The problem with using 1st places for normalizing anything is that sometimes the 1st place is an obvious outlier. That should be dubbed "the Gretzky effect" I think, but it happens with other players too (peak Howe, peak Hulls, etc.)
That's the reason why, for example, the VsX measure normalizes goals/points to #2 totals, not #1 (otherwise everyone who played during Gretzky's prime would look undeservedly bad).
Similarly, one can argue that 1970-1996 had such great leads of #1 over #10 not because it was easier to open large leads back then compared to now, but because this period had Gretzky, Lemieux, Esposito, and peak Hull, and the 1997-2017 does not have as many great goal-scorers.
Looking at #2 leads over #10 and comparing them across eras avoids this criticism, though averaging #1, #2, and #3 can at least greatly attenuate this concern as well.
That seems to be an interesting observation, because it suggests that leads of #3 over #10 (and, presumably, #5 over #10) are more comparable across eras and probably should not be adjusted by the same multiplier we apply to peak seasons. On the other hand, the difference in the adjustment factor seems to be minor, and probably we are close to the point where the measurement error of the method is greater than the extra adjustments we can make (I mean, if we are adjusting O6-era leads by multiplying them by 0.6, we are not sure it is 0.6 precisely, it is somewhere in the ballpark - but then 0.65 is somewhere in the ballpark too, and we will never know which one it is, so changing the adjustment factor to 0.65 for some or all leads may be not that important).
I think it is beneficial to have the Adjusted % Leads Over 10th model. What I would like to do next is create a new Adjusted % Leads Over 15th model, although I am also contemplating a different approach.
As I said, the goal-scoring totals on the leader board tend to level out around the 15th scorer on the list. Sometimes, it evens out a few spots later.
Instead of setting the benchmark specifically at 15th place, perhaps it would be a worthwhile endeavor to average out the goal totals of the 15th to 20th placed scorers and use that as a benchmark instead for each individual season (i.e., the benchmark would be a different number from season to season -- 100 benchmarks, logged so that we could use them in future calculations).
Let's look very quickly at what I mean when I say that the scoring levels out around 15th. Here are the 10th to 20th places on the goal-scoring leader boards from 1919-20, 1929-30, 1939-40, 1949-50, 1959-60, 1969-70, 1979-80, 1989-90, 1999-00, and 2009-10 (scroll down for the list).
We see that this happens every year. The scoring levels out among many players. 15th to 20th is, in almost all cases, the placement of the pool of players that the top goal scorers have entirely separated themselves from. The difference between 15th and 20th place is almost always one or two goals.
If we were to use the average of that 15th-to-20th pool as the benchmark instead of a a static 15th-placed finish, I think the calculations would yield even more accurate measurements of dominance over one's peers. We would be measuring them against the "average" NHL goal scorer in each respective season.
The only other consideration would be how we would want to determine a multiplier to convert numbers across eras. We could, once again, average out the multipliers for the 1st place, 2nd place, and 3rd place finishes. This was the method we used with the Adjusted % Leads Over 10th model partly because, in the case of the 1970-1997 era, the 3rd-placed % Lead Over 10th was 100% of the corresponding figure in the modern era. Using the average of the top three finishes allows us to incorporate data about the trend of dominance within an era into our conversions.
1919-20:
10. Punch Broadbent* • OTS 19
11. Cy Denneny* • OTS 16
Sprague Cleghorn* • OTS 16
13. Harry Cameron* • 2TM 15
14. Didier Pitre* • MTL 14
15. Thomas McCarthy • QBC 12
16. George Carey • QBC 11
Bert Corbeau • MTL 11
Babe Dye* • TRS 11
Mickey Roach • TRS 11
20. Ken Randall • TRS 10
Average: 11
1929-30:
10. Pit Lepine • MTL 24
Bun Cook* • NYR 24
12. Ace Bailey* • TOR 22
13. Hooley Smith* • MTM 21
Johnny Gottselig • CBH 21
Frank Finnigan • OTS 21
Baldy Cotton • TOR 21
17. Charlie Conacher* • TOR 20
Herbie Lewis* • DTC 20
19. Aurele Joliat* • MTL 19
Butch Keeling • NYR 19
Average: 20.25
1939-40:
10. Murray Armstrong • NYA 16
11. Kilby MacDonald • NYR 15
12. Syd Howe* • DET 14
13. Syl Apps* • TOR 13
Bill Cowley* • BOS 13
Dutch Hiller • NYR 13
16. Tom Anderson • NYA 12
Doug Bentley* • CBH 12
Busher Jackson* • NYA 12
Lynn Patrick* • NYR 12
Lou Trudel • MTL 12
Average: 12.375
1949-50:
9. Max Bentley* • TOR 23
Ted Lindsay* • DET 23
Paul Ronty • BOS 23
12. Sid Smith • TOR 22
Edgar Laprade* • NYR 22
14. Doug Bentley* • CBH 20
Ted Kennedy* • TOR 20
Bert Olmstead* • CBH 20
17. Milt Schmidt* • BOS 19
Billy Reay • MTL 19
Bud Poile* • 2TM 19
Tony Leswick • NYR 19
Leo Gravelle • MTL 19
Harry Watson* • TOR 19
Average: 19.3333333333333
1959-60:
10. Norm Ullman* • DET 24
Bob Pulford* • TOR 24
Jean-Guy Gendron • BOS 24
13. George Armstrong* • TOR 23
14. Gary Aldcorn • DET 22
Dickie Moore* • MTL 22
16. Phil Goyette • MTL 21
17. Don McKenney • BOS 20
Doug Mohns • BOS 20
Murray Oliver • DET 20
Tod Sloan • CBH 20
Average: 20.71428571428571
1969-70:
8. Dave Balon • NYR 33
Red Berenson • STL 33
Bobby Orr* • BOS 33
11. Jean Ratelle* • NYR 32
Jacques Lemaire* • MTL 32
Dave Keon* • TOR 32
14. John Bucyk* • BOS 31
Gordie Howe* • DET 31
16. Pit Martin • CBH 30
17. Bill Collins • MNS 29
Phil Goyette • STL 29
Danny Grant • MNS 29
John McKenzie • BOS 29
Average: 29.71428571428571
1979-80:
10. Steve Shutt* • MTL 47
11. Blair MacDonald • EDM 46
12. Rick Martin • BUF 45
13. Mike Rogers • HAR 44
14. Real Cloutier • QUE 42
Al MacAdam • MNS 42
Steve Payne • MNS 42
Bryan Trottier* • NYI 42
18. Gilbert Perreault* • BUF 40
Kent Nilsson • ATF 40
Rick Middleton • BOS 40
Peter McNab • BOS 40
Lanny McDonald* • 2TM 40
Bill Barber* • PHI 40
Darryl Sittler* • TOR 40
Average: 40.72727272727273
1989-90:
9. Joe Nieuwendyk* • CGY 45
Mark Messier* • EDM 45
Mario Lemieux* • PIT 45
Mike Gartner* • 2TM 45
13. Pat Verbeek • HAR 44
14. John Ogrodnick • NYR 43
15. John MacLean • NJD 41
Dino Ciccarelli* • WSH 41
17. Dave Andreychuk* • BUF 40
Dave Gagner • MNS 40
Wayne Gretzky* • LAK 40
Steve Thomas • CHI 40
Pierre Turgeon • BUF 40
Average: 40.28571428571429
1999-00:
9. Milan Hejduk • COL 36
Luc Robitaille* • LAK 36
11. Steve Yzerman* • DET 35
Patrik Elias • NJD 35
Mariusz Czerkawski • NYI 35
Valeri Bure • CGY 35
15. Jeremy Roenick • PHX 34
16. Miroslav Satan • BUF 33
Teemu Selanne* • MDA 33
18. Mats Sundin* • TOR 32
19. Chris Simon • WSH 29
Glen Murray • LAK 29
Shawn McEachern • OTT 29
Jarome Iginla • CGY 29
Marian Hossa • OTT 29
Jonas Hoglund • TOR 29
Ray Whitney • FLA 29
Average: 30.45454545454545
2009-10:
10. Alexandre Burrows • VAN 35
Bobby Ryan • ANA 35
12. Anze Kopitar • LAK 34
13. Rick Nash • CBJ 33
Jeff Carter • PHI 33
Nicklas Backstrom • WSH 33
16. Jarome Iginla • CGY 32
Dustin Penner • EDM 32
18. Mike Richards • PHI 31
19. Matt Moulson • NYI 30
Phil Kessel • TOR 30
Patrick Kane • CHI 30
Jussi Jokinen • CAR 30
Patric Hornqvist • NSH 30
Mikael Samuelsson • VAN 30
Average: 31.16666666666667