This post has gone through a bunch of variations, I thought I had a good format but the tables turned out to be way too wide, so I had to re-write it to try and fit it better. I also apologize for the length, and the number of tables in it. It initially started off from daver asking for a comparison of Brett Hull's 86 goal season to Ovechkin's 65 goal season, which once you have the ancillary aspects of the model nailed down, is fairly easy to do.
Year | GP | GF | GF/G | PGP | G | A | Pts | G% | P% | LAGF | LAGF/G | %LA |
90-91 | 80 | 310 | 3.875 | 78 | 86 | 45 | 131 | 0.277 | 0.423 | 276 | 3.45 | 1.123 |
[TBODY]
[/TBODY]
For a glossary - GP is number of games in that season, GF is the number of goals that player's team scored, GF/G is per game, PGP is the number of games that player played in, G/A/Pts are his counting stats, G% is the percentage of goals that player scored of his team's total, P% is the percentage of points that player accrued on goals his team scored, LAGF is league average goals for that season, LAGF/G is per game, and %LA is the percentage above or below league average the team scored that season.
Once you have your inputs, to place that Hull season in 07-08, 08-09, or 09-10 is just a matter of editing LAGF, like so.
Year | GP | LAGF | LAGF/G | New GF | NGF/G | G | A | Pts |
07-08 | 82 | 223 | 2.720 | 250.471 | 3.055 | 69.5 | 36.4 | 105.8 |
08-09 | 82 | 234 | 2.854 | 262.826 | 3.205 | 72.9 | 38.2 | 111.1 |
09-10 | 82 | 227 | 2.768 | 254.964 | 3.109 | 70.7 | 37.0 | 107.7 |
[TBODY]
[/TBODY]
Further glossary - New GF is generated by multiplying LAGF by %LA. Hull's new point total is generated by taking that New GF number, and applying (in Hull's case) 27.7% of goals and 42.3% of points.
Now you can take these numbers in two different ways - the first being to try to assess whether or not 70+36=106 would be adjudged to be more valuable than 65+47=112 by MVP voters in 07-08, or the second to accept that Hull's goal scoring season is marginally more impressive than Ovechkin's, but again quite closer than the raw 86-65 total.
You don't need to stick to single seasons either, you could input multiple years and output multiple years, for example if you wanted to compare Wayne Gretzky's 3 year peak to Ovechkin's 3 year peak.
Input:
Year | GP | GF | GF/G | PGP | G | A | Pts | G% | P% | LAGF | LAGF/G | %LA |
81-82 | 80 | 417 | 5.213 | 80 | 92 | 120 | 212 | 0.221 | 0.508 | 321 | 4.013 | 1.299 |
82-83 | 80 | 424 | 5.3 | 80 | 71 | 125 | 196 | 0.167 | 0.462 | 309 | 3.863 | 1.372 |
83-84 | 80 | 446 | 5.575 | 74 | 87 | 118 | 205 | 0.195 | 0.460 | 316 | 3.95 | 1.411 |
[TBODY]
[/TBODY]
Output:
Year | GP | LAGF | LAGF/G | New GF | NGF/G | G | A | Pts |
07-08 | 82 | 223 | 2.720 | 289.692 | 3.533 | 63.9 | 83.4 | 147.3 |
08-09 | 82 | 234 | 2.854 | 321.087 | 3.916 | 53.8 | 94.7 | 148.4 |
09-10 | 82 | 227 | 2.768 | 320.386 | 3.907 | 62.5 | 84.8 | 147.3 |
[TBODY]
[/TBODY]
Quite ridiculous numbers, but then Wayne was a ridiculous player. Even though I'm already tackling outlier seasons, I feel like my model breaks down when looking at the most outlier of outlier seasons. But you still can throw a whole bunch of peak forward seasons as input - Ovechkin 07-08 as your benchmark, Howie Morenz 27-28, Maurice Richard 44-45, Gordie Howe 52-53, Bobby Hull 65-66, Phil Esposito 70-71, Mike Bossy 81-82, and Mario Lemieux 88-89, and see what they would look like in 07-08.
Input:
| GP | GF | GF/G | PGP | G | A | Pts | G% | P% | LAGF | LAGF/G | %LA |
07-08 | 82 | 238 | 2.902 | 82 | 65 | 47 | 112 | 0.273 | 0.471 | 223 | 2.720 | 1.067 |
27-28 | 44 | 116 | 2.636 | 43 | 33 | 18 | 51 | 0.284 | 0.440 | 84 | 1.909 | 1.381 |
44-45 | 50 | 228 | 4.560 | 50 | 50 | 23 | 73 | 0.219 | 0.320 | 184 | 3.680 | 1.239 |
52-53 | 70 | 222 | 3.171 | 70 | 49 | 46 | 95 | 0.221 | 0.428 | 168 | 2.400 | 1.321 |
65-66 | 70 | 240 | 3.429 | 65 | 54 | 43 | 97 | 0.225 | 0.404 | 213 | 3.043 | 1.127 |
70-71 | 78 | 399 | 5.115 | 78 | 76 | 76 | 152 | 0.190 | 0.381 | 244 | 3.128 | 1.635 |
81-82 | 80 | 385 | 4.813 | 80 | 64 | 83 | 147 | 0.166 | 0.382 | 321 | 4.013 | 1.199 |
88-89 | 80 | 347 | 4.338 | 76 | 85 | 114 | 199 | 0.245 | 0.573 | 299 | 3.738 | 1.161 |
[TBODY]
[/TBODY]
The outputs of Morenz and Esposito are the ones that I'd consider 'off', because even though Morenz's percentages aren't out of line with the others, the fact remains that Montreal's top 3 forwards that year accounted for ~70% of Montreal's goals, whereas by M. Richard's year, his line with Lach/Blake accounted for only 46%. In Esposito's case, the fact that Boston was 63.5% above league average that year is just an absurd number.
Output:
Year | GP | LAGF | LAGF/G | New GF | NGF/G | G | A | Pts |
07-08 | 82 | 223 | 2.720 | 238.000 | 2.902 | 65.0 | 47.0 | 112.0 |
27-28 | 82 | 223 | 2.720 | 307.952 | 3.756 | 87.6 | 47.8 | 135.4 |
44-45 | 82 | 223 | 2.720 | 276.326 | 3.370 | 60.6 | 27.9 | 88.5 |
52-53 | 82 | 223 | 2.720 | 294.679 | 3.594 | 65.0 | 61.1 | 126.1 |
65-66 | 82 | 223 | 2.720 | 251.268 | 3.064 | 56.5 | 45.0 | 101.6 |
70-71 | 82 | 223 | 2.720 | 364.660 | 4.447 | 69.5 | 69.5 | 138.9 |
81-82 | 82 | 223 | 2.720 | 267.461 | 3.262 | 44.5 | 57.7 | 102.1 |
88-89 | 82 | 223 | 2.720 | 258.799 | 3.156 | 63.4 | 85.0 | 148.4 |
[TBODY]
[/TBODY]
Ignoring Morenz and Esposito, Hull's 69.5 goal total still the highest of all the seasons. I'd also like to point out the repeatability of G%/P% - I've shown 10 different peak seasons from 10 different players over almost the entirety of NHL history, in a variety of scoring environments, and yet goal scoring clusters between 20-25% in almost all of them, and point totals from 38-48%.
Finally, despite the fact that I've converted every season here into 07-08 doesn't mean that is the only option. We could in fact take Ovechkin's year and put it into every other year we've chosen to convert his peak year into theirs.
Year | GP | LAGF | LAGF/G | New GF | NGF/G | G | A | Pts |
07-08 | 82 | 223 | 2.720 | 238.000 | 2.902 | 65.0 | 47.0 | 112.0 |
27-28 | 44 | 84 | 1.909 | 89.650 | 2.038 | 24.5 | 17.7 | 42.2 |
44-45 | 50 | 184 | 3.680 | 196.377 | 3.928 | 53.6 | 38.8 | 92.4 |
52-53 | 70 | 168 | 2.400 | 179.300 | 2.561 | 49.0 | 35.4 | 84.4 |
65-66 | 70 | 213 | 3.043 | 227.327 | 3.248 | 62.1 | 44.9 | 107.0 |
70-71 | 78 | 244 | 3.128 | 260.413 | 3.339 | 71.1 | 51.4 | 122.5 |
81-82 | 80 | 321 | 4.013 | 342.592 | 4.282 | 93.6 | 67.7 | 161.2 |
88-89 | 80 | 299 | 3.738 | 319.112 | 3.989 | 87.2 | 63.0 | 150.2 |
90-91 | 80 | 276 | 3.450 | 294.565 | 3.682 | 80.4 | 58.2 | 138.6 |
[TBODY]
[/TBODY]