Simulation Hockey League
Regression Curve - Printable Version

+- Simulation Hockey League (https://simulationhockey.com)
+-- Forum: Community (https://simulationhockey.com/forumdisplay.php?fid=17)
+--- Forum: SHL Discussion (https://simulationhockey.com/forumdisplay.php?fid=49)
+---- Forum: Suggestion Box (https://simulationhockey.com/forumdisplay.php?fid=50)
+---- Thread: Regression Curve (/showthread.php?tid=88338)

Pages: 1 2 3 4 5 6 7


RE: Regression Curve - DeletedAtUserRequest - 09-12-2018

if you want this to work and have it done fairly then use the current situation as the standard... if 25 seasons is the longest career to date.. then have the 25% begin at season 24

Current regression table as is...
24th season +: 25%
26th season+30

this would keep the current historial stats safe from any asterisk as each player moving forward will have the same chance as the longest current players to ever play.. while getting buried by the same new regression table at the same point of there careers.

thats something that i could def back. theres logic and fairness to that.


RE: Regression Curve - King - 09-12-2018

09-12-2018, 07:00 PMMike Izzy Wrote: if you want this to work and have it done fairly then use the current situation as the standard... if 25 seasons is the longest career to date.. then have the 25% begin at season 24

Current regression table as is...
24th season +: 25%
26th season+30

this would keep the current historial stats safe from any asterisk as each player moving forward will have the same chance as the longest current players to ever play.. while getting buried by the same new regression table at the same point of there careers.

thats something that i could def back

Hey man, could you expand on this asterisk argument, I don't think I fully understand it yet.


RE: Regression Curve - Tomen - 09-12-2018

09-12-2018, 07:03 PMKing Wrote:
09-12-2018, 07:00 PMMike Izzy Wrote: if you want this to work and have it done fairly then use the current situation as the standard... if 25 seasons is the longest career to date.. then have the 25% begin at season 24

Current regression table as is...
24th season +: 25%
26th season+30

this would keep the current historial stats safe from any asterisk as each player moving forward will have the same chance as the longest current players to ever play.. while getting buried by the same new regression table at the same point of there careers.

thats something that i could def back

Hey man, could you expand on this asterisk argument, I don't think I fully understand it yet.

I am obviously not Izzy but how I interpret his post is that not everyone will be able to have the same "settings" to achieve the same career goals. But that has never been the case in the SHL if you go back through the history of the SHL. TPE inflation, regression scale changes and Update Scale changes made it that players never had the exact same chances to achieve high career numbers.

You could say that those who created around S17-S25 are in a sweet spot for the highest career numbers because of TPE inflation, they still had the older & easier update scale which allowed for higher attributes and therefore a better in sim performance. S21-S25 thre were 52 games played in a season. Rule changes also happened which limited double shifting star players so overall ice time got less and therefore production.


RE: Regression Curve - Samsung virtual assistant - 09-12-2018

Wayne Gretzky. Needs an asterisk IRL


RE: Regression Curve - DeletedAtUserRequest - 09-12-2018

09-12-2018, 07:03 PMKing Wrote:
09-12-2018, 07:00 PMMike Izzy Wrote: if you want this to work and have it done fairly then use the current situation as the standard... if 25 seasons is the longest career to date.. then have the 25% begin at season 24

Current regression table as is...
24th season +: 25%
26th season+30

this would keep the current historial stats safe from any asterisk as each player moving forward will have the same chance as the longest current players to ever play.. while getting buried by the same new regression table at the same point of there careers.

thats something that i could def back

Hey man, could you expand on this asterisk argument, I don't think I fully understand it yet.

Realism Era (S12-S20; S25-Present)
52-Game Era (S21-S24)
Inflation Era (S9-S11)
Experimental Era (S6-S8)
Dead Puck Era (S3-S5)

these are the 5 eras of the SHL.... if we make a change to the regression table which all but ends a players chance at any historical numbers then the era must be changed to allow that new generation of players to set there own standards under the new rules... (active players caught in between these changes are subject to debate IMO)

But... if we use the longest career (24 seasons) as the template to make these change then theres an inherant fairness to the rules.


RE: Regression Curve - raymond3000 - 09-12-2018

09-12-2018, 06:17 PMDrunkenTeddy Wrote:
09-12-2018, 06:09 PMraymond3000 Wrote: Maybe we can simplify regression altogether and just make the regression percentage equal to the number of seasons played?

Wow.. I said this half an hour ago to King in discord. Only issue is it makes Eggy's life a bit harder when he has to write the regression post. I like the idea of it, but if we are really just trying to tackle the outliers who drag on their career, just making regression go to 30% after season 25 would do it.

Slightly changing topics here, but it can't be too difficult to whip up a spreadsheet that automatically generates the regression list. King already has a list with all the SHL players, their draft year, and their TPE. Of course, someone would still have to manually input TPE totals, but that should make Eggy's life easier.


RE: Regression Curve - DeletedAtUserRequest - 09-12-2018

btw.. im happy that the league is getting back to real SHL discussion.. last few days have been rough reading  Cool


RE: Regression Curve - Sleepy - 09-12-2018

09-12-2018, 07:03 PMKing Wrote:
09-12-2018, 07:00 PMMike Izzy Wrote: if you want this to work and have it done fairly then use the current situation as the standard... if 25 seasons is the longest career to date.. then have the 25% begin at season 24

Current regression table as is...
24th season +: 25%
26th season+30

this would keep the current historial stats safe from any asterisk as each player moving forward will have the same chance as the longest current players to ever play.. while getting buried by the same new regression table at the same point of there careers.

thats something that i could def back

Hey man, could you expand on this asterisk argument, I don't think I fully understand it yet.

I'm sure Izzy will explain, but from my perspective of it, he's concerned that changing the regression scale so much will somehow change the validity of previous record holders and the ability for new players to break those records because the scale affected them differently than players under a previous regression curve.


RE: Regression Curve - DeletedAtUserRequest - 09-12-2018

09-12-2018, 07:14 PMSleepy Wrote:
09-12-2018, 07:03 PMKing Wrote:
09-12-2018, 07:00 PMMike Izzy Wrote: if you want this to work and have it done fairly then use the current situation as the standard... if 25 seasons is the longest career to date.. then have the 25% begin at season 24

Current regression table as is...
24th season +: 25%
26th season+30

this would keep the current historial stats safe from any asterisk as each player moving forward will have the same chance as the longest current players to ever play.. while getting buried by the same new regression table at the same point of there careers.

thats something that i could def back

Hey man, could you expand on this asterisk argument, I don't think I fully understand it yet.

I'm sure Izzy will explain, but from my perspective of it, he's concerned that changing the regression scale so much will somehow change the validity of previous record holders and the ability for new players to break those records because the scale affected them differently than players under a previous scale regression curve.



+1


RE: Regression Curve - King - 09-12-2018

09-12-2018, 07:07 PMTomen Wrote: I am obviously not Izzy but how I interpret his post is that not everyone will be able to have the same "settings" to achieve the same career goals. But that has never been the case in the SHL if you go back through the history of the SHL. TPE inflation, regression scale changes and Update Scale changes made it that players never had the exact same chances to achieve high career numbers.

You could say that those who created around S17-S25 are in a sweet spot for the highest career numbers because of TPE inflation, they still had the older & easier update scale which allowed for higher attributes and therefore a better in sim performance. S21-S25 thre were 52 games played in a season. Rule changes also happened which limited double shifting star players so overall ice time got less and therefore production.

09-12-2018, 07:12 PMMike Izzy Wrote: Realism Era (S12-S20; S25-Present)
52-Game Era (S21-S24)
Inflation Era (S9-S11)
Experimental Era (S6-S8)
Dead Puck Era (S3-S5)

these are the 5 eras of the SHL.... if we make a change to the regression table which all but ends a players chance at any historical numbers then the era must be changed to allow that new generation of players to set there own standards under the new rules... (active players caught in between these changes are subject to debate IMO)

But... if we use the longest career (24 seasons) as the template to make these change then theres an inherant fairness to the rules.

09-12-2018, 07:14 PMSleepy Wrote: I'm sure Izzy will explain, but from my perspective of it, he's concerned that changing the regression scale so much will somehow change the validity of previous record holders and the ability for new players to break those records because the scale affected them differently than players under a previous scale regression curve.

Oh okay, yeah I got confused cause I thought that was already address.

if thats whats meant by it, then it's not an issue, based on just what you guys shared. As you said this is already the case. Every era is very different from every other, and there's already no way to compare them.


RE: Regression Curve - Tomen - 09-12-2018

09-12-2018, 07:12 PMMike Izzy Wrote:
09-12-2018, 07:03 PMKing Wrote:
09-12-2018, 07:00 PMMike Izzy Wrote: if you want this to work and have it done fairly then use the current situation as the standard... if 25 seasons is the longest career to date.. then have the 25% begin at season 24

Current regression table as is...
24th season +: 25%
26th season+30

this would keep the current historial stats safe from any asterisk as each player moving forward will have the same chance as the longest current players to ever play.. while getting buried by the same new regression table at the same point of there careers.

thats something that i could def back

Hey man, could you expand on this asterisk argument, I don't think I fully understand it yet.

Realism Era (S12-S20; S25-Present)
52-Game Era (S21-S24)
Inflation Era (S9-S11)
Experimental Era (S6-S8)
Dead Puck Era (S3-S5)

these are the 5 eras of the SHL.... if we make a change to the regression table which all but ends a players chance at any historical numbers then the era must be changed to allow that new generation of players to set there own standards under the new rules... (active players caught in between these changes are subject to debate IMO)

But... if we use the longest career (24 seasons) as the template to make these change then theres an inherant fairness to the rules.

technically if we introduce a new era for every regression/update scale change there would be quite a few more Era's.
Like the last update scale change was in the offseason from S27 to S28 I think. Then another one when double shifting got ruled out.


RE: Regression Curve - King - 09-12-2018

09-12-2018, 07:12 PMraymond3000 Wrote:
09-12-2018, 06:17 PMDrunkenTeddy Wrote:
09-12-2018, 06:09 PMraymond3000 Wrote: Maybe we can simplify regression altogether and just make the regression percentage equal to the number of seasons played?

Wow.. I said this half an hour ago to King in discord. Only issue is it makes Eggy's life a bit harder when he has to write the regression post. I like the idea of it, but if we are really just trying to tackle the outliers who drag on their career, just making regression go to 30% after season 25 would do it.

Slightly changing topics here, but it can't be too difficult to whip up a spreadsheet that automatically generates the regression list. King already has a list with all the SHL players, their draft year, and their TPE. Of course, someone would still have to manually input TPE totals, but that should make Eggy's life easier.

I would be able to do that pretty easily.


RE: Regression Curve - raymond3000 - 09-12-2018

09-12-2018, 07:20 PMKing Wrote:
09-12-2018, 07:12 PMraymond3000 Wrote:
09-12-2018, 06:17 PMDrunkenTeddy Wrote:
09-12-2018, 06:09 PMraymond3000 Wrote: Maybe we can simplify regression altogether and just make the regression percentage equal to the number of seasons played?

Wow.. I said this half an hour ago to King in discord. Only issue is it makes Eggy's life a bit harder when he has to write the regression post. I like the idea of it, but if we are really just trying to tackle the outliers who drag on their career, just making regression go to 30% after season 25 would do it.

Slightly changing topics here, but it can't be too difficult to whip up a spreadsheet that automatically generates the regression list. King already has a list with all the SHL players, their draft year, and their TPE. Of course, someone would still have to manually input TPE totals, but that should make Eggy's life easier.

I would be able to do that pretty easily.

@Eggy216


RE: Regression Curve - Sleepy - 09-12-2018

09-12-2018, 07:20 PMKing Wrote:
09-12-2018, 07:07 PMTomen Wrote: I am obviously not Izzy but how I interpret his post is that not everyone will be able to have the same "settings" to achieve the same career goals. But that has never been the case in the SHL if you go back through the history of the SHL. TPE inflation, regression scale changes and Update Scale changes made it that players never had the exact same chances to achieve high career numbers.

You could say that those who created around S17-S25 are in a sweet spot for the highest career numbers because of TPE inflation, they still had the older & easier update scale which allowed for higher attributes and therefore a better in sim performance. S21-S25 thre were 52 games played in a season. Rule changes also happened which limited double shifting star players so overall ice time got less and therefore production.

09-12-2018, 07:12 PMMike Izzy Wrote: Realism Era (S12-S20; S25-Present)
52-Game Era (S21-S24)
Inflation Era (S9-S11)
Experimental Era (S6-S8)
Dead Puck Era (S3-S5)

these are the 5 eras of the SHL.... if we make a change to the regression table which all but ends a players chance at any historical numbers then the era must be changed to allow that new generation of players to set there own standards under the new rules... (active players caught in between these changes are subject to debate IMO)

But... if we use the longest career (24 seasons) as the template to make these change then theres an inherant fairness to the rules.

09-12-2018, 07:14 PMSleepy Wrote: I'm sure Izzy will explain, but from my perspective of it, he's concerned that changing the regression scale so much will somehow change the validity of previous record holders and the ability for new players to break those records because the scale affected them differently than players under a previous scale regression curve.

Oh okay, yeah I got confused cause I thought that was already address.

if thats whats meant by it, then it's not an issue, based on just what you guys shared. As you said this is already the case. Every era is very different from every other, and there's already no way to compare them.

personally I'm not concerned about it. I think changing the regression scale is needed. It'll obviously effect career stats and the like but I think that should be secondary, if not tertiary, to other issues like league health.


RE: Regression Curve - DeletedAtUserRequest - 09-12-2018

09-12-2018, 07:30 PMSleepy Wrote:
09-12-2018, 07:20 PMKing Wrote:
09-12-2018, 07:07 PMTomen Wrote: I am obviously not Izzy but how I interpret his post is that not everyone will be able to have the same "settings" to achieve the same career goals. But that has never been the case in the SHL if you go back through the history of the SHL. TPE inflation, regression scale changes and Update Scale changes made it that players never had the exact same chances to achieve high career numbers.

You could say that those who created around S17-S25 are in a sweet spot for the highest career numbers because of TPE inflation, they still had the older & easier update scale which allowed for higher attributes and therefore a better in sim performance. S21-S25 thre were 52 games played in a season. Rule changes also happened which limited double shifting star players so overall ice time got less and therefore production.

09-12-2018, 07:12 PMMike Izzy Wrote: Realism Era (S12-S20; S25-Present)
52-Game Era (S21-S24)
Inflation Era (S9-S11)
Experimental Era (S6-S8)
Dead Puck Era (S3-S5)

these are the 5 eras of the SHL.... if we make a change to the regression table which all but ends a players chance at any historical numbers then the era must be changed to allow that new generation of players to set there own standards under the new rules... (active players caught in between these changes are subject to debate IMO)

But... if we use the longest career (24 seasons) as the template to make these change then theres an inherant fairness to the rules.

09-12-2018, 07:14 PMSleepy Wrote: I'm sure Izzy will explain, but from my perspective of it, he's concerned that changing the regression scale so much will somehow change the validity of previous record holders and the ability for new players to break those records because the scale affected them differently than players under a previous scale regression curve.

Oh okay, yeah I got confused cause I thought that was already address.

if thats whats meant by it, then it's not an issue, based on just what you guys shared. As you said this is already the case. Every era is very different from every other, and there's already no way to compare them.

personally I'm not concerned about it. I think changing the regression scale is needed. It'll obviously effect career stats and the like but I think that should be secondary, if not tertiary, to other issues like league health.



well thats debatable too eh. how could the current system be unhealthy when the majority of players start thinking about recreating when they hit regression, most have retired before there 17th season and only a select few ever reach season 20. The nature of the SIM career has members grow bored/tired of there current player way before any regression table forces them out. The data suggests that we have a healthy system in place.