Serge Bernier’s offensive value
I got the impression that I was being challenged/questioned on my assertion about Bernier so I thought I’d take some time to put a WHA comparison system (and what I’ve said about Bernier) to the test. In the past in this section, I have seen WHA conversion factors of anywhere from 0.60 to 0.75 suggested, and both were based on some degree of research and real-life results, not just gut feelings. That said, they also come pretty close to matching the gut feelings of most of us in this section, from what I can tell. However, that’s a fairly wide range, so maybe I can narrow it down.
I recall in 2010 I did a study where I compared the career PPG of players who played a certain number of games in both the WHA and NHL, and averaged out the decline in point totals that were typically observed. It was the first time I had seen such a thing done, and it yielded a result closer to that 0.60 mark; however, I can’t find the spreadsheet and I think I only publicly released the final results as opposed to “showing my work”. Also, IIRC, it was not that scientific as it was based on career totals; I think specifically targeting the seasons directly before/after a player’s WHA stint makes more sense even if it is more difficult.
I took a long list of players to determine the appropriate factor to use. To qualify, a player had to have at least two WHA seasons if they were primarily an NHL player, and at least two NHL seasons if they were primarily a WHA player. They had to be a player typically discussed in this section (arguably ATD/MLD/AAA/AA caliber) and had to be a forward, as forwards can be judged primarily on offensive numbers, and defensemen can’t (and their offensive totals are hugely dependent on whether they got PP time in a given season or not). Players whose positional integrity was uncertain were dropped to be safe (Mark Howe, for example). If a player went “NHL-WHA-NHL” then I used their last two in the NHL, first two in the WHA, last two in the WHA, first two in the NHL. If their career started or ended with the WHA, then there is only one point of comparison instead of two, so I used their first/last two in the respective leagues. If one of the “adjacent” seasons was less than a half season I added the next one in as well, to ensure I was getting a solid sample size. The smaller of the two samples (either WHA or NHL) had to be at least 125 games; if it wasn’t, I would take the next season to make it at least that large. If that was not possible, the player was thrown out (Tom Webster, Gordie Howe). I also left out Wayne Gretzky because he is an extreme outlier in just about any model and this one is no exception.
After collecting who I could off the top of my head, I found the rest of these players by scanning the top-50 all-time leaders for career WHA GP and Pts, then looking at the year by year all-star teams and top-20 in points. Then I looked at the “skater statistics” pages on WHA, checking the oldest and youngest blocks of players for guys who were working their way up to the NHL, or taking it easy after their prime was over. I hope I got everyone. If I didn’t, please let me know.
Name | NHL GP | NHL Pts | WHA GP | WHA Pts | NHL PPG | WHA PPG | Diff
Tardif | 272 | 225 | 303 | 409 | 0.83 | 1.35 | 0.61
Keon | 308 | 221 | 301 | 291 | 0.72 | 0.97 | 0.74
Walton | 292 | 247 | 211 | 281 | 0.85 | 1.33 | 0.64
Dudley | 303 | 173 | 270 | 277 | 0.57 | 1.03 | 0.56
Hull | 183 | 206 | 253 | 373 | 1.13 | 1.47 | 0.76
Bernier | 223 | 145 | 273 | 368 | 0.65 | 1.35 | 0.48
Lacroix | 158 | 70 | 312 | 436 | 0.44 | 1.40 | 0.32
Ftorek | 145 | 131 | 293 | 406 | 0.90 | 1.39 | 0.65
Houle | 290 | 186 | 214 | 257 | 0.64 | 1.20 | 0.53
Stoughton | 272 | 227 | 219 | 179 | 0.83 | 0.82 | 1.02
Henderson | 212 | 159 | 279 | 235 | 0.75 | 0.84 | 0.89
Cloutier | 168 | 217 | 150 | 258 | 1.29 | 1.72 | 0.75
K.Nilsson | 160 | 222 | 158 | 214 | 1.39 | 1.35 | 1.02
Nilsson | 160 | 163 | 144 | 250 | 1.02 | 1.74 | 0.59
Hedberg | 160 | 149 | 145 | 253 | 0.93 | 1.74 | 0.53
Flett | 199 | 95 | 195 | 187 | 0.48 | 0.96 | 0.50
Rogers | 160 | 210 | 160 | 143 | 1.31 | 0.89 | 1.47
McKenzie | 142 | 146 | 179 | 190 | 1.03 | 1.06 | 0.97
Mahovlich | 149 | 173 | 148 | 171 | 1.16 | 1.16 | 1.00
Lukowich | 158 | 141 | 160 | 174 | 0.89 | 1.09 | 0.82
MacDonald | 143 | 151 | 160 | 139 | 1.06 | 0.87 | 1.22
Preston | 202 | 125 | 153 | 110 | 0.62 | 0.72 | 0.86
Nedomansky | 143 | 101 | 174 | 172 | 0.71 | 0.99 | 0.71
Ruskowski | 146 | 129 | 153 | 158 | 0.88 | 1.03 | 0.86
Bordeleau | 144 | 84 | 153 | 176 | 0.58 | 1.15 | 0.51
Williams | 198 | 131 | 139 | 89 | 0.66 | 0.64 | 1.03
Connelly | 147 | 86 | 156 | 165 | 0.59 | 1.06 | 0.55
Lawson | 156 | 39 | 156 | 194 | 0.25 | 1.24 | 0.20
MacMillan | 182 | 113 | 153 | 88 | 0.62 | 0.58 | 1.08
Ullman | 158 | 104 | 144 | 130 | 0.66 | 0.90 | 0.73
Backstrom | 155 | 110 | 148 | 122 | 0.71 | 0.82 | 0.86
Napier | 130 | 80 | 159 | 161 | 0.62 | 1.01 | 0.61
Tonelli | 150 | 100 | 145 | 119 | 0.67 | 0.82 | 0.81
MacGregor | 193 | 110 | 135 | 75 | 0.57 | 0.56 | 1.03
Joyal | 139 | 62 | 194 | 103 | 0.45 | 0.53 | 0.84
McDonald | 171 | 107 | 147 | 70 | 0.63 | 0.48 | 1.31
Hampson | 156 | 59 | 154 | 117 | 0.38 | 0.76 | 0.50
Boudrias | 149 | 106 | 140 | 70 | 0.71 | 0.50 | 1.42
Gendron | 132 | 55 | 127 | 69 | 0.42 | 0.54 | 0.77
Fonteyne | 138 | 32 | 149 | 61 | 0.23 | 0.41 | 0.57
Semenko | 125 | 32 | 142 | 36 | 0.26 | 0.25 | 1.01
Total | 7371 | 5422 | 7548 | 7776 | 0.74 | 1.03 | 0.71
It’s not particularly hard to put together something even more “scientific” than this; however, this is a great improvement on what I did before and it appears intuitively sound, fitting in with what the consensus tends to be about WHA offensive totals.
A couple of notes:
- I probably should have included all players, not just the ones who were at least AA draft caliber. I’m not interested in going back and finding them all, but to be honest I can’t quite justify why their results should not be relevant in a study of how much easier it is to score in the WHA than the NHL. These players would tend to drag that 0.71 figure down, as they are guys who were not special NHL players, hence why we’ve never selected them here.
- Another factor that would technically drag this 0.71 down is that so many of these players played at ages where we know there were likely not in their offensive primes – either quite old or quite young. And then they’re being compared to NHL seasons that were closer to, or in, their prime. If an age-adjusted analysis was possible it would surely drag the final answer down just a bit.
- If you were to “weigh” each player’s factor based on the number of games played and points (i.e. which samples are more significant) the final factor that the table would determine would be 0.76. However, these were arbitrarily selected samples to begin with so I didn’t see the point in adding that factor into it. It could be argued that even if this is valid, the two points above offset that to about the same degree.
- My conclusion as of now (and I welcome comments and criticisms) is that WHA scoring achievements are worth about 71% what NHL scoring achievements in the 1970-1982 range are.
With that all done, how does this apply to Bernier?
What we tend to use for NHL players is a Vs2, VsX, or “percentage” method, whatever you want to call it, there are a couple systems in use but they have the same principle to them. Most of us tend to speak in terms of “best 6 scores” and though I try not to adhere to that too strictly in cases of outstanding (or poor) consistency/longevity, it has proven to be a solid barometer of offensive peak.
Bernier’s best NHL scores by my methodology are: 68, 60, 42. If I was to calculate his WHA percentage scores I would do the following:
1974: 86/100 = 86 (both 1st and 2nd are outliers this year)
1975: 122/122 = 100 (used Bernier as the benchmark as 1st and 2nd are outliers from him)
1976: 102/123 = 83
1977: 96/131 = 73
1978: 78/129 = 60
1979: 82/116 = 71
Multiplied by 0.71 to account for the relative ease in posting these figures, Bernier’s 6 WHA percentage scores would be 61, 71, 59, 52, 43, and 50.
So his best six percentage scores with his NHL days also counted, would be 71, 68, 61, 60, 59, 52, for a total of 371. That’s pretty strong, and no one available comes close to this (I see 350 or so as the gold standard heading into the AA) so it’s safe to say that Bernier absolutely belongs at least in the AAA. But where?
Ignoring for now the facts that there are mitigating factors leading to some of the percentage scores of these wingers, both good and bad (some of them achieved these scores despite injuries cutting some of their best seasons a bit short, some of them benefitted from a clearly superior center), here are the best north American AAA scoring line wingers based on the same metric. Everyone listed here is currently on a AAA scoring line, except for Havlat and McDonald, the only two wingers I can see on the bottom half of anyone’s roster who have the offensive credentials to hang with these guys.
Pre-expansion players have a 15% boost added to their score. This has been discussed many times and not everyone agrees with it; however, I still believe it intuitively makes sense as it tends to reflect at any given time in the MLD or AAA what the best available post-expansion players seem to outpace the best available pre-expansion players by. (no one disagrees that VsX will favour the modern player from a bigger league, particularly when we’re at this caliber of player).
Note that Boldirev is mostly a center and should be discounted to some degree based on personal preference. Note that starting around Christmas, Kessel will be making his “best 6” score go up by one every time he scores another point, as his 2014 season will eclipse his 2008 season as his sixth best. Also, note that Tobin’s score of 373 really means “at least 373” because he played defense 20% of the time (though I am not sure how much defense he played in the seasons identified as his six best, it could be none and in all likelihood it is lower than his career average of 20%) Fourthly, note that Liscombe is adjusted for WW2 so as not to give him too much, or too little credit for those seasons. And lastly, I should mention that I see three other wingers in the AAA who, despite being next to impossible to compare straight up, certainly belong somewhere on this list – Golikov, Richardson and Routh.
Also please do note that this is not meant to be a definitive ranking of AAA wingers, for one thing it only takes offensive numbers into account, doesn’t consider playoffs, doesn’t consider who was helped by superior linemates versus being a catalyst themselves, and it’s only based on six seasons. Players who are in the same range for best 6 seasons should be given a closer look, and the most important thing to look at is probably what else happened outside those six seasons. For a guy like Earl Robinson, the answer is absolutely nothing, for a guy like Bernier, it’s decent, and for Kehoe and Boldirev it’s quite a lot. On the other side of the coin, it weighs a player’s best couple of seasons only as highly as their fifth and sixth best, and from what I can tell, most of us do place a higher value on how good a player was at his very best (see: the assertion of Kessel as possibly the top winger here, the fact that Neal and Kunitz were selected at all, the emphasis on one season of a few all-star votes for a defenseman)
Anyway, here’s the list:
Sorrell | 435
Murray | 418
Grant | 415
Robinson | 403
McDonald | 392
Liscombe | 391
Kaleta | 389
Brunette | 382
Kessel | 381
Poile | 377
Havlat | 374
Kehoe | 374
Tobin | 373
Bernier | 371
Courtnall | 370
Boldirev | 370
Sanderson | 370
R.Sheppard | 367
Depending on some other factors, it’s relatively safe to say Bernier is the 10th-21st best scoring winger in the AAA. Considering there are 14 teams and therefore 28 scoring line wingers, even after accounting for some complementary players, Bernier is a good fit as an AAA 1st line winger – but also not by any means a special one. Somewhere between average and bargain basement. It certainly would not be a misappropriation of votes to put him on the 3rd all-star team for this draft… 2nd is pushing it, and 1st team I don’t see as a particularly defendable when you look at the other names there. I think you’ve got him more or less where he belongs.