | Dale Hawerchuk | 729
37. | Reg Noble | 719
38. | Ted Kennedy | 710
39. | Joe Thornton | 708
40. | Red Kelly | 699
41. | Adam Oates | 698
42. | Peter Stastny | 692
43. | Ron Francis | 688
44. | Roy Conacher | 682
44. | Peter Forsberg | 682
46. | Syl Apps | 681
47. | Brett Hull | 679
48. | Denis Savard | 676
49. | Sweeney Schriner | 662
50. | Marty Barry | 653

Any thoughts, comments or criticisms?

Edit: Changed minor errors in the tables.

Edit2: Updated through the 2011-12 season.

Edit3: Updated through the 2012-13 season.

Click to read more...

 

[SidGenoMario]

What an awesome idea, I've never seen this before.

 

[Zauper]

This goes back to the central issue: you're rewarding players for not playing full seasons.

Is 77 points in 67 games (94.2p pace) better than 94 points in 82 games? (Malkin last year vs MSL)

But you're not even assuming that a player has to have played half the season. Though you do assume that the person you're comparing them to has to -- why? They could still be way off of the actual top performers in a given year. What about Forsberg in 07-08? He played 9 games and put up 14 points. Why are we rewarding him for that, exactly?

Then what are you using to determine the top PPG ratios? I've noticed that Ovechkin, who has finished t5 in scoring in 4 of his 5 years in the league only has three years worth of numbers. Which means your argument is that in spite of him finishing in the top 5 in whichever year you're including -- not bothering to check PPGs -- he wasn't on a t5 finish pace. Even though I doubt whoever you're replacing him with finished in the top 10 or even the top 15.

I'm sure the same problem occurs with several other people in other seasons, as well.

 

[TheDevilMadeMe]

This is fascinating.

Surprised to see Maurice Richard so high, considering he never won an Art Ross (but was 2nd 5 times). I guess things just didn't fall right for him in any particular season.

Johnny Bucyk's numbers really drive home his reputation - never a game breaker, but among the game's very good for a long time. Seems like you could count on him being an 11-20 scorer on a regular basis.

Surprised to see Frank Mahovlich with a big disconnect between "top 10" and "top 20" pace. I always thought of him as truly elite performer, rather than an "almost elite for a long time" guy. I guess he gets extra credit for playing in Toronto's defensive system and then producing big numbers in the playoffs.

I'm actually really surprised to see Ron Francis not even on the list of "top 20" finishes. Does this really drive home just how non-elite his offense was at any given time?

 

[TheDevilMadeMe]

This goes back to the central issue: you're rewarding players for not playing full seasons.

Is 77 points in 67 games (94.2p pace) better than 94 points in 82 games? (Malkin last year vs MSL)

But you're not even assuming that a player has to have played half the season. Though you do assume that the person you're comparing them to has to -- why? They could still be way off of the actual top performers in a given year. What about Forsberg in 07-08? He played 9 games and put up 14 points. Why are we rewarding him for that, exactly?

Then what are you using to determine the top PPG ratios? I've noticed that Ovechkin, who has finished t5 in scoring in 4 of his 5 years in the league only has two years worth of numbers. Which means your argument is that in spite of him finishing in the top 5 in whichever year you're including -- not bothering to check PPGs -- he wasn't on a t5 finish pace. Even though I doubt whoever you're replacing him with finished in the top 10 or even the top 15.

I'm sure the same problem occurs with several other people in other seasons, as well.

Good point.

I think this is a fascinating idea, but you definitely need to remove guys who played less than a certain number of games. No reason for Forsberg's pace in 07-08 to take away from what anyone else in that season did, when he could only maintain it for 14 games.

I'm sure there are a rookie or two somewhere in there who scored at a better than PPG pace in less than 5 games.

 

[matnor]

This goes back to the central issue: you're rewarding players for not playing full seasons.

Is 77 points in 67 games (94.2p pace) better than 94 points in 82 games? (Malkin last year vs MSL)

But you're not even assuming that a player has to have played half the season. Though you do assume that the person you're comparing them to has to -- why? They could still be way off of the actual top performers in a given year. What about Forsberg in 07-08? He played 9 games and put up 14 points. Why are we rewarding him for that, exactly?

Then what are you using to determine the top PPG ratios? I've noticed that Ovechkin, who has finished t5 in scoring in 4 of his 5 years in the league only has two years worth of numbers. Which means your argument is that in spite of him finishing in the top 5 in whichever year you're including -- not bothering to check PPGs -- he wasn't on a t5 finish pace. Even though I doubt whoever you're replacing him with finished in the top 10 or even the top 15.

I'm sure the same problem occurs with several other people in other seasons, as well.

As for your first question. Martin St. Louis would get credited with 82 games and Malkin with 67 games. So Malkin does not get credit for a full season but instead only for the games he did play. The same with Forsberg, he would get credit with 9 games in 07/08, which is negligible in the grand scheme of things. Finally, I just spotted an error in my code which meant players who scored exactly at a top-5 pace were not included. Therefore, Ovehckin's first season where he had the 5th highest ppg should be included. I'll update the tables.

 

[Hawkey Town 18]

Then what are you using to determine the top PPG ratios? I've noticed that Ovechkin, who has finished t5 in scoring in 4 of his 5 years in the league only has two years worth of numbers. Which means your argument is that in spite of him finishing in the top 5 in whichever year you're including.

While this post is slightly inaccurate (Ovechkin is given credit for 233 games, which is 3 seasons worth, not 2), it does bring up a good point. Why isn't Ovechkin being given top 5 credit for the 05'-06' season? According to hockeyreference.com he finished 5th in PPG that year.

http://www.hockey-reference.com/leagues/NHL_2006_leaders.html


EDIT: You beat me to it. Question Answered in the post above.

 

[Hawkey Town 18]

Good point.

I think this is a fascinating idea, but you definitely need to remove guys who played less than a certain number of games. No reason for Forsberg's pace in 07-08 to take away from what anyone else in that season did, when he could only maintain it for 14 games.

I'm sure there are a rookie or two somewhere in there who scored at a better than PPG pace in less than 5 games.

I don't think Forsberg takes away from anyone with the way this is set up. The Top 5 pace is determined by a guy who played half the games, so Forsberg does not factor in. He just gets credit for his 14 games. Theoretically, it is possible for 6, 7, 8, 9, 10, etc. guys to get credited with a Top 5 pace if many of them played less than half the games.

 

[matnor]

I don't think Forsberg takes away from anyone with the way this is set up. The Top 5 pace is determined by a guy who played half the games, so Forsberg does not factor in. He just gets credit for his 14 games. Theoretically, it is possible for 6, 7, 8, 9, 10, etc. guys to get credited with a Top 5 pace if many of them played less than half the games.

Exactly, Forsberg gets credit for 9 games but he doesn't count when calculating who the player with the fifth highest scoring pace was. There is a slight problem with regression-to-the-mean for players with few games. This is especially problematic for the earlier era since those games have higher weights but I don't know how to get around that problem.

 

[matnor]

Updated original post with (hopefully) correct tables.

 

[Zauper]

As for your first question. Martin St. Louis would get credited with 82 games and Malkin with 67 games.

In a world where MSL was 6th in PPG pace and Malkin 5th (or 11th and 10th), MSL wouldn't be credited with top5/top10 pace, according to your methodology since because Malkin played over 50% of the games in the season he is included in the pace calculation. Thus because he scored at a slightly higher pace (.2 PPG), he would be ranked above MSL in spite of (IMO) having an inferior season. Here's a better example, perhaps:
In 07-08, Mike Richards had 75p in 73GP. Kovalev had 84P in 82GP. Richards is the cutoff for 20th in PPG (of people who played over 50% of the season), which means Kovalev doesn't get any credit under your system. Is Richards season really better than Kovalevs?
Is Spezza's 06-07 season really better than Heatley's? 67GP-87P vs 82GP-105P. That's the cutoff for 5-6 in PPG.

Personally? I'd rather have the player that plays an extra 10-15 games over the player that would give me one or two (or even only a fraction of one) more point(s) if they had played those games.

So Malkin does not get credit for a full season but instead only for the games he did play. The same with Forsberg, he would get credit with 9 games in 07/08, which is negligible in the grand scheme of things.

Yes, but Malkin also pushes down the people who had better seasons than him but less PPG. The same with Forsberg? Or is he excluded from pushing people down because he only played 9 games? In either case; giving him 9 games worth of credit because he only played 9 games in a season, though at a high level, seems off to me.

Finally, I just spotted an error in my code which meant players who scored exactly at a top-5 pace were not included. Therefore, Ovehckin's first season where he had the 5th highest ppg should be included. I'll update the tables

Gotcha. Wonder if that moves anyone else. It should since presumably it meant every player to ever be #5 in PPG would have that season excluded.

 

[JackSlater]

Very good stuff. I like that it shows that Lindros was more dominant than most give him credit for. I agree with the idea that players below a certain games played threshold (I would set it maybe around 20 games personally) should not impact the finishes of other players though. I would not remove those games from the injured player's totals though.

 

[matnor]

I'm actually really surprised to see Ron Francis not even on the list of "top 20" finishes. Does this really drive home just how non-elite his offense was at any given time?

Francis numbers are 152/287/688. Obviously, his best seasons were with Jagr (and Lemieux on the PP) so you have to take his elite numbers with a pinch of salt.

 

[matnor]

Very good stuff. I like that it shows that Lindros was more dominant than most give him credit for. I agree with the idea that players below a certain games played threshold (I would set it maybe around 20 games personally) should not impact the finishes of other players though. I would not remove those games from the injured player's totals though.

That's exactly how I thought. The limit I set such that a player is required to have played at least half the season in order to affect the threshold but the games the players played still count for that particular player (see Forsberg in 03/04 and 07/08, Kariya in 97/98, Lemieux in 90/91, 93/94, 01/02 etc.)

 

[pluppe]

this is very good stuff.

I would be very interested to see top-1, 2 and 3 finishes as well as I think that would paint an interesting picture of absolute peak. If you ever get the time of course.

 

[matnor]

In a world where MSL was 6th in PPG pace and Malkin 5th (or 11th and 10th), MSL wouldn't be credited with top5/top10 pace, according to your methodology since because Malkin played over 50% of the games in the season he is included in the pace calculation. Thus because he scored at a slightly higher pace (.2 PPG), he would be ranked above MSL in spite of (IMO) having an inferior season. Here's a better example, perhaps:

In 07-08, Mike Richards had 75p in 73GP. Kovalev had 84P in 82GP. Richards is the cutoff for 20th in PPG (of people who played over 50% of the season), which means Kovalev doesn't get any credit under your system. Is Richards season really better than Kovalevs?

Is Spezza's 06-07 season really better than Heatley's? 67GP-87P vs 82GP-105P. That's the cutoff for 5-6 in PPG.

Personally? I'd rather have the player that plays an extra 10-15 games over the player that would give me one or two (or even only a fraction of one) more point(s) if they had played those games.

I see your point but I don't really know how to correct for that. It only affects players right on the margin though. I could possibly raise the required number of games played to affect the threshold to be as high as 70-75 games (on a 82-game schedule).

Yes, but Malkin also pushes down the people who had better seasons than him but less PPG. The same with Forsberg? Or is he excluded from pushing people down because he only played 9 games? In either case; giving him 9 games worth of credit because he only played 9 games in a season, though at a high level, seems off to me.

Forsberg would not affect the threshold since he didn't play half the games. And 9 games is 9 games so I think he should get credit for that.

Gotcha. Wonder if that moves anyone else. It should since presumably it meant every player to ever be #5 in PPG would have that season excluded.

Yes, it moved a lot of players around.

 

[matnor]

this is very good stuff.

I would be very interested to see top-1, 2 and 3 finishes as well as I think that would paint an interesting picture of absolute peak. If you ever get the time of course.

I think one should be careful in interpreting number of games played at top-1, 2 and 3-pace since there would be high variance in such numbers where players who were injured often might be favored in a regression-to-the-mean sense. Also, players playing at the same time as Gretzky, Howe etc would have it real hard to be number 1. But, since you asked nicely, here's the numbers :

Top 50 in games played scoring at a top-1 pace:

|Player|GP at top-1 pace
1. | Wayne Gretzky | 872
2. | Gordie Howe | 567
3. | Mario Lemieux | 497
4. | Phil Esposito | 406
5. | Stan Mikita | 326
6. | Bill Cowley | 274
7. | Guy Lafleur | 234
8. | Alex Ovechkin | 233
9. | Jean Beliveau | 231
10. | Joe Malone | 225
11. | Jaromir Jagr | 221
12. | Sweeney Schriner | 164
13. | Bill Cook | 164
14. | Elmer Lach | 164
15. | Max Bentley | 159
16. | Bobby Hull | 158
17. | Howie Morenz | 153
18. | Charlie Conacher | 152
19. | Peter Forsberg | 123
20. | Maurice Richard | 111
21. | Bobby Orr | 92
22. | Bryan Hextall | 82
23. | Busher Jackson | 82
24. | Ace Bailey | 82
25. | Andy Bathgate | 82
26. | Nels Stewart | 82
27. | Cooney Weiland | 82
28. | Punch Broadbent | 82
29. | Gordie Drillon | 82
30. | Joe Thornton | 81
31. | Ted Lindsay | 81
32. | Doug Bentley | 79
33. | Sidney Crosby | 79
34. | Eric Lindros | 79
35. | Bryan Trottier | 78
36. | Bobby Clarke | 78
37. | Newsy Lalonde | 77
38. | Babe Dye | 75
39. | Bernie Geoffrion | 75
40. | Todd Bertuzzi | 72
41. | Georges Boucher | 72
42. | Aurel Joliat | 68
43. | Syl Apps | 46
44. | Marc Savard | 45
45. | Ziggy Palffy | 35
46. | Jack McGill | 22
47. | Paul Kariya | 22
48. | Bill Mosienko | 21
49. | Mike Walton | 10
50. | Leth Graham | 3
51. | Eddie Wares | 3
52. | Kamil Piros | 3
53. | Bobby Bauer | 3

Top 50 in games played scoring at a top-2 pace:

|Player|GP at top-2 pace
1. | Wayne Gretzky | 1103
2. | Mario Lemieux | 788
3. | Gordie Howe | 649
4. | Jaromir Jagr | 548
5. | Jean Beliveau | 536
6. | Phil Esposito | 488
7. | Maurice Richard | 424
8. | Stan Mikita | 402
9. | Bobby Orr | 400
10. | Bobby Hull | 397
11. | Cy Denneny | 389
12. | Marcel Dionne | 328
13. | Guy Lafleur | 316
14. | Charlie Conacher | 296
15. | Bernie Geoffrion | 289
16. | Bill Cowley | 274
17. | Ted Lindsay | 245
18. | Newsy Lalonde | 238
19. | Max Bentley | 236
20. | Busher Jackson | 236
21. | Alex Ovechkin | 233
22. | Joe Malone | 225
23. | Dickie Moore | 203
24. | Syl Apps | 196
25. | Peter Forsberg | 195
26. | Bill Cook | 164
27. | Elmer Lach | 164
28. | Nels Stewart | 164
29. | Sweeney Schriner | 164
30. | Joe Thornton | 163
31. | Andy Bathgate | 158
32. | Bryan Trottier | 157
33. | Babe Dye | 154
34. | Howie Morenz | 153
35. | Aurel Joliat | 150
36. | Jari Kurri | 140
37. | Sidney Crosby | 132
38. | Eric Lindros | 131
39. | Bill Mosienko | 86
40. | Norm Ullman | 82
41. | Bryan Hextall | 82
42. | Mike Bossy | 82
43. | Pat LaFontaine | 82
44. | Ace Bailey | 82
45. | Roy Conacher | 82
46. | Punch Broadbent | 82
47. | Milt Schmidt | 82
48. | Martin St. Louis | 82
49. | Cooney Weiland | 82
50. | Evgeni Malkin | 82
51. | Jarome Iginla | 82
52. | Henrik Sedin | 82
53. | Buddy O'Connor | 82
54. | Toe Blake | 82
55. | Gordie Drillon | 82

Top 50 in games played scoring at a top-3 pace:

|Player|GP at top-3 pace
1. | Wayne Gretzky | 1103
2. | Gordie Howe | 895
3. | Mario Lemieux | 815
4. | Stan Mikita | 706
5. | Jean Beliveau | 700
6. | Jaromir Jagr | 680
7. | Phil Esposito | 651
8. | Maurice Richard | 588
9. | Bobby Orr | 480
10. | Guy Lafleur | 474
11. | Bobby Hull | 469
12. | Cy Denneny | 468
13. | Babe Dye | 460
14. | Ted Lindsay | 395
15. | Doug Bentley | 373
16. | Charlie Conacher | 371
17. | Bernie Geoffrion | 358
18. | Marcel Dionne | 328
19. | Andy Bathgate | 322
20. | Busher Jackson | 301
21. | Newsy Lalonde | 290
22. | Sweeney Schriner | 289
23. | Syl Apps | 278
24. | Bill Cowley | 274
25. | Dickie Moore | 269
26. | Eric Lindros | 267
27. | Elmer Lach | 246
28. | Joe Thornton | 240
29. | Max Bentley | 236
30. | Alex Ovechkin | 233
31. | Howie Morenz | 231
32. | Joe Malone | 225
33. | Joe Sakic | 215
34. | Sidney Crosby | 209
35. | Peter Forsberg | 195
36. | Bill Cook | 164
37. | Evgeni Malkin | 164
38. | Nels Stewart | 164
39. | Denis Savard | 162
40. | Norm Ullman | 162
41. | Frank Nighbor | 161
42. | Larry Aurie | 161
43. | Mark Messier | 160
44. | Paul Thompson | 159
45. | Peter Stastny | 159
46. | Bryan Trottier | 157
47. | Joe Primeau | 155
48. | Milt Schmidt | 155
49. | Gordie Drillon | 154
50. | Jarome Iginla | 152

 

[canucks4ever]

Surpised to see bathgate and dionne so high on that list, especially when you consider that they were competing against howe and gretzky respectively.

 

[TheDevilMadeMe]

Yeah, the fact that Richard goes from 20th in Top 1 finishes to 7th in Top 2 finishes shows that they probably aren't a very meaningful metric. (Just the first example I saw).

I do find the top 5, 10, and 20 metrics interesting.

 

[Zauper]

I believe that top1 finishes largely is a lot like looking at the Art Ross winners, with few exceptions due to injuries, but in most cases they're injuries that only prevented an art ross finish (still finished high in points in a given year).

But yeah, the 1/2/3 metric probably have limited meaning themselves.

 

[TheDevilMadeMe]

Surpised to see bathgate and dionne so high on that list, especially when you consider that they were competing against howe and gretzky respectively.

Bathgate and Dionne were absolutely dominant regular season offensive players. Their downsides are very similar - neither played a lick of defense, and both have serious question marks about their playoff resumes (Dionne moreso on the second one).

 

[seventieslord]

very interesting stuff!

 

[WilliamRanford]

This is a very cool idea.

One possible way to express this data would be to show what percentage of their normalized careers they played at a top-5, top 10, etc pace. Howe played much longer than Gretzky, so you'd expect his absolute number to be higher. But it Gretz played 78% of his career at a top 5 pace while Howe played 72% at a top 5 pace, that would be the paydirt, IMO.

Thoughts?

 

[SidGenoMario]

Shows how godly Malkin can/should be.

 

[canucks4ever]

Bathgate and Dionne were absolutely dominant regular season offensive players. Their downsides are very similar - neither played a lick of defense, and both have serious question marks about their playoff resumes (Dionne moreso on the second one).

Yeah i know thier downsides, i just didnt expect them to rank so high on these lists, dionne was no defensive wizard but he was gritty and could get his nose dirty, its not like he's pierre turgeon.

Bathgate played on the worst team in the o-6, not much of an opportunity to develop a playoff resume.