Tuesday, August 29, 2006

False Normalization of 1920's Assists

During the off season I have looked at some hockey sabermetrics questions. One problem is adjusting scoring from different eras to produce a top 10 goal scoring seasons and a top 10 assist scoring seasons list. One method to do this comes from the hockey outsider (Peter Albert) and I am exploring his results.

The adjusted goal scoring list gave pretty good results, with Brett Hull leading the way. It tended to have too many players from years where the quality of play is lower (like expansion years) and it had none from the original six era. Because of this, I suggest it needs a quality of opposition adjustment but I am uncertain that existing statistics allow a reliable method to do this.

The adjusted assist list is different. It has Wayne Gretzky on it four times and the other six entries are all players from the 1920's. In the 1920's assists were different from today. Forward passing was slowly being allowed into the game. In 1929/30 it was first allowed within the offensive zone (but not across the blue line into the offensive zone as that would be offside). Thus there was less passing in the attacking zone. Any pass had to be a pass backwards. The official scorers were very stingy with assists. They tended to give out a maximum of one assist per goal - and only if the pass clearly led to the goal. As a result there were significantly less assists. Players who got a good number of assists tended to get them from drop passes and cross ice passes that were slightly behind them. The nature of their assists was different from that of today. The number of players who got these assists was less than it is today. Some players who were goal scorers tended to shoot instead of pass in these situations and rarely got assists. As a result, a player could have an assist total in the teens and lead the league.

Is that achievement really the same as a 100 plus assist season in more modern times? With this adjusted assist system it is viewed as such. Is it even logical to believe that the best assistmen of all time are a bunch of guys who were not allowed to use forward passes in the offensive zone and then nobody else until Wayne Gretzky came along sixty years later? No. This is a false normalization.

A false normalization occurs if player's totals are normalized against their era, when they clearly have very different totals from very different eras and it appears after normalization that in one era, an achievement is worth a lot more than the same achievement in another era and this difference is unreasonable.

As an example we can look at the best two seasons of all time in the adjusted assist list. Number one is Wayne Gretzky who in 1985/65 scored 163 assists for the Edmonton Oilers. Number two is Cy Denneny who in 1924/25 scored 15 asists for the Ottawa Senators. Denneny's assist total is normalized upwards two times by significant amounts. It is normalized upwards once because he only played 28 games (compared to Gretzky's 80) and a second time because assist frequency was so much lower in his era. After normalization, Gretzky has 129 normalized assists and Denneny has 128. Each real assist by Gretzky translates to 0.79 normalized assists and each real assist by Denneny translates to 8.5 normalized assists. Therefore each Denneny assist is worth 10.8 Gretzky assists. Is that reasonable? Could one assist really be worth more than ten times as much in one era then it is in another era? Afterall each assist ony leads to one goal regardless of era. Could that goal be ten times more valuable? Obviously not.

The problem is one of normalization. It is the wrong way to approach this problem. Unfortunately, it is the only approach where a framework exists to try to approach it. I suggest approaching it with a different framework - unfortunately one that is probably not entirely possible given the nature of hockey stats. Despite the fact that the method is not entirely possible, it is instructive to look at the way it would be approached.

Assist scoring produces goals. One method to approximate how many goals it produces comes from goals created formalism. This is merely an approximation and with better statistics about how the goal was created, we could probably break it down better. Anyway, we need to know how many goals were created by the assists of a player in question. Once we have that number, we can translate it into wins. Since wins correlate with goals scored (neglecting complications like shootouts and points for losing in overtime, a team's winning percentage is very well approximated by their goals scored squared divided by goals scored plus goals allowed squared - this is often referred to as the Pythagorean Formula - not to be confused with right angle triangles), we can turn the number of goals created into a number of wins created by those goals. Then we can compare the wins created into different eras. The wins created need to be normalized once by the number of games played to do this. This will reduce the false normalization where one assist is worth ten times as much as an assist in another era. There is no way one assist can be worth that much more even taking into account schedule length. Now after this is all done, we must correct one more time for the quality of opposition. It is easier to get assists or win games in a lower quality league. There is likely no clear way to undertake this step as it requires a way to quantify precisely the quality of opposition in different eras. Following this method there is no way six of the ten best assist seasons would come from the 1920's before forward passing was fully allowed.

In the 1920's few players got assists. A select handful was capable of getting them in double digit numbers. That select handful gets falsely normalized to be equivalent to 120 to 160 assist seasons of Wayne Gretzky. There is no way that 15 assists is the same as 160 assists. The 160 assists lead to far more wins.

Another example of this sabermetric problem can be found in baseball. Prior to Babe Ruth (the deadball era), there were very few homeruns. Sometimes people produce some normalization to say that a 10 home run season was as good or better than a 60 home run season. Its not. It doesn't produce nearly as many wins for a team.

Given the current framework of hockey sabermetric calculations it is necessary to normalize to compare different seasons. This leads to problems particularly with assists. Players from the 1920's before the full implementation of forward passing who were capable of getting a few assists (maybe a double digit number) normalize to be as good as somebody who gets 100 plus assists in later years. This doesn't make sense. It doesn't make sense that several players in the pre-forward passing era could get have better assist seasons than Bobby Orr or Mario Lemieux or Adam Oates.

Sunday, August 27, 2006

Adjusted Assist Scoring: Single Season

In sabermetrics and hockey one problem is to adjust scoring between different eras. The hockey outsider has a method to do this that I have already used to look at single season adjusted goals and am mostly satisfied with the results.

Adjusting assists is slightly different than adjusting goals because the frequency of asists being given out by scorers has changed significantly over time. In the earoy days of hockey, there were more goals than assists. Most goals were unassisted. Today, most goals have two assists. This gives one more factor that needs adjustment.

Here are the top 10 adjusted assist seasons according to the hockey outsider (Peter Albert):

Top 10 Adjusted Assist Scoring Seasons of All Time
Name Team Year Games Played Adjusted Assists Actual Assists
Wayne GretzkyEdm1985/8680129163
Cy DennenyOtt1924/252812815
Duke KeatsEdm- WCHL1921/222511925
Bill BoucherMon1924/253011013
Wayne GretzkyLA1990/9178109122
Wayne GretzkyEdm1984/8580107135
Dick IrvinChi1926/274410618
Wayne GretzkyEdm1986/8779103121
Howie MorenzMon1927/284310318
Frank BoucherNYR1928/294410316

This list is not nearly as satisfactory as the adjusted goals list. It contains Wayne Gretzky four times. Since he has far more assists than anyone else in history this is not suprising. The problem is the other six guys on the list are all from the 1920's. Until the 1929/30 season, there was a maximum of one assist per goal and the scorer would only give out an asists if he though it clearly led to the goal. This can clearly be seen looking at the all time assist leaders by season. Frank Boucher lead the NHL in assists in both 1929 and in 1930 (after the change was made to today's system). In 1929 he has 16 assists (good enough for 10th all time on the hockey outsider list). In 1930 he had 36 assists but now placed 23rd on the hockey outsider list. That is an amazing piece of normalization. In back to back years a player gets 20 MORE assists but is normalized to have more normalized assists in the year when he had 20 LESS assists.

These 1920's players at the top of the list are due to normalization effects. They benefit because their games played was low and they benefit because there were few assists. Each real assist Cy Denneny got in 1925 was worth more than 8.5 normalized assists. Each real assist that Wayne Gretzky got in 1986 was worth less that 0.8 of a normalized assist. Was a Cy Denneny assist really worth about ten Wayne Gretzky assists to his team? Obviously it wasn't. This is a false normalization problem. I will write in the future about this and why it dominates this normalized assist list.

Friday, August 25, 2006

Progressive Sports Challenge Bites The Dust

There is one less alternative in the fantasy sports world. Progressive Sports Challenge has stopped running. They ran fantasy baseball and football leagues for six seasons. Despite a court ruling saying leagues dont own their stats and cannot demand a fantasy sport liscence, they received threatening letters from the NFL demanding they get liscenced. This is following baseball's lead. The people at Progressive Sports Challenge are in the right as far as I can tell but have neither the money nor patience to fight the NFL and MLB so they are closing up shop. It's a sad day when any fantasy sports alternative is killed by greed of the major sports leagues. I hope this doesn't spread to the NHL.

Thursday, August 24, 2006

Why We Need A Quality of Opposition Adjustment To Rank All Time Goal Scoring Seasons

One sabermetrics and hockey problem is rating the best goal scoring seasons of all time. One very good attempt at solving this problem was done by the hockey outsider (Peter Albert). He shows Brett Hull's 1990/91 86 goal season to be the best goal scoring year ever. This is a very plausible result.

When one looks further at the top 10 best goal scoring seasons of all time, it becomes clear that there are systematic problems. Nine of the ten best goal scoring seasons are shown to have occurred since 1970. None occurred in the original six years. The other year was Babe Dye's 38 goal year back in 1924/25. Seven of the ten best years on the list were within three years of an expansion. Its clear that people tended to score more adjusted goals in expansion weakened years. During the original six years, there were no expansions, so they didn't make this list.

There are three specific years that I was expecting to see on or near the top ten from the original six era. I was expecting that Maurice Richard's 50 goals in 50 games season in 1945, Bernie Geoffrion's 50 goal season tying his record in 1961 and Bobby Hull's 1966 54 goal season breaking that 50 goal mark for the first time might feature prominantly on the list. Surprisingly, none of these three years are near the top in Peter Albert's top 50 list (only Hull made it in 40th spot). There are a couple other original six seasons represented on the list. Gordie Howe's 49 goal year in 1953, Jean Beliveau's 47 goal year in 1956 and Bobby Hull's find spots on the list in 12th and 30th position respectively. How could an entire era be so underrepresented? Was there really no great goal scorer in that era or does the analysis systematically overlook them? I think the situation is that the era is overlooked. With only six teams in the NHL, there were fewer bottomfeeding teams to play against where a goal scorer could pad his totals. In more recent times, particularly during expansions, some bottomfeeder teams existed. This is not to say that the average team or average player is any worse in either era, it is a reflection of the fact that more bad players and teams have existed in the larger league.

How would one attempt to adjust for the quality of opposition? Specifically for the presence or absence of bad players and teams that one could pad their statistics against. Its not an easy question. In baseball (where sabermetrics is a much more exact science) there are a few indicators of a league quality that naturally come out of the statistics. For example, one can use the ratio of double plays to errors as a measure. In a good league (ie the majors) there are more double plays then errors, but in a beer league errors are common and double plays are rare. There isn't a clear example of a statistic that has been accurately kept since the beginning of pro hockey that can capture the quality of the league.

The best I came up with after thinking about this problem for a while is the standard deviation of the player's ages from a defined mean age for a pro hockey player. In weak years of the NHL more really young player make the league and more aging players are able to hang on longer to continue their careers after their value is depleted. In strong years, there are few young players and the aging players tend to be forced into retirement instead of having the opportunity to stick around for another year or two. This is shown in the example of Gordie Howe. He was 18 when he first made the NHL in 1946. This was a league that was beginning to grow in strength after the Second World War had depleted its ranks for a few years. Gordie played until age 43 when he retired from the Detroit Red Wings in 1971. Three years later, he was lured out of retirement to play in the weaker WHA. He continued to have successful years in the WHA until it folded in 1979. He played one more year in the NHL when it expanded to include the surviving WHA teams and retired in 1980 at age 52. These final years would likely not have happened if the NHL (and WHA) were not weakened due to the rapid expansion of the 1970's.

I think this technique of using a standard deviation in age as a proxy for league quality would fail in the early days of pro hockey. In the early days, pro players often played around 20 games and their incomes from hockey were not enough to sustain them the whole season and players had a second job that they also worked. It was not uncommon for a still talented player to retire because of pressures from the other job. Moving to a new city for hockey would interfere with their career. Maybe they were at a point in their career where taking the time off for a 20 game season was too much. So many players retired to pursue their second often more lucrative career. This keeps the standard deviation of ages low and also keeps the quality of the league low.

Although I think using the standard deviation of player's ages as a proxy for league quality is probably the best solution to try to correct data for quality of opposition, I don't think it would fully solve the problem. I think it is a lot of work for little gain. I think the problem would still exist after this attempted correction.

The list of the best goal scoring seasons that the hockey outsider produced is a good one. It is good work. Its most glaring problem is the lack of a correction for quality of opposition. This correction is not an easy one to make. It may not be possible to make it in an unbiased manner from existing statistics. I would love some smart person to prove me wrong, but I am not sure its likely. This method produced a list of players who tend to be from expansion seasons and overlooks those players from the original six era. Nevertheless, it is a pretty good attempt at solving a complex problem.

Saturday, August 19, 2006

Is Brett Hull's 1990/91 86 Goal Season The Best Goal Scoring Year Ever?

Whenever we do sabermetrics it is important to ask ourselves if the results we get make sense. It is important to check to make sure that we are not getting unintended results because of circumstances we did not properly take into consideration.

Recently, I have written about the adjusting scoring from different eras to do player comparisons using a method devised by the hockey outsider (Peter Albert). The top 10 single season adjusted goal scoring list produced by this method lists Brett Hull's 1990/91 86 goal season with the St Louis Blues as the best goal scoring season ever. Is this result accurate?

Brett Hull played 78 games that season. He scored 86 goals, 45 assists for 131 points and won the Hart Trophy as NHL MVP. His lead in the goal scoring department was dramatic. In second place was a three way tie between Theo Fleury of Calgary, Cam Neely of Boston and Steve Yzerman of Detroit who each scored 51 goals. Hull scored 68.6% more goals than his closest opponent. Hull finished second in scoring in the NHL that season, finishing behind Wayne Gretzky of the Los Angeles Kings, who had 163 points (due to a staggering 122 assists). Brett Hull was teamed with Adam Oates who was his setup man. Oates finished second in assists behind Gretzky. In Hull's career which ended last season, he never again managed as many goals in a season, though it may be partly because he never had a full season with Adam Oates as his centreman. Oates was traded to Boston midway through the next season.

Brett Hull came within 6 goals of breaking Wayne Gretzky's single season record of 92 goals and he did in in a time when scoring had dropped from when Gretzky set the record. Nobody has scored as many goals in a season as Hull did that season in any year since. This is partly because scoring has dropped leaguewide since that time. Most shockingly, Hull did not have this season in an expansion weakened NHL - since quality of opposition is not taken into account in this statistics adjustment players in weaker NHL seasons tend to dominate. There had been no NHL expansion since 1979 and the next expansion (the San Jose Sharks) happened the following season. Hull leads the top ten adjusted scoring goal scoring single season list by a full five goals over Phil Esposito's 1970/71 season (which was an expansion year). This is a 7% increase over Esposito's adjusted goal scoring numbers, which is probably large enough to be significant (when considering the fact that Esposito played in an expansion weakened NHL). I think the pick of Brett Hull's 1990/91 season as the best goal scoring year ever is a good one.

Now does this mean Brett Hull is the best goal scorer ever? No. One year doesn't make a career. Also, goal scoring is not a solo act. Most great goal scoring seasons occurred because there was a very successful linemate who was setting up the goal scorer. In Hull's case Adam Oates filled that role. It is not an easy problem to separate how much of the credit should go to Hull and how much should go to Oates (or other teammates). This is one of the problems in hockey sabermetrics. Player's contributions interact in complicated ways. When players have good "chemistry" both of their statistics are improved. When a player plays equally well without the "chemistry" with a linemate his numbers are not as good. This is true of his linemate as well. How much credit goes to either player so that he would hold the same value with or without his all star linemate is an unsolved problem. Ideally, a player should have the same value to his team regardless of who his linemates are. Otherwise, his "value" is not truly due to his contribution. It is unsatisfactory to compare players with one another when the value of the player is due to his teammates and not the player in question. Can this problem be solved? I don't think hockey collects enough statistics to do so (I would love to be proven wrong). I don't know which other statistics are needed. We need to know more about the circumstances of when a player gives and receives passes from his teammates for example. Right now we can compare (for example) the best goal scoring seasons, but we cannot compare the best seasons. We know when circumstances came together that somebody had a top goal scoring season, but that doesn't show that how much of that season was due to the player who scored the goals and how much came from teammates setting them up. Clearly, to have the best goal scoring season ever, one must be a very good goal scorer, but it is not necessary to be the best goal scorer ever. A top setup man like Adam Oates might make up the difference between Hull's season and (for example) Mario Lemieux's 1988/89 season where he scored 85 goals (75 adjusted goals) but did not have a player as good as Adam Oates to feed him the puck. In fact, Lemieux was also the best assist man in the NHL that season. Teammates Paul Coffey (on defence) and Rob Brown were the forth and sixth place finishers respectively in assists. It is widely believed that Brown's success was due to being Lemieux's linemate and Coffey and Lemieux often appeared on the ice separately (though not so often during power plays). So do we would Lemieux's season have a better goal scoring value when decoupled from teammates? Quite possibly, but we lack a reliable method to do this.

Brett Hull's 1990/91 season was a very good year. It was the best goal scoring season in the history of hockey. I think the sabermetrics are accurate. How much of this credit should go to Hull and how much to teammates (specifically Adam Oates) is unclear. It is entirely possible that Hull's season while it was the best (adjusted) goal scoring season ever, it was not the best goal scoring season that can be credited to one player ever. Decoupling the contribution of teammates to a combined success is an unsolved sabermetric problem.

Friday, August 18, 2006

My Top 50 List

Since I critiqued the Hockey News Top 50 player list yesterday I thought I would make an attempt at making a list of my own. In order to begin, I decided I needed a better definition. Top 50 players - top 50 how? My definition is these are the 50 players that I would most want to build a winning hockey team right now (that may not be exactly what Mike Brophy had in mind, but it is my best attempt to define the question). I will only leave a comment with the top 10 players and any players on my list who Brophy left off his list. I think it would be redundant otherwise. I dont have too many new things to say about player number 30 that I didn't say yesterday.

Here is my list:

1. Joe Thornton NHL MVP and at age 27 he might still be getting better. Why wasn't he ranked higher by Brophy?

2. Nicklas Lidstrom Consistently the best defender in hockey. Sure he is 36, but he is not showing any signs of decline. I still don't know why he didn't get nominated for the Hart Trophy.

3. Jaromir Jagr Had a great year and could have won Hart as well. Had shoulder surgery in the off season, which keeps him from the very top of this list.

4. Alexander Ovechkin Third highest scorer in the NHL as a rookie. This guy will be incredible. How can you score more than 50 goals when Dainius Zubrus is the best guy to pass you the puck?

5. Scott Niedermayer A great defenceman in his prime. Lidstrom kept him from repeating as Norris winner.

6. Roberto Luongo The best goaltender in the game. When he steps out of Florida and into Vancouver he should post some eye catching numbers He faced more shots than anyone else in the last two NHL seasons and still posted very good numbers. Give him a defence and he will take a run at Vezina.

7. Jarome Iginla A dominant scorer who should be among the top scorers in the NHL now that Tanguay helps him carry the load.

8. Ilya Kovalchuk The best goal scorer in the NHL and he's only 23 so he has many great years ahead.

9. Sidney Crosby Another great young talent who is already a great scorer. If he has a weakness its his defence.

10. Chris Pronger Dominant performance in the playoffs should have won him the Conn Smythe, but voters couldn't see past the fact his team didn't win the cup.

11. Markus Naslund
12. Miikka Kiprusoff
13. Peter Forsberg
14. Martin Brodeur
15. Patrik Elias
16. Zdeno Chara
17. Erik Staal
18. Marian Hossa
19. Joe Sakic
20. Dany Heatley
21. Daniel Alfredsson
22. Rick Nash
23. Brad Richards
24. Marian Gaborik
25. Henrik Lundqvist
The forgotten rookie star from last season. He was better than Phaneuf but couldn't get a Calder nomination. One of the better goalies in the game.

26. Todd Bertuzzi An absolute monster. He's been dealing with the mental issues that go with the Steve Moore hit. A new start in Florida and he could bounce right back.

27. Mats Sundin
28. Vincent LeCavalier
29. Wade Redden
30. Pavel Datsyuk
31. Sergei Zubov
32. Jason Spezza
33. Jonathan Cheechoo
34. Alex Tanguay
35. Simon Gagne
36. Mike Modano
Led Dallas in scoring and very solid defensively. Sure he is 36, but he's still very good.

37. Ed Jovanovski Sure he has suffered some injuries in the past 3 seasons, but its nothing recurring. He should give a very good year and much more injury free.

38. Milan Hejduk Last year may not have been his best offensive season, but he is still 30 and was good enough to score 98 points in the low scoring 2002/03 season. He has too good a scoring touch not to find it again.

39. Martin Havlat
40. Marty Turco
41. Keith Tkachuk
Sure he came to camp overweight and he suffered through injuries all season. But he still managed to score nearly a point per game and dominate games physically with little or no supporting cast

42. Pavol Demitra Has put up some very consistent numbers for several years. Better than point per game again last year. Should fit in well in Minnesota.

43. Bryan McCabe
44. Daniel Briere
58 points in only 48 games last year. He's always had a scoring touch and he looks ready to show us with a big full season.

45. Ryan Smyth
46. Rod Brind'Amour
47. Tomas Vokoun
48. Shane Doan
49. Rob Blake
50. Martin St Louis

Missing from my list but appearing on the Mike Brophy list are Brian Gionta, Olli Jokinen, Patrick Marleau, Brendan Morrow, Dion Phaneuf, Robyn Regehr, Lubomir Vishnovsky and Henrik Zetterberg.

Malkin "defects"

With the end of communism in Russia and the fall of the Berlin Wall, I naively believed it would be the end of foreign players who wanted to play in the NHL but had significant legal and illegal hurdles placed in their way. That isn't the case. There is no player transfer agreement between Russia and the NHL. That means that players under contract in one league can jump to the other league without any formal legal recourse.

The Russians are unhappy at the low player transfer fees for talent that they produced. Evgeni Malkin would net a $200,000 transfer fee, which is likely a small amount of what he would be worth to the Russian Elite League if he stayed there. The Russians have been trying to get transfer fees raised to a level that compensates their system for te loss of a player.

The NHL's position has been this is our offer and if you don't take it we will take your players and you won't get any money in return at all. They have attempted some strongarm tactics to get Russia to agree to the deal, but it hasn't worked.

This standoff led to court cases regarding the status of Alexander Semin and Alexander Ovechkin. It is only getting worse. Players want to leave Russia to join the NHL, but it is unclear how to do it.

Alexei Mikhnov and Andrei Taratukhin have give notice that they will break their contracts and come to the NHL (under Russian labor law it takes 14 days notice to break any signed contract). Evgeny Malkin is going one step further. Although Russia is technically a free country, its not quite so simple in practise. The Russian mafia owns large portions of the country and have interest in most successful hockey franchises. If a player tries to leave they will try to use ther influences to change his opinion. In Malkin's case he was forced to sign his most recent contract at 3AM in the morning after some coercion. He was not allowed access to his passport until his Metallurg Magnitogorsk team went to their training camp in Finland. During this training camp, Malkin fleed Russia. He has now surfaced in Los Angeles.

The future holds costly legal battles between the NHL and Russia over player transfer rights. Malkin will likely play in Pittsburgh, all that isn't known is how much Pittsburgh will pay to Russia for his rights. Either both sides will settle on a figure or the courts will decide (with some possibility the figure will be zero and some possibility it will be much higher than we are used to). The Russians want a system similar to soccer where teams negotiate individually transfer costs on a player by player basis. Sometimes one player can be worth many million dollars. This structure would be far more expensive for the NHL and would upset the salary cap system. What is capped is player costs. A $20 million player transfer fee would be almost half of a team's salary cap and they won't have any players signed to play hockey.

This fight will influence the relations between the NHL and Europe in general. For example, higher transfer fees to Russia will lead to higher transfer fees to other countries.

Here is TSN's story on the matter and here is what the Moscow Times has to say.

Thursday, August 17, 2006

Hockey News Top 50 Players

The Hockey News came out with their yearbook which lists the top 50 players in the NHL according to Mike Brophy. This list always brings up some contreversy. Here is their list along with the rating the player had last season and my comments.

1. Miikka Kiprusoff (31st overall in 2005) In order to jump 30 points and land in first overall, one would think Kiprusoff would have won the Hart Trophy by a large margin. That is not so. While Kiprusoff won the Vezina and finished third in the Hart Trophy voting (though I would not have nominated him for the award). I cannot find any logical way to justify Kiprusoff as the best player in the NHL. Would you expect he is the most likely guy to win the 2007 Hart Trophy? I wouldn't.

2. Chris Pronger (up from 4th). While Pronger is clearly a very good player and deserved (but did not win) the Conn Smythe trophy is he the best defender in hockey? If so why was he not a Noris Trophy nominee?

3. Jaromir Jagr (up from 20th). His time in Washington was unimpressive and led to a drop in his ranking, but he jumped back to the standard we'd expect from Jagr this year. I picked him as the man most deserving of the Hart Trophy. This ranking is about right.

4. Joe Thornton (up from 6th). A Hart Trophy didn't give Thornton as big a bump up the list as I would have expected.

5. Alexander Ovechkin (debut). The rookie of the year is poised to have a great career. This ranking is about right.

6. Sidney Crosby (debut). The runner up to the rookie of the year. He is almost two years younger than Ovechkin and probably projects to having a better career, but I am not sure he is quite this high yet. The fact Crosby is a centerman makes his rokie defensive lapses a bigger deal than Ovechkin's on the wing. When that is coupled with the fact Crosby is a poor faceoffman and not a huge imposing force, I wouldn't rate Crosby quite this high yet.

7. Jarome Iginla (down from 2nd). Last year wasn't Iginla's best. Sure he led the Flames in scoring, but he wasn't among the top scorers in the NHL. Some are down on him because he has never had a huge 100 point year (his huge scoring years happened in a lower scoring NHL than today), but I think this rating is justified.

8. Scott Niedermayer (up from 13th). Hard to make sense of the fact he is rated higher in a year he is Norris trophy runner up than last year when he was defending champ.

9. Rick Nash (up from 50th). Huge jump for a guy who only got 54 points last year (granted it was in 54 games). Sure Nash has loads of potential, but he has never shown himself to be as good as the nineth best player in the NHL.

10. Dany Heatley (down from 5th). It was a mistake to rate Heatley as high as 5th last year when he was coming off an awful season recovering from the mental and physical injuries of the car accident that killed Dan Snyder. This is shown by the fact Heatley had a 103 point season and dropped in the rankings.

11. Martin Brodeur (down from 7th). Sure Brodeur is a very good goalie, but he hasn't been the best goalie in the NHL in a while. He starts to look beatable now that the defence in front of him is no longer as good as the Scott Niedermayer/ Scott Stevens Devils.

12. Patrik Elias (up from 26th) His 45 points in only 38 games while recovering from hepatitis should have earned him the Masterton trophy. Not sure why the Devils media didn't even nominate him.

13. Peter Forsberg (down from 1st) If he was healthier I'm sure he would be ranked higher, but Forsberg rarely has injury free years. His off-season ankle surgery is the latest in a long line of injuries. Just imagine what a healthy Forsberg could have accomplished

14. Eric Staal (debut) Sophomore center had a great breakout year and capped it off as the top scorer in the Stanley Cup playoffs.

15. Nicklas Lidstrom (down from 12th) What must Lidstrom do to move up? He was a runaway Norris Trophy winner who deserved a Hart nomination (but didn't get it) and for this he drops in the rankings?

16. Wade Redden (debut) Redden was a Norris favorite before injuries dropped him in the stretch drive. How could he have been omitted from last year's list?

17. Joe Sakic (down from 10th) He may be 37, but he is still going strong. He led Colorado in points last year. At his age, things may start to slow down, but they haven't yet.

18. Ilya Kovalchuk (down from 8th) I think he's the best pure goal scorer in the NHL today. He tied for 3rd in goal scoring last year and he drops?

19. Zdeno Chara (up from 21st) I think he was the best Ottawa defenceman last year, despite Redden being slightly higher rated by THN.

20. Daniel Alfredsson (down from 16th) Tied for fourth in points in the NHL last year. If he is dropping its only because he was slighlty overrated last year.

21. Vincent LeCavalier (up from 27th) Why isn't Brad Richards the highest rated member of the Lightning?

22. Jason Spezza (debut) A great passer (second in assists last season), but a bit too one dimensional for this high a ranking. He is not too strong defensively and benefits from the offensive depth in Ottawa to pad his assist totals.

23. Brad Richards (up from 25th) Should be the highest rated Lightning player.

24. Marian Hossa (down from 14th) He deserves a better ranking than this. I think Hossa suffers because he plays in the hockey backwater of Atlanta now and gets seen fewer times a year by most writers.

25. Roberto Luongo (up from 38th) I have no idea why he was rated a low as he was two years ago when he could have won the Vezina trophy had voters noticed him. 2005/06 was still a good year for Luongo, but he was unhappy with Keenan in Florida. Look for new life now that Luongo got a new start in Vancouver. I think 25th is still too low for him.

26. Martin Havlat (up from 28th) A great talent, but I'd like to see him have a year where he approaches 90 or 100 points before ranking him this high. Will he get lost in Chicago without the same calibre of linemates he was used to in Ottawa?

27. Marian Gaborik (up from 42nd) Another great talent who is yet to have an elite offensive season, however playing under Jacques Lemaire in Minnesota has made Gaborik much stronger defensively than Havlat and thus should be ranked ahead of him.

28. Henrik Zetterberg (debut) Although came through with an 85 point year (which is much bigger than Havlat or Gaborik did), but doesn't carry the same uside of either of them and has benefitted from from having many strong teammates. I am not convinced he can keep up that level of offense now that opposing teams will be keying in on him much more strongly.

29. Tomas Vokoun (debut) Worthy of a Vezina nomination, but I am, uncertain of his future given the blood clot that kept him out at the end of the year. That uncertainty would keep me from rating him this highly.

30. Rob Blake (down from 17th) At age 36, I am not sure he still has enough left to justify this high a ranking. The phyiscal part oBlake's game is gone now and often he relies on the reputation he has earned in past years. If players start to test him in his own zone will he stand up? He is still a top offensive defenceman.

31. Simon Gagne (debut) Gelled well with Peter Forsberg to start the season off with a lot of goals. His scoring rate slowed a bit when Forsberg ran into injury problems.

32. Dion Phaneuf (debut) He's not ready for this kind of a ranking. Sure he's a fine young defender who had an (undeserved) Calder nomination, but he is still beatable in his own zone and this became painfully obvious when Anaheim eliminated the Flames in the playoffs.

33. Ryan Smyth (up from 49th) Smyth really wasn't a better player in 2005/06 than he was in previous years, but his trip to the Stanley Cup finals got him more exposure in the eastern media. His real ranking should be in between these two extremes.

34. Markus Naslund (down from 3rd) Naslund played through injury most of last season and it kept him out of the Olympics. I think a healthy Naslund will bounce back and show this ranking is unreasonably low.

35. Pavel Datsyuk (up from 36th) I see him as the best Detroit forward. He should be ranked ahead of Zetterberg.

36. Jonathan Cheechoo (debut) He meshed extremely well with Joe Thornton and led the NHL in goals. Can he keep up that pace and how good is he if Thornton is not feeding him?

37. Patrick Marleau (up from 46th) The best number two centreman in the NHL last year. Needs to continue developing his defensive skills as Joe Thornton will play the other team's checking lineleaving Marleau to more often than not play the number one line of the opposing team. This wasn't an easy transition for him, as was shown by his -12 +/- rating.

38. Rod Brind'Amour (debut) The Selke winner and captain of the Stanley Cup winning Carolina Hurricane. He was embarrassingly left off the list of players eligible for the Olympics when he should have made the team. Since he is 36, he might slow down this year, but last year he showed no signs of it.

39. Shane Doan (debut) Suprisingly left off last year's list. Top scorer in Phoenix but he should be able to score more than he has so far. This past season was a bit of a disappointment for him both offensively and defensively (I suppose if you can disappoint and lead your team in scoring then you are a pretty good player).

40. Lubomir Visnovsky (debut) His 67 point year on defence was a suprise given he hadn't scored more than 30 points since 2001. I am not sure he can keep up this scoring rate, especially given that he won't be the clear go to offensive guy in LA with Rob Blake around.

41. Olli Jokinen (debut) Has developed into a very good forward since he arrived in Florida, but I am not confident in his consistency.

42. Sergei Zubov (debut) He is too good a player to be this low in the rankings. He was a deserving Norris trophy nominee and has a long string of solid seasons.

43. Alex Tanguay (up from 44) A better than point per game forward the last two seasons he played in the NHL. He deserves a better ranking than this one. I expect he and Iginla could really ignite a Calgary offence.

44. Robyn Regehr (debut) The best one dimensional defensive defenceman in the NHL today. The best defender in Calgary (why is Phaneuf rated higher?). That said he is not an offensive contributor so I don't think I would rank him this highly.

45. Bryan McCabe (debut) A somewhat one-dimensional offensive defenceman, but one who offers more than Regehr having made second team all star in 2004 and having been a leading Norris contender before injuries slowed him in the second half of the season.

46. Brendan Morrow (down from 41st) A very good two way forward, but since he has never approached a point per game season, I think his inclusion on this list is a (repeated) mistake. Why isn't Mike Modano the top rated Dallas forward?

47. Martin St Louis (down from 9th) The 2004 Hart trophy winner has dropped hard. I think he lost a lot of momentum losing a year during the lockout. Quite possibly it cost him a Hall of Fame career.

48. Mats Sundin (down from 15th) Sure he's 36, but he led Toronto in scoring last year so that kind of drop is unwarranted.

49. Marty Turco (down from 39th) Third in wins last year, but his saves percentage suffered. Still he has a track record as a top goaltender.

50. Brian Gionta (debut) Suddenly scored 89 points last year (previous career best was 29). I am not convinced he can repeat now that he will draw more coverage from opposing teams.

There is a thread on hfboards.com that discusses this list in detail if you want to see what more people have to say about it.

Tuesday, August 15, 2006

Adjusted Goal Scoring: Single Season

In sabremetrics and hockey one problem is to adjust scoring between different eras. The hockey outsider has a method that does this as well as any. It is not the only method. Similar methods are used by Darryl Shilling and Total Hockey.

All methods try to normalize away differences in games played, ice time and scoring rates in different eras of hockey. None are successful in removing the effect of quality of opposition, which is a significant factor that remains in the data.

One problem is to figure out the ten best single seasons in terms of adjusted goal scoring. Accoring to the hockey outsider method these are:

Top 10 Adjusted Goal Scoring Seasons of All Time
Name Team Year Games Played Adjusted Goals Actual Goals
Brett HullStL1990/91788286
Phil EspositoBos1970/71787776
Mario LemieuxPit1988/89767585
Wayne GretzkyEdm1983/84747387
Wayne GretzkyEdm1981/82807292
Mario LemieuxPit1995/96707169
Babe DyeTor1924/25297138
Pavel BureFla2000/2001826959
Pavel BureFla1999/2000746858
Phil EspositoBos1971/72766766

Only six different people appear in this table. Wayne Gretzky, Mario Lemieux, Pavel Bure and Phil Esposito appear twice each. According to this statistic, Brett Hull had the best goal scoring year ever. The only other player appearing in this list is Babe Dye. Nobody in the original six era makes this top ten goal scoring list. Seven of the ten years occured within 3 years of an expansion year (the exceptions are Brett Hull, Mario Lemieux in 1989 and Wayne Gretzky in 1984). This shows the effect of decreased calibre of opposition. In a couple future articles, I will look deeply into why these results were produced and if any flaws can be fixed.

Saturday, August 12, 2006

Adjusting Scoring From Different Eras

One problem in sabremetrics and hockey is how to compare scoring stats from one era to another. There are several strategies that have been used. I have introduced one used by Darryl Shilling and another used by Total Hockey. Today I will introduce a third method. Its conceptually similar to the other two, but shows the differences that can occur with slightly different methodology. It also gives me an opportunity to introduce a great sabremetrics site the hockey outsider (Peter Albert). This adjusted scoring method is his way of comparing players between different eras.

The first variable that influences scoring statistics from one era to the next is schedule length. In the early days on professional hockey, teams played as few as 16 games in a regular season. Today they play 82. Obviously, the more games a player plays the more chances he has to score. To adjust this, a statistic is multiplied by a schedule adjustment of 82/games played. For example, in a 50 game season it is multiplied by 1.64.

The second variable that influences scoring statistics from one era to another is the level of offence. In some eras, teams scored more goals per game than in others. A player from a higher scoring era would appear to have higher scoring numbers than a similar player in a lower scoring era, despite the fact they are similar. To adjust for this, a player's statistics need to be adjusted for the scoring rate in his era. This is complicated further, especially in the early days of hockey, where one high scoring player could account for a large portion of the goals in a league. In order to account for this, the player whose stats are being adjusted needs to have his offensive contribution subtracted from the league. This is done by subtracting the "goals created" by a given player from the league stats. In this analysis, a slightly different formula is used for goals created than in the Daryl Shilling goals created analysis. Here the formula used is
Goals Created = 1/2 * (Goals + Assists * League Goals / League Assists).
The intent of the formula is the same, but it is a slightly different approach. To adjust for the scoring rate in an era, statistics are next multiplied by a level of offense adjustment of 6.4/(League Goals - Player's Goals Created). Here 6.4 is taken as a value to approximate the longterm average of goals per game in professional hockey.

The next adjustment is that of assist frequency. In the early days of professional hockey there are very few assists per goal. That reduced player's offensive totals relative to today where there are more assist per goal handed out. This adjustment is 1.55/(League Assists/ League Goals). 1.55 is taken as a value to approximate the longterm average of assists per goal in professional hockey. Obviously this adjustment is not necessary when comparing goal scoring only, but when comparing assists or points from different eras it is.

The final adjustment is that of icetime. In the early days of professional hockey, it was no uncommon for a player to play the entire game. Today there are four lines that are often given relatively even ice times. To adjust playing time from one era to the next, it is necessary to adjust average playing times for a player of that time. Individual players will be different from the averages, but in the earlier days of hockey, there is no record of how much playing time a player had. Today, there are 60 minutes in a standard NHL game. There are 18 non-goalies competing for icetime and five players are on the ice at any given time. So 60 * 5 / 18 = 16.7 minutes per player per game. This analysis assumes no shorthanded situations or goalies getting pulled and no overtime, so it is not exact, but it gives a number quite close to the average ice time per player. Ice time will vary for forwards vs. defencemen (defenders tend to get more ice time in a game), but again this complication is ignored. In a given era, the same calculation can be carried out. 60 minutes * number of skater positions on the ice / number of skaters on a roster (= era adjusted ice time per player). The roster size changed over time and in the very early days so did the number of skater positions since they had a seventh man on the ice called the rover. This ice time adjustment is 16.7 / Era adjusted ice time per player.

With these four adjustments we can attempt to compare statistics from different eras. This does not capture all the possible era adjustments that might exist. The biggest one missing is that of quality of opposition. In some years, the NHL wa better than in others. It was easier to score in the weaker (for example expansion or war) years than in the higher quality years, since you had weaker players to beat in order to score. This is a large adjustment that is missing. It is missing, because it is extremely hard to find an independant measure for the quality of a professional league in a given season. I will soon post giving some results of this analysis.

Wednesday, August 09, 2006

Ownership of Stats For Fantasy Sports

Major league baseball has been forcing internet fantasy baseball leagues to pay a licence fee to use statistics and player names. This has killed some fantasy baseball games including Swirve Baseball. This shortsighted plan may have made MLB some money but it has alienated fans of fantasy games.

CBC Marketing and Distribution sued MLB for the right to hold their CDM fantasy baseball game unlicenced. They won. This practise of Major League Baseball has been found to be illegal.

This is good news for anyone who plays fantasy sports.

TSN has a story about this here.

Revenue Sharing and $36 Million

One aspect on the NHL CBA that will have a strong effect on the way some teams conduct their affairs is the revenue sharing agreement. This agreement is very complicated, but a summary of it can be found here.

This agreement isn't new this season, but it will affect this season differently than it did last year.

Last year the pot of revenue sharing monies potentially came from four sources. It came from centrally generated league revenues (TV contracts and mechandising), it came from left over escrow money due to the owners, it came from the playoff revenues of those teams that had playoff runs and it was directly contributed by the top ten revenue generating markets. There is a complicated formula for exactly how much comes from each source, but that is where the money comes from. The money is distributed to the teams that are among the 15 lowest revenue producing markets (as long as they are not from the very largest markets New York - Islanders and Rangers, Chicago or Los Angeles - Kings and Ducks). Exactly how much each team gets is again determined by a complex formula.

Now here is the change this season. Likely, escrow money that goes to the owners will be much larger than it was last year. This is because real revenue figures are used in the revenue projections instead of the lowball guesses used last year (in part to have an easy goal to exceed). In fact, the expectation of huge escrow payments worried the NHLPA so much that Ted Saskin got the salary reduced from the figure it would have been using the CBA formula. It is clear that more than half of the NHL's teams will pay more than the $36 million midpoint (halfway between the salary cap and floor) that would lead to zero escrow payments if the NHL exactly meets revenue projections. It is clear that those teams that exceed the midpoint will do so by a larger margin on average than those who fall below it. It is clear that the escrow account will be much larger this year than it was last year. Now the revenue sharing formula asks for 33% of money (which will be a minimum of 4.5% of the revenue of the NHL) coming from escrow - should this money exist in the escrow account. If we project an NHL revenue of $2.1 billion this is $94.5 million. $31.5 million would come from escrow accounts. Again using a $2.1 billion total revenue, this means the 54% of revenue that goes to the players is $1.134 billion in total. That works out to $37.8 million in player costs per team. Contracts are not all signed for next season, but total player costs obtained by adding up all the team salary cap hits at Irish Blues is already $1.194 billion. Even if we assume that there are no more contracts at all this season, that leave $60 million in escrow money at the end of the year. That makes a total escrow account of $60 million. $31.5 million goes directly to revenue sharing. What happens to the other $28.5 million (which we have likely underestimated)? It also gets shared but in a different scheme. It is gets shared prefernetially among the teams that don't pay their team as much as the salary midpoint. Right now only 5 teams (who may still sign players and add payroll) would qualify. They are: Edmonton, Nashville, Pittsburgh, St Louis and Washington. These teams would stand to get a few million dolars each for keeping payroll below $36 million. That is why teams are keeping the $36 million payroll figure in their thoughts.

Should the season go badly, it makes sense to dump enough payroll to get below the salary midpoint. That is part of the thinking of Boston and Buffalo when they walk away from salary arbitration deals. Should we have a bad season, we want to be able to reduce our payroll enough to gather these additional revenue sharing dollars. If we don't think we can have a playoff run, it makes financial sense to chase this money instead. Even if our payroll is too high at season's start, because we think we can compete, we will reduce it later if it turns out we do not compete. That is why taking on additional salaries that are larger than we planned is a bad idea.

This idea is one Tom Benjamin has kicked around on his blog, but most fans are unaware of the implication of this revenue sharing. That is why I thought it was a good idea to spell it out.

I predict that the mid-season salary dumping trades of at least one also ran team will make sense in this scheme. They want the additional revenue sharing money and are reducing payroll to get it.

Tuesday, August 08, 2006

Buffalo Walks Away From Dumont

The Buffalo Sabres have become the second team this year to walk away from an arbitration ruling and let their player become an unrestricted free agent. Buffalo joins Boston in making this move. Buffalo has walked away from the $2.9 million salary arbitrators gave to Jean-Pierre Dumont.

Dumont is a solid offensive talent who has put up 40 or more points in four of his last five NHL seasons. He is one piece of the offensive depth that helped Buffalo have a good playoff run that took them within one game of a Stanley Cup finals berth.

A $2.9 million salary seems a little high for Dumont (especially when compared the the $3 million given to Ladislav Nagy - who is a much better player), but this move does nothing to improve the Buffalo Sabres. This is the kind of move that a small market team has to make from time to time. They cannot afford all their good players (they cannot afford to spend to the salary cap) so they have to let some go. I seem to recall Gary Bettman promising something about this CBA helping small market teams keep their talent - but I guess that is public relations and not reality.

Buffalo (much like Boston) is trying to keep their player costs below $36 million (and not the $44 million figure) because this allows the Sabres a shot at revenue sharing money.

As the salary cap continues to rise, we will see more and more instances of small market teams who cannot afford to spend at the salary cap level and let their talent get away since they cannot afford them. The bigger markets will not have this problem and will be able to bring in some of the talent. Sure a salary cap limits how much of the talent they can acquire, but this isn't an issue of quantity of talent, its an issue of quality of talent. As free agency rules liberalize, more and more talent will be available to be purchased. The best players in the NHL will be for sale - and at a young enough age that they will still have many great years left.

It is true that Dumont's $2.9 million salary is on the high side. It is also true that a team in a big market with salary cap room to spare (Buffalo has salary cap room to spare) who thinks they have a good chance to contend (Buffalo came very close to making the 2006 Stanley Cup finals) would bite the bullet and pay Dumont. Buffalo doesn't have the same options in their small market. Such is life in the NHL. This CBA won't change it and if it leads to Buffalo's talent reaching free agency at a younger age, it will only make the problem worse.

Here is TSN's story on the Buffalo announcement.

Saturday, August 05, 2006

Bruins Walkaway From Tanabe

The Boston Bruins made a big leap into the free agency pool adding Zdeno Chara and Marc Savard. Just when it looked like they were about to get past the cheapness that surrounded the Mike O'Connell reign of error, they show us that maybe this isn't true. Boston announced today that they are walking away from David Tanabe's one year $1.275 million arbitration award.

Tanabe is a 26 year old defenceman that was expected to fill a role on the Bruins 2nd or 3rd line. He has almost 400 games NHL experience and is a solid but so far unspectacular defenceman. His game is an offensive game and he has struggled to learn the finer points of defensive play. He was a first round pick by Carolina in 1999 and is reaching the age where he will likely enter the prime of his career. Defenceman often take a big step forward around Tanabe's age. I don't think it is unreasonable to expect that he might as well. If he does, he would certainly be worth $1.275 million. He most certainly would provide more than youngsters Andrew Alberts and Milan Jurcina, who will be asked to fill his minutes.

Tanabe is now an unrestricted free agent. He can sign with any team that wants him. He suddenly becomes one of the most attractive ufa defenceman left on the market (disagree? pick another - not too many options left are there and those that exist are well into their 30's).

Boston is probably shooting for revenue sharing money if they keep payroll below $36 million. Tanabe probably improves the bruins marginally over any other defenceman they can give his minutes to at this point, but who cares? Any fans who have been enticed to buy back into the Bruins because of the Chara and Savard signings probably won't have second thoughts over a cheap decision to not pay a depth defender who might be about to develop into a decent defender.

Here is a TSN story on the issue.

Mikhnov's Play To Get To The NHL

Since there is no Russia - IIHF player transfer deal this year the status of players under contract in the Russian leagues who want to play in the NHL is unclear. This leaves some high profile prospects like Evgeny Malkin unclear on what it would take for them to come to the NHL.

The Russians believe that player transfer fees are too low and want to negotiate directly with NHL teams for transfer fees of specific players. Gary Bettman has refused to allow NHL teams to directly negotiate transfer fees. Currently, transfer fees are hundreds of thousands of dollars per player and vary with the position the player was drafted.

One player stuck in this limbo is Alexei Mikhnov, who was drafted by the Edmonton Oilers in the first round in 2000. He will be the test case for a end run around the situation. Russian labor law states that anybody can leave any job with two weeks written notice even if they have a contract. Mikhnov has submitted his written notice that he is leaving.

So what happens next? Probably court cases on this subject. Cases occurred last year regarding the status of Alexander Semin and Alexander Ovechkin. Mikhnov's team Yaroslaval Lokomotiv may be left with nothing for his departure or might be able to garner a large settlement that dwarfs anything the IIHF deal would offer. There are rumored links between the Yaroslavl team and the Russian mafia which could make Mikhnov's departure from Russia a complicated one if he doesn't play ball with them.

This is a story to watch this summer.

Here is a Jim Matheson story on the subject.

Thursday, August 03, 2006

No Russian IIHF Deal For This Year Either

The teams in the Russian Elite League have decided not to sign the IIHF player transfer deal that Russian Ice Hockey Federation president Vladislav Tretiak agreed to in June. This will be the second year where there is no NHL- Russia IIHF transfer deal. This deal governs the co-existence of the Russian leagues and the NHL. Among other things, it outlines how much money the NHL must pay to the Russian leagues if they sign a player who is under contract in Russia. The Russians have maintained that these payments are not enough (they are based on the draft position of the player and typically several hundred thousand dollars).

This leaves a situation where there will likely be legal disputes when over the status of Russian players bewteen the two leagues. Last year, disputes occurred on the status of Alexander Semin and Alexander Ovechkin. This year, the most likely target for such legislation is Evgeni Malkin of the Pittsburgh Penguins who is an early Calder favorite.

Without a player transfer deal, NHL players may leave in mid-contract to join Russian teams and Russian league players may leave in mid-contract to join NHL teams. It is up to the individual teams involved to negotiate or otherwise determine the details of the player transfers. It is possible that a player like Malkin could require a multi-million dollar payment to free him from his Russia team Metallurg Magnitogorsk. It is also possible that he could leave with no payment whatsoever. It complicates the transfer of players such as Malkin to the NHL.

Here is an ESPN story on this situation.

Pnep is at it Again

Pnep (Roman Nepomnyaschev) has recently updated his Hall of Fame monitor. Previous incarnations have been mentioned in previous posts in this blog (here are forwards, defenders and goalies. Although these systems do a good job of ranking players based on what they accomplished in their careers, they can run into problems as they rely on awards won by players (which may carry poor voting from awards voters into rankings) and because some aspects of players skill are easier to measure than others (offence is easier to measure than defence).

Here are the updated formulas:

Adjusted PTS/2 +
Adjusted PO PTS +
Captainship - 15 pts per season +
Top 10 in Goals after Season - 10 pts +
# 1 in Goals after Season - 100 pts +
# 2 in Goals after Season - 70 pts +
Top 10 in PTS after Season - 10 pts +
# 1 in PTS after Season - 150 pts +
# 2 in PTS after Season - 100 pts +
Cup - 50 pts (minimum 5 Adj PO GMS) +
Final - 25 pts (minimum 5 Adj PO GMS) +
All Star Game - 5 pts +
HART - 150 pts +
HART Runner Up - 100 pts +
BYNG - 75 pts +
1 ALL STAR TEAM - 75 pts +
2 ALL STAR TEAM - 50 pts +
CALDER - 50 pts +
CALDER Runner Up - 30 pts +
SELKE - 50 pts +
SELKE Runner Up - 35 pts +
CONN SMYTHE - 100 pts

Adjusted Games/5 +
Adjusted Pts/10 +
Adjusted PO Games/2 +
Adjusted PO Pts/4 +
Captainship - 15 pts per season +
Top 10 in Goals after Season - 10 pts +
# 1 in Goals after Season - 100 pts +
# 2 in Goals after Season - 70 pts +
Top 10 in PTS after Season - 10 pts +
# 1 in PTS after Season - 150 pts +
# 2 in PTS after Season - 100 pts +
Cup - 50 pts (minimum 5 Adj PO GMS) +
Final - 25 pts (minimum 5 Adj PO GMS) +
All Star Game - 5 pts +
HART - 150 pts +
HART Runner Up - 100 pts +
1 ALL STAR TEAM - 75 pts +
2 ALL STAR TEAM - 50 pts +
CALDER - 50 pts +
CALDER Runner Up - 30 pts +
NORRIS - 150 pts +
NORRIS Runner UP - 125 pts +
CONN SMYTHE - 100 pts

Adjusted Wins/2 +
Adjusted PO Wins*4 +
Cup - 50 pts (minimum 4 Adj PO Wins) +
Final - 25 pts (minimum 4 Adj PO Wins) +
All Star Games - 5 pts +
Calder - 50 pts +
Calder Runner Up - 30 pts +
1 ALL STAR TEAM - 150 pts (seasons 1917-81); 75 pts (seasons 1981-06) +
2 ALL STAR TEAM - 100 pts (seasons 1917-81); 50 pts (seasons 1981-06) +
Vezina - 100 pts (seasons 1917-81); 150 pts (seasons 1981-06) +
Vezina Runner Up - 75 pts (seasons 1917-81); 125 pts (seasons 1981-06) +
Conn Smythe - 100 pts +
Hart - 150 pts +
Hart Runner Up - 100 pts

In the years before awards existed, he uses projected award winners that can be downloaded here. Statistics are adjusted as described here.

This is a lot of work and it is very good work, but it still has problems. There is a non-symmetry across positions required to calibrate between different positions. For example, defenceman get points for games played and no other positions do. Does this make sense?

The biggest changes are adding points for being a team captain and for runs in the Stanley Cup playoffs. These are generally things that are done by good players - but not always. There are bad players who play on good teams and have Stanley Cup runs and there are good players on weak teams who do not. A team captain is a political position usually given to a top player on a team, but it isn't the best way to determine who is a top player. Its benefit is that it is easy to determine who was captain, so its not too difficult to add this to the HOF monitor.

It is argued that the Hall of Fame monitor is intended to determine who will be inducted into the Hall of Fame given the preferences of voters and not who should be inducted and they tend to look at things like captaincy and Stanley Cup playoff runs, so they should be included.

I disagree. I disagree because we have seen recent Hall of Fame inductions. They select unworthy players because they are nice guys who are borderline candidates who are friendly with the Hall of Fame committee (Dick Duff). If you really want to capture the biases of the Hall of Fame committee you should be given points for friendship with committee members. This is a rather subjective thing to do. How do I quantify if player X is a friend of Hall of Fame committee member Y. Did they appear on the same team at one time? Were they both in the same celebrity golf tournament? It's not something that should be analyzed.

I think pnep has done well using statistics that are available to rank players. He adds things that might be seen as important to Hall of Fame voters that are easy to quantify. It is a good list, but it should be used not to try to mimic the often irrational decisions of the Hall of Fame committee, it should be used to rank the best players of all time.

The problem with ranking the best players of all time is that often the contribution of a player does not fully show up in his statistics. This is the primary problem with sabermetrics and hockey. In order to have a truly definitive player ranking system, we must be able to quantitatively show a player's value toward making his team win games. The results of this ranking system should be used to predict award winners. It shouldn't use award wins as an input.

Is such a system possible? I don't think it is with the hockey statistics that exist right now. Too much of the game is undocumented statistically. Even if future games were better statistically analyzed, the old games will not have the "modern statistics". Thus ranking players of the past will be a guesstimate at best on how they would have fared. I don't have any such system myself. I doubt it is possible to have a system as successful as those that exist in baseball. I am very interested in any such attempts and will post and critique them here. This pnep system is the best that exists (that I have run across). I hope it can lead to bigger and better results in the future.

Here is an hfboards discussion of the newest hall of fame monitor.

This page is powered by Blogger. Isn't yours?