Proposal to change extension method

[Rob]

I hear you but when we're extending contracts for 5 years at a time I think the previous 3 are completely relevant. Certainly they would be relevant in an actual contract negotiation or arbitration (good, bad, and ugly). What if we're talking about a Joffery Lupul type player that is injured all the time? Base his contract off one high year or build in some accounting for lost games?

But maybe it will take some numbers to compare.

If Stamkos misses a season, then puts up 3.0PPG the next 2 seasons - no GM here or in the NHL is considering that injured year in the contract length. 

[jmtrops]

I did a few teams that are closest to the cap this year. this is their cap for 20/21 with the new points calculation. I applied the prospect discount to those players.
TOR- 2M
NYI- (-3.6M)
PHL- (-13.7M)
PIT- (-1.3M)
STL- (-15M)
WIN- (-9.9M)
COL- (-33.3M)
EDM- 13.5M
VGK- (-2.3M)
iLL DO THE OTHER TEAMS LATER

[SlackJack]

If Stamkos misses a season, then puts up 3.0PPG the next 2 seasons - no GM here or in the NHL is considering that injured year in the contract length.

Exactly. Take the best two years of the three and average them. So 3.0PPG would stand.

[jmtrops]

well I dont feel so bad now. I did the rest except for ARI, BOS, NYR, OTT
BUF- 9.7M
CHI- (-21.7)
DET- 17.2M
MON- 6.8M
NAS- (-4.7M)
VAN- 6.6M
WAS- 21.6M

[SlackJack]

Posting some numbers here for consideration. I've done this by hand but I think the over-all picture is clear: the changes are not huge. I haven't quite figured out how to nicely align tables in BBC but if you pick through you will see that average fantasy points for the best of the last 3 years are in bold along with the extension value at $25k per point.

Putting players like Aho and Marner in this list isn't quite right because their rookie production drags them down and they're not up for extension anyways. They are rightly covered by prospect extension contracts which is a different thing all-together.

Over-all there is a downward trend for these top 25 players which I think will be more than made up in the middle so I don't see a watering down effect. (Even if there was you can always bump the multiplier to $26-$27k per point.) Anyways, my argument is precisely that players should not be extended based solely on their single best year of production and as I said, the changes are not huge.

• Nikita Kucherov   
  328.2   366.3   428.3   397.3   $9.90   $10.70
• Alex Ovechkin   
  337.25   395.3   423.25   409.27   $10.20   $10.60
• Nathan MacKinnon   
  199.15   358   401.35   379.67   $9.50   $10.00
• Patrick Kane  
  333.8   258.55   398.3   366.05   $9.20   $10.00
• Connor McDavid   
  355.65   390.9   384.75   387.82   $9.70   $9.80
• Leon Draisaitl   
  270.25   238.5   378.8   324.5   $8.10   $9.50
• John Tavares  
  263.7   299.15   371.85   335.5   $8.40   $9.30
• Steven Stamkos   
  79.35   305.05   362.8   333.92   $8.30   $9.10
• Sidney Crosby 
  364.35   311.9   360.65   362.5   $9.10   $9.00
• Johnny Gaudreau   
  198.35   276.15   349.85   313   $7.80   $8.70
• Andrei Vasilevskiy   
  249.25   397   348.25   372.62   $9.30   $9.90
• Brad Marchand   
  327.15   311.45   346.65   336.9   $8.20   $8.70
• Brayden Point  
  158.85   272.35   345.7   309.02   $7.70   $8.60
• Frederik Andersen   
  359.75   390.25   345   375   $9.40   $9.80
• Tyler Seguin   
  269.05   345.95   343.2   344.58   $8.60   $8.60
• Carey Price   
  336   194.75   340.75   338.37   $8.50   $8.50
• Connor Hellebuyck   
  251.5   406   338   372   $9.30   $10.20
• Jake Guentzel  
  139.5   205.05   325.4   265.22   $6.60   $8.10
• Sebastian Aho 
  208.65   255.85   325.4   290.6   $7.20   $8.10
• Mitchell Marner 
  210.45   234.65   325.3   279.97   $7.00   $8.10
• Gabriel Landeskog   
  148.8   269.3   320.65   295.2   $7.40   $8.00
• Aleksander Barkov   
  203.25   289.1   316.7   302.9   $7.60   $7.90
• Sergei Bobrovsky   
  383.75   362.75   315.75   373.25   $9.30   $9.10
• David Pastrnak   
  299.1   311.55   313.8   312.65   $7.80   $7.80

Cherry picking some cases now to really show where an average helps account for injury or different on-ice usage:

• Andrej Sekera   
  143.95   16.15   21.65   82.8   $2.10   $0.50
• Jason Zucker   
  230.25   273.1   189.45   251.65   $6.30   $6.80
• Kevin Shattenkirk   
  188.55   72.10   98.25   143.4   $3.60   $2.5
• Kris Letang   
  124.8   188.15   232.7   210.42   $5.30   $5.80
• Oscar Klefbom   
  167.4   101.2   106.15   136.78   $3.40   $2.70
• Ryan Nugent-Hopkins   
  170.85   200.75   247.65   224.2   $5.60   $6.20
• Sven Baertschi   
  105.1   140.7   51.6   122.9   $3.10   $3.50
• Tomas Hertl   
  98.6   194.65   278.45   236.55   $5.90   $7.00
• Zach Parise   
  176   112.85   242.1   209.05   $5.2   $6.1

[Rob]

Posting some numbers here for consideration. I've done this by hand but I think the over-all picture is clear: the changes are not huge. I haven't quite figured out how to nicely align tables in BBC but if you pick through you will see that average fantasy points for the best of the last 3 years are in bold along with the extension value at $25k per point.

Putting players like Aho and Marner in this list isn't quite right because their rookie production drags them down and they're not up for extension anyways. They are rightly covered by prospect extension contracts which is a different thing all-together.

Over-all there is a downward trend for these top 25 players which I think will be more than made up in the middle so I don't see a watering down effect. (Even if there was you can always bump the multiplier to $26-$27k per point.) Anyways, my argument is precisely that players should not be extended based solely on their single best year of production and as I said, the changes are not huge.

• Nikita Kucherov   
  328.2   366.3   428.3   397.3   $9.90   $10.70
• Alex Ovechkin   
  337.25   395.3   423.25   409.27   $10.20   $10.60
• Nathan MacKinnon   
  199.15   358   401.35   379.67   $9.50   $10.00
• Patrick Kane  
  333.8   258.55   398.3   366.05   $9.20   $10.00
• Connor McDavid   
  355.65   390.9   384.75   387.82   $9.70   $9.80
• Leon Draisaitl   
  270.25   238.5   378.8   324.5   $8.10   $9.50
• John Tavares  
  263.7   299.15   371.85   335.5   $8.40   $9.30
• Steven Stamkos   
  79.35   305.05   362.8   333.92   $8.30   $9.10
• Sidney Crosby 
  364.35   311.9   360.65   362.5   $9.10   $9.00
• Johnny Gaudreau   
  198.35   276.15   349.85   313   $7.80   $8.70
• Andrei Vasilevskiy   
  249.25   397   348.25   372.62   $9.30   $9.90
• Brad Marchand   
  327.15   311.45   346.65   336.9   $8.20   $8.70
• Brayden Point  
  158.85   272.35   345.7   309.02   $7.70   $8.60
• Frederik Andersen   
  359.75   390.25   345   375   $9.40   $9.80
• Tyler Seguin   
  269.05   345.95   343.2   344.58   $8.60   $8.60
• Carey Price   
  336   194.75   340.75   338.37   $8.50   $8.50
• Connor Hellebuyck   
  251.5   406   338   372   $9.30   $10.20
• Jake Guentzel  
  139.5   205.05   325.4   265.22   $6.60   $8.10
• Sebastian Aho 
  208.65   255.85   325.4   290.6   $7.20   $8.10
• Mitchell Marner 
  210.45   234.65   325.3   279.97   $7.00   $8.10
• Gabriel Landeskog   
  148.8   269.3   320.65   295.2   $7.40   $8.00
• Aleksander Barkov   
  203.25   289.1   316.7   302.9   $7.60   $7.90
• Sergei Bobrovsky   
  383.75   362.75   315.75   373.25   $9.30   $9.10
• David Pastrnak   
  299.1   311.55   313.8   312.65   $7.80   $7.80

Cherry picking some cases now to really show where an average helps account for injury or different on-ice usage:

• Andrej Sekera   
  143.95   16.15   21.65   82.8   $2.10   $0.50
• Jason Zucker   
  230.25   273.1   189.45   251.65   $6.30   $6.80
• Kevin Shattenkirk   
  188.55   72.10   98.25   143.4   $3.60   $2.5
• Kris Letang   
  124.8   188.15   232.7   210.42   $5.30   $5.80
• Oscar Klefbom   
  167.4   101.2   106.15   136.78   $3.40   $2.70
• Ryan Nugent-Hopkins   
  170.85   200.75   247.65   224.2   $5.60   $6.20
• Sven Baertschi   
  105.1   140.7   51.6   122.9   $3.10   $3.50
• Tomas Hertl   
  98.6   194.65   278.45   236.55   $5.90   $7.00
• Zach Parise   
  176   112.85   242.1   209.05   $5.2   $6.1

The top 25 see a fairly steep deflation, but my fear with your model is what the bottom looks like.  Your model would truncate the price range, but add more prices to the range.  In the new model I described, of the 675 contract prices that I compared with the old extension model, there's something like 10 contracts at the bottom of the range at $0.9m - I have a feeling that range would end at a higher number in your model - since more players wouldn't price out of the list with that 3rd season keeping them in the range.  It seems like a lot of work to quantify that, though - having to average out 3 years.  It's beyond my expertise to make a spreadsheet that automates it, especially when there's 3 data points for some, 2 for some and 1 for others. 

[Rob]

Sekera is the perfect example.  He's player number 676 - how many more like him to do add to the range, forcing the bottom of the range to go up and up and up?  There'd have to be fewer than 10 to keep the bottom of the range at $0.9m.

[SlackJack]

You can always bump the multiplier up to $26k (or whatever) if you need to shift up or down but I can look at the numbers at the bottom. Sekera at $2.1m or $500k is cheap either way.

[Rob]

You can always bump the multiplier up to $26k (or whatever) if you need to shift up or down but I can look at the numbers at the bottom. Sekera at $2.1m or $500k is cheap either way.

If the theoretical range goes from $10m-$1m in the "best of 2 year" method and the ranges goes from, let's say, $9m-$2m in the "average of 3" method you'd have to add a 2nd tier where the multiplier is different.  Otherwise the range just shifts downwards and the top gets less expensive.  So a tier 1 player (however we define that) could be $26k while a tier 2 player could be $24k.  We could tinker to find the right combination to keep the price range relative to the current settings. 

Getting the right figure is the tough part - cause sampling some low ranking players isn't enough.  We'd need to rank them all to see how fat the range would be and how to adjust it correctly. 

[SlackJack]

If the theoretical range goes from $10m-$1m in the "best of 2 year" method and the ranges goes from, let's say, $9m-$2m in the "average of 3" method you'd have to add a 2nd tier where the multiplier is different.  Otherwise the range just shifts downwards and the top gets less expensive.  So a tier 1 player (however we define that) could be $26k while a tier 2 player could be $24k.  We could tinker to find the right combination to keep the price range relative to the current settings. 

Getting the right figure is the tough part - cause sampling some low ranking players isn't enough.  We'd need to rank them all to see how fat the range would be and how to adjust it correctly.

More complicated than I was aiming for but you obviously have lots of tools to work with. I think over-all the scheme is brilliant in how it corrects or homogenizes the current quirks or imbalances. Mind you I like the current flavour of DNHL. It's a bit weird and I love it that way.

With these vast structural changes it will be fun to see just how manic free agency is in a few years. I can see huge pressure on roster construction when everyone is actually getting paid what they are worth. Especially when pinned to only one good year.

But that works for me just fine