Rerate Question...

Florida Panthers · Senior Member

Rerate Question...

BryceBruins · BRHL2 Co-Commish

DU is the only statistical rating based on more than the last season, as it is based on the last 2 NHL seasons.

For the other stats it is based purely on the last NHL season at a per game clip. So if Super Joe has a 45 point year he will be statistically rated as a 45 point player. The reasoning? We are based on the last NHL season, if he was only a 45 point player for the Avalanche. He would be rated as a 45 point player because thats how he performed. Although historically his SK and SP woudl be higher which tends to help offensive stats, although there is no qualitative analysis on how.

With players missing parts of the season, the only attribute that will be crippled would be there DU as it is based on a per game basis (although players under 10 GP in the NHL would be very likely to see a few manual tweaks to their adjusted ratings). Goalies it does affect more, Cloutier would be an example based on our current ratings. Although goalie ratings are something we are always tweaking, hopefully to eliminate the peaks and valleys for those injury prone goalies.

Hopefully that answers your questions, if not lemme know, although my response could be dealyed cause I'm at work today

Florida Panthers · Senior Member

BryceBruins wrote:

With players missing parts of the season, the only attribute that will be crippled would be there DU as it is based on a per game basis (although players under 10 GP in the NHL would be very likely to see a few manual tweaks to their adjusted ratings). Goalies it does affect more, Cloutier would be an example based on our current ratings. Although goalie ratings are something we are always tweaking, hopefully to eliminate the peaks and valleys for those injury prone goalies.

Hopefully that answers your questions, if not lemme know, although my response could be dealyed cause I'm at work today

I personally think that playing a limited number of games should have some sort of impact a little more than that... sure, a forward getting 20 points in 20 games of the season is great, but that's a long way from being able to do it over an 80 game schedule. Hell, even if someone does 40 in 40, yeah, good job, but I don't think it should have the same impact as someone getting 80 in 80. Performing at the level for the full season takes more than a partial season...

CBJackets · Guru

Florida Panthers wrote:

BryceBruins wrote:

With players missing parts of the season, the only attribute that will be crippled would be there DU as it is based on a per game basis (although players under 10 GP in the NHL would be very likely to see a few manual tweaks to their adjusted ratings). Goalies it does affect more, Cloutier would be an example based on our current ratings. Although goalie ratings are something we are always tweaking, hopefully to eliminate the peaks and valleys for those injury prone goalies.

Hopefully that answers your questions, if not lemme know, although my response could be dealyed cause I'm at work today

I personally think that playing a limited number of games should have some sort of impact a little more than that... sure, a forward getting 20 points in 20 games of the season is great, but that's a long way from being able to do it over an 80 game schedule. Hell, even if someone does 40 in 40, yeah, good job, but I don't think it should have the same impact as someone getting 80 in 80. Performing at the level for the full season takes more than a partial season...

Sure but there has to be some kind of games benchmark. I agree that 10 games is kind of low, maybe make it 20, if a player plays any less than that they will see a lowering of their ratings.

However what you're arguing for is usually a question of durability, How can you really say if a player with 30 points in 30 games would end up with 82. You could go the other way and say that if a player like Thornton gets injured after a slow start that he could have put up more points and got a little shafted. I think 20 games is a decent benchmark, while 10 is definitely too low. Around a quarter mark of the season point leaders tend to even themselves out.

Historically a guy who puts up 40 points in 40 games is far more likely to get around a ppg than to taper off in any significant way.

-- Edited by CBJackets at 02:21, 2006-11-02

Florida Panthers · Senior Member

CBJackets wrote:

Historically a guy who puts up 40 points in 40 games is far more likely to get around a ppg than to taper off in any significant way.
-- Edited by CBJackets at 02:21, 2006-11-02

I just look at it and compare to the absolutely retarded contract that Briere got this year in arbitration (yes I realize we're talking rerates not dollars, but it's similar in principle). He got that $5 million contract for 58 points in 48 games. Prior to that his best year was 65 points and he hadn't shown an increase and progression in points year after year.

Make no mistake, he's certainly showing this year that he can continue to put up that 58 point in 48 games pace, but my point is that at the time that contract was awarded, it was absolutely retarded.

Clearly 40 points in 40 games is more likely an indicator than 10 points in 10 games.

Philippe27 · Guru

Florida Panthers wrote:

CBJackets wrote:

Historically a guy who puts up 40 points in 40 games is far more likely to get around a ppg than to taper off in any significant way.
-- Edited by CBJackets at 02:21, 2006-11-02

I just look at it and compare to the absolutely retarded contract that Briere got this year in arbitration (yes I realize we're talking rerates not dollars, but it's similar in principle). He got that $5 million contract for 58 points in 48 games. Prior to that his best year was 65 points and he hadn't shown an increase and progression in points year after year.

Make no mistake, he's certainly showing this year that he can continue to put up that 58 point in 48 games pace, but my point is that at the time that contract was awarded, it was absolutely retarded.

Clearly 40 points in 40 games is more likely an indicator than 10 points in 10 games.

Personally in my ratings I use goals per game for scoring but there is still a small percentage awarded to total goals. So for example Elias who has 16 goals in 38 games would have been projected to get 35 but he got 81 scoring instead of 83 like Satan who actually had 35 goals in 82 games.

OilersBrent · Senior Member

Pro rating scoring and passing ratings is very hard to do.

Relating to your example of Elias and Satan: I think it's clear that in today's day Elias is the better scorer of the two. I don't think Elias's scoring rating should take a hit because he missed a chunk of the season due to injury. If he is healthy and able to play he will score at a greater rate than Satan. His DU rating should be the one to take the hit because he can't play a full season, and therefore in the sim he will miss the games but score when he is healthy. If both DU and SC/PA were both lowered due to missed games then injury plagued star players won't be realistic in this league.

As for the issue of difference between 10 points in 10 games and 40 points in 40 games: It is hard to rate when it comes to this. Who says the guy who has 10 points wasn't going to continue at that rate, better or just slightly worse? Being the re rater you can't use your own opinions of the player because everyone has a different view of a player and their skill. Again it comes to the DU rating. Having a guy with a low DU is a big disadvantage in sim leagues. Everyone knows that Forsberg is arguably the best player in the NHL when healthy but he is rarely healthy. When he plays he produces big time and that is the way it should be in the sim.

-- Edited by OilersBrent at 16:52, 2006-11-02

Philippe27 · Guru

OilersBrent wrote:

Pro rating scoring and passing ratings is very hard to do.

Relating to your example of Elias and Satan: I think it's clear that in today's day Elias is the better scorer of the two. I don't think Elias's scoring rating should take a hit because he missed a chunk of the season due to injury. If he is healthy and able to play he will score at a greater rate than Satan. His DU rating should be the one to take the hit because he can't play a full season, and therefore in the sim he will miss the games but score when he is healthy. If both DU and SC/PA were both lowered due to missed games then injury plagued star players won't be realistic in this league.

As for the issue of difference between 10 points in 10 games and 40 points in 40 games: It is hard to rate when it comes to this. Who says the guy who has 10 points wasn't going to continue at that rate, better or just slightly worse? Being the re rater you can't use your own opinions of the player because everyone has a different view of a player and their skill. Again it comes to the DU rating. Having a guy with a low DU is a big disadvantage in sim leagues. Everyone knows that Forsberg is arguably the best player in the NHL when healthy but he is rarely healthy. When he plays he produces big time and that is the way it should be in the sim.

-- Edited by OilersBrent at 16:52, 2006-11-02

DU is very random in the sim and over 1 season will have very little impact so you can't rely on that to give an edge to players who played more games.

OilersBrent · Senior Member

I agree DU is random, but so are injuries in the NHL. Elias and Forsberg aren't guarunteed to get injured every year. If they don't, they will rack up the points, just as they should in the BRHL.

CBJackets · Guru

After your certain standard (say 20 games) DU is the only way to pusnish a player who has played fewer games. Otherwise it gets way too subjective, how are you to say that that player wouldn't have put up more points in more games played.

A lot of random/unexplained stuff happens, a guy who sucks may put up 30 points in 30 games and then get injured (rare but it could happen) but that's kind of the same as a lesser player playing on an awesome line and racking up the points. It's just probably better to try and avoid subjectivity.

Philippe27 · Guru

CBJackets wrote:

After your certain standard (say 20 games) DU is the only way to pusnish a player who has played fewer games. Otherwise it gets way too subjective, how are you to say that that player wouldn't have put up more points in more games played.

A lot of random/unexplained stuff happens, a guy who sucks may put up 30 points in 30 games and then get injured (rare but it could happen) but that's kind of the same as a lesser player playing on an awesome line and racking up the points. It's just probably better to try and avoid subjectivity.

Agreed, which is why the penalty in my ratings is very minor (only 1 or 2 pts) but it's still there because 30 pts in 30 games does not equal 80 pts in 80 games. We can't get in the subjective but we can give a very minor penalty to players that played fewer games.

CBJackets · Guru

Philippe27 wrote:

CBJackets wrote:

After your certain standard (say 20 games) DU is the only way to pusnish a player who has played fewer games. Otherwise it gets way too subjective, how are you to say that that player wouldn't have put up more points in more games played.

A lot of random/unexplained stuff happens, a guy who sucks may put up 30 points in 30 games and then get injured (rare but it could happen) but that's kind of the same as a lesser player playing on an awesome line and racking up the points. It's just probably better to try and avoid subjectivity.

Agreed, which is why the penalty in my ratings is very minor (only 1 or 2 pts) but it's still there because 30 pts in 30 games does not equal 80 pts in 80 games. We can't get in the subjective but we can give a very minor penalty to players that played fewer games.

That's actually a nice way to do it. And Maybe a guy who has 50 in 50 sees his SC decrease by 1.

Florida Panthers · Senior Member

CBJackets wrote:

That's actually a nice way to do it. And Maybe a guy who has 50 in 50 sees his SC decrease by 1.

It's a nice simple way, but how about something at least with a little more bite... say 1 point off for every 10 games missed. It may sound like a lot, but I really don't think it is, in fact I'd say it's pretty generous.

Someone who would normally get an 85 in scoring gets an 82 if they miss 30 games. An 82 is still pretty good considering you've missed more than a 3rd of the season...

Philippe27 · Guru

Florida Panthers wrote:

CBJackets wrote:

That's actually a nice way to do it. And Maybe a guy who has 50 in 50 sees his SC decrease by 1.

It's a nice simple way, but how about something at least with a little more bite... say 1 point off for every 10 games missed. It may sound like a lot, but I really don't think it is, in fact I'd say it's pretty generous.

Someone who would normally get an 85 in scoring gets an 82 if they miss 30 games. An 82 is still pretty good considering you've missed more than a 3rd of the season...

If you do that for every rating it becomes a lot. I'm not exactly sure how my formulas for other ratings do it but somehow they just do. Anyways that's something to consider for Eric and Bryce.

CBJackets · Guru

Philippe27 wrote:

Florida Panthers wrote:

CBJackets wrote:

That's actually a nice way to do it. And Maybe a guy who has 50 in 50 sees his SC decrease by 1.

It's a nice simple way, but how about something at least with a little more bite... say 1 point off for every 10 games missed. It may sound like a lot, but I really don't think it is, in fact I'd say it's pretty generous.

Someone who would normally get an 85 in scoring gets an 82 if they miss 30 games. An 82 is still pretty good considering you've missed more than a 3rd of the season...

If you do that for every rating it becomes a lot. I'm not exactly sure how my formulas for other ratings do it but somehow they just do. Anyways that's something to consider for Eric and Bryce.

Ya, I really don't support that, DU is supposed to take care of that. It's not perfect but players with low DU do get injured more often than players with higher DU.

A perhaps better suggestion is to adjust the DU in the ratings, right now I don't think there is a big enough gap between high durability and low durability.

If a player misses 30 games and you dock say PC, PA, and SC, well that's going to have a significant impact. Whereas if you play with durability, you can hopefully have a player play 40 or 50 games as they did in the NHL. Nothing is perfect but taking that many points away from the ratings is getting way too subjective.