Era Adjustment and VHL Defensemen

[Alex 3,248]

So, with the conversation around my previous article, I set out on a mission to figure out a way to adjust for era. After a few hours exploring the internet, the only accurate way to adjust for era within the VHL revealed itself, and was fairly complicated, but I will try to explain it below:

 

Step 1: Identify each season’s top 5 point scorers and the number of points they scored.

Step 2: Average out those points for each season.

Step 3: Set up a 5 season rolling average for each season, this reduces the impact of outliers. For example, the number used for season 3 was the average of seasons 1 through 5, two seasons each way. For seasons 1 and 96 this was 2 seasons forward and back respectively.

Step 4: Determine the overall average number of points for each season.

Step 5: Divide a season’s rolling average (step 3) by the overall average to give each season a ratio.

Step 6: Divide a player’s total points by the average ratio for the seasons they played in.

 

The main issue with this is that defensemen scoring doesn’t necessarily mimic overall scoring, however we can assume that an increase to the top end of scoring does mean that defensemen scoring would increase, or at a minimum would have had the potential to increase. What this did do is clearly identify 3 kind of well-known peaks in league scoring, specifically from seasons 1-10 (by far the largest), S50-S59, and from S79-S83 (the meta era). The interesting thing here is that while players who played in the meta era did have their totals drop, it really wasn’t by much, given that they would have played on either side of it as well, which were both below average scoring seasons. It is worth noting that in today’s VHL while we do still have a bunch of players crossing the 90-point mark, at the top end it is a below average scoring era in the VHL (by a hair). From what I’ve seen the reasoning here is simply because we do have a bunch of high earning players, who will produce offensively, but will be unlikely to ever hit the high marks set in the earlier seasons. Since the goalie change a few seasons ago scoring has been trending towards average. Please see my pretty graph below:

 

 

I then went through this with the top 50 scoring defensemen plus a couple Hall of Famers from the S60-S70 window as per Victor’s request. Of the 54 players assessed 25 saw a increase in their point total, and 29 players had their point totals decrease as a result of era adjustment. All of the top 5 players to increase played between S65 and S78, the player that saw the largest increase was Condor Adrienne who played from S68 to 75, which would have had the middle of their career in the early 70’s which was the lowest scoring era in VHL history by a massive margin.

 

Largest Increase:

1.      Condor Adrienne (712 +114) @OrbitingDeath

2.      Smitty Werbenjagermanjensen (637 +104) @Fire Tortorella

3.      Lincoln Tate (611 +103) @Tate

4.      Erik Killinger (658 +68) @Peace

5.      Roque Davis (638 +66) @Josh

 

As for players seeing their points decrease when adjusted for era, they all played within the first 11 seasons of the VHL, notably when essentially every offensive record was set, so that makes sense. Most notably here is that Sterling Labatte saw the largest decrease as their point total dropped by 103 points.

 

Largest Decrease:

1.      Sterling Labatte (669 -103) @sterling

2.      Shawn Crowley (519 -82)

3.      Danny Tremblay (506 -80)

4.      Joey Kendrick (483 -78) @Kendrick

5.      Jochen Walser (574 -43) @marshall_222

 

We did also see some considerable movement in the defensemen scoring leaderboard when we adjust for era, namely that Skor McFleury goes up to the number 1 spot, Pierre Emile Bouchard jumps into the 2^nd spot, Japindeer Singh is the only player to have their point total decrease and remain in the top 5, and Condor Adrienne moves up to 4^th.

 

Era Adjusted Scoring Leaders:

1.      Skor McFleury (768 +50) @Alex

2.      Pierre Emile Bouchard (726 +56) @Gaikoku-hito

3.      Japindeer Singh (714 -37) @8Ovechkin8

4.      Condor Adrienne (712 +114) @OrbitingDeath

5.      Bo Johansson (711 +49) @Shindigs

 

With that being said I made the decision to re-work the formula to determine a statistical score, essentially using the VHFL scoring but inserting era adjusted scoring instead of assists and keeping the initial goal totals to still have more weight there, and because I don’t have the time to do more math. I did drop the weight of a goal to 1.5 (instead of 3) since they are counted in the era adjusted point totals already.

 

Era Adjusted Stat Rank:

1.      Skor McFleury @Alex

2.      Brian Payne @Scurvy

3.      Hard Markinson @STZ

4.      Condor Adrienne @OrbitingDeath

5.      Alexander Valiq @Koradek

6.      Jake Thunder @Thunder

7.      Pierre Emile Bouchard @Gaikoku-hito

8.      Japinder Singh @8Ovechkin8

9.      Battre Sandstrom @Acydburn

10. Daniel Braxton @Jericho

 

As for awards through conversations with Victor I made the decision to adjust the weight of individual awards to better show their overall value (Continental Cup = 100, Labatte/MVP = 75, ROTY/Funk = 35, Other = 50), since this changes things slightly, here’s your top 10.

 

Award Leaders:

1.      Conner Low @Smarch

2.      Condor Adrienne @OrbitingDeath

3.      Sterling Labatte @sterling

4.      Phil Hamilton @Phil

5.      Smitty Werbenjagermanjensen @Fire Tortorella

6.      Jochen Walser @marshall_222

7.      Skor McFleury @Alex

8.      Ryan Sullivan @Advantage

9.      Elijah Incognito @Strummer

10. Daniel Braxton @Jericho

 

As for overall ranking I made the decision to add a bonus for statistical leaders (actual goals, assists, points, hits, and SB) and players with awards named after them, giving said players a bonus of 100 (maximum of 1 bonus) as kind of an impact/legacy bonus. The top 10 defensemen with this era adjusted and improved system are:

 

1.      Condor Adrienne @OrbitingDeath

2.      Skor McFleury @Alex

3.      Sterling Labatte @sterling

4.      Phil Hamilton @Phil

5.      Brian Payne @Scurvy

6.      Smitty Werbenjagermanjensen @Fire Tortorella

7.      Hard Markinson @STZ

8.      Japinder Singh @8Ovechkin8

9.      Jake Thunder @Thunder

10. Daniel Braxton @Jericho

 

Lastly since the conversation around last article, Conner Low ends up with 478 era adjusted points (+4), ends up 53^rd statistically, 1^st for awards, gains a legacy bonus for having the most Labattes, and ends up 14^th overall.

 

Thank you for reading, while the era adjustment formula is sound, if you have any comments or suggestions to improve the scoring system I’m all ears!

 

1041 words (2 weeks)

 

[Ahma 1,444]

Sorry but Labatte still clears!

Edited November 29 by Ahma

[Gustav 6,469]

1 hour ago, Ahma said:

Sorry but Labatte still clears!

 

Yeah and honestly putting Low at #14 is probably even harder evidence that numbers aren't perfect. Obviously I understand that's true in a general sense outside of this article--my last one ranked players based on nothing but my own personal judgment of how my own made-up metric should work, so that isn't a criticism of this one. I'm not about to go around telling everyone that McFleury was the second-greatest of all time (sorry).

 

This is still really fun, though! Props for the era adjustment because that's something I've always been too afraid to try quantitatively.

[Victor 11,015]

Just to copy some of my thoughts while talking with Alex below. I do echo Gus' point that the era adjusting formula is a good thing and looks to have caught the dips and peaks accurately enough. Except maybe the post meta S80s which I really don't think were low scoring.

 

- The thing is though, is it different players lead the scoring but the scoring still roughly the same? Looking through Valiq Trophy winners, with a few exceptions it's always 90-100 points. Ergo if you're a high scoring defenceman you're still hitting the same mark as guys in the past.

 

- The dip on the graph from I think S85 to S93 - there were still 10 players with 100 points per season. That's the same as S65 which is above the average? Something feels off. Much like Markinson and Sandstrom still scoring high despite being meta boosted.

 

I do want to see adjusted scoring totals per game as well.

[Dil 1,763]

this may be the saddest fucking copium I have ever seen in my entire life give it a rest your player was mid

[Alex 3,248]

4 hours ago, Gustav said:

Yeah and honestly putting Low at #14 is probably even harder evidence that numbers aren't perfect. Obviously I understand that's true in a general sense outside of this article--my last one ranked players based on nothing but my own personal judgment of how my own made-up metric should work, so that isn't a criticism of this one. I'm not about to go around telling everyone that McFleury was the second-greatest of all time (sorry).

The general idea was to make it purely objective, which is why low ends up low. Essentially just because they had approximately 250 less points in an almost dead on average scoring era. There’s no doubt that they were the most dominant defensemen ever, but their shortened time at the position results in their lower stats. I could maybe try adjusting the weight of awards to be exactly 50% of overall, would just be more math and the numbers would look weird.

 

3 hours ago, Victor said:

The thing is though, is it different players lead the scoring but the scoring still roughly the same? Looking through Valiq Trophy winners, with a few exceptions it's always 90-100 points. Ergo if you're a high scoring defenceman you're still hitting the same mark as guys in the past.

I think a large part of the issue here is that the general assumption is that if the highest scoring players are scoring more the highest scoring defensemen at least have the potential to as well. Which while an assumption should be correct in this case.

 

3 hours ago, Victor said:

The dip on the graph from I think S85 to S93 - there were still 10 players with 100 points per season. That's the same as S65 which is above the average? Something feels off. Much like Markinson and Sandstrom still scoring high despite being meta boosted.

So as mentioned because the meta period was relatively brief, it did effect their era adjusted scoring, but the impact was dampened due to both playing on either side of it.

 

 Another issue is that while we do have so many players scoring over 100 points, it could be feasible that it’s simply due to their being so many high earning players now compared to before? The main point is that those high scoring players are not scoring near as much as early players or meta.

 

It is also worth noting that the early seasons were so high scoring that it skews the entire ratio system, which makes them all outliers and they should just be excluded. This is why Labatte drops so much, but it’s not wrong that the entire reason they lead defensemen in scoring is because of the era they played in.

Edited November 29 by Alex

[OrbitingDeath 3,228]

#1 

[FrostBeard 2,061]

Condor is where it is at!

[Gustav 6,469]

5 hours ago, Alex said:

Another issue is that while we do have so many players scoring over 100 points, it could be feasible that it’s simply due to their being so many high earning players now compared to before? The main point is that those high scoring players are not scoring near as much as early players or meta.

 

My (unproven) hypothesis on this point is that we just don't have a super packed league anymore. That's not a bad thing, but having a handful of teams that are far from full gives more opportunities for other players to beat up on those teams.

 

It's also likely that high-level players generally aren't competing for ice time right now and are more clearly the "best" on their teams--of course Simon will default to giving someone with 1100 TPA the points over their linemate who could be an IA roster filler or even a bot.

 

I think the early S50s is an extreme example of this, where a handful of both skaters and goalers were putting up insane numbers just because the league was empty and they were better than all the players around them on either end (and probably average only against each other). I would rule out both the start of the league (where the "best" point was really wherever the S1 players were for their entire careers) and meta (which was kind of its own situation) as fair comparison points. Early S50s is a fair example IMO, as is that huge dip in your graph in the late S60s where we were pre-meta and rosters were fuller than ever. We had a couple seasons around then where no one hit 100 points--which I'm sure would make some people today lose their minds. IMO that's because of roster sizes--I'd venture a guess that we'll be able to point out that effect again with hybrid attributes in place to limit meta.

[Lemorse7 1,021]

cough bias cough XD

[Alex 3,248]

Week 2

[Victor 11,015]

On 11/29/2024 at 12:06 AM, Alex said:

Era Adjusted Stat Rank:

1.      Skor McFleury @Alex

2.      Brian Payne @Scurvy

3.      Hard Markinson @STZ

4.      Condor Adrienne @OrbitingDeath

5.      Alexander Valiq @Koradek

6.      Jake Thunder @Thunder

7.      Pierre Emile Bouchard @Gaikoku-hito

8.      Japinder Singh @8Ovechkin8

9.      Battre Sandstrom @Acydburn

10. Daniel Braxton @Jericho

Just coming back to this because I was updating career stats so naturally looking at things including defencemen.

 

Something, and I am very confident it's blocked shots, is throwing the entire formula because 7 out of the 10 best statistical defenceman can't have played in the last 30 seasons.

 

89 out of 120 players with 1,000+ career SBs (and 41 of the top 50) have retired in the last 30 seasons so that requires era adjusting as well without a doubt.

[Alex 3,248]

1 hour ago, Victor said:

89 out of 120 players with 1,000+ career SBs (and 41 of the top 50) have retired in the last 30 seasons so that requires era adjusting as well without a doubt.

I don't have the time to check this right now, but I do think SB has increased over the past so many seasons, theoretically it could be due to the increase in the number of players? Especially when you consider that STHS really hasn't changed over the seasons. More high earning players = more shots = more shots blocked?

[Victor 11,015]

11 hours ago, Alex said:

I don't have the time to check this right now, but I do think SB has increased over the past so many seasons, theoretically it could be due to the increase in the number of players? Especially when you consider that STHS really hasn't changed over the seasons. More high earning players = more shots = more shots blocked?

I would have expected shots and save percentage to go up as well then. We did have the save % spike but that was a brief thing at the start of the hybrid era when goalies were overpowered.

 

My theory is its more to do with the balance of teams so even if you're a defenceman on a good team you're facing more competitive teams. Either way I don't think that makes modern defencemen better that they have that opportunity. 

[JCarson 744]

Not sure if the article itself or the following discussion makes this stream worth the read.  unfortunately due to so many factors I am not sure it is possible to do a good comparison across all players across all time.  It is like that in the NHL as well.  Let's just take Ovechkin vs Gretzky for all time scoring.  Now I expect as long as Ovechkin can stay healthy that he will beat Gretzky's scoring record but is it really comparable.  The game itself has changed drastically between the two era's.  Even if you look at only the stick.  Players now can have custom curves on their stick, Gretzky played in an era where you could get a penalty if your stick was outside certain guidelines.  Lets then discuss wood versus composite.  How better would have Gretzky been if he had today's sticks.  Sometimes you just have to step back and say they are both great scorers.

 

I give this article a 9 out of 10 for attempting something most people would avoid.