2005-06 Hart Trophy Revisit

[wetcoast]

But if your season is divided into two teams, how is the player "most valuable to his team"?

He was by far the best and most valuable Bruin during his time there and even more valuable in San Jose.

I also have to take issue with the comment above from a poster I usually agree with 99% of the time but Jonathan Cheechoo and his goal total that year are simply amazing.

sure he scored 28 goals 2 years before but he was a grinder hard effort guy not a sniper and this is a tale of 2 seasons.

Pre Thornton 24-7-8-15

Post Thornton 58-49-29-78

That 49 goals is 12 more than Ovechkin in the same time period.

Jonathan also went from being a minus 9 in 24 GP to a plus 32 in 58 GP.

I venture to say that Thornton had a huge impact here.

 

[wetcoast]

Funny thing about Jagr is this isn't even his closest call to me - I thought he deserved the Harts in 2000 and 1995 even more than this one

It's funny how with Jagr - he only has 1 hart, but changing very little to history he could so easily have had 5 or 6. Really shows how much competition and timing of competition can impact trophy counting.

Some time ago I starting thinking that all time Hart (or whatever) trophy is really important to focus on, as often too much can be made out between guys who win versus who didn't even if the differences are so minute that they don't really matter.

 

[The Panther]

It is an interesting question about who had the better season (never mind "more valuable to his team" / Hart stuff) -- Thornton or Jagr.

Jagr won the Pearson, which (I think) means he was voted "best player in the NHL" by his peers.

Goals
54 - Jagr
29 - Thornton
Assists
96 - Thornton
69 - Jagr
Points (PPG)
125 - Thornton (1.54)
123 - Jagr (1.50)
"Primary" points (i.e., minus secondary assists)
101 - Jagr
99 - Thornton
Scoring at 20 games
30 - Thornton
29 - Jagr
Scoring at 41 games
63 - Jagr
61 - Thornton
Scoring at 61 games
92 - Thornton
91 - Jagr
Plus / Minus
+34 Jagr
+31 Thornton
ES points
72 - Thornton
71 - Jagr
PP points
52 - Jagr
51 - Thornton
Percentage of player's points from team's total goals
48% - Jagr
44% - Thornton

_____________________

You could really flip a coin between them in terms of who had the better season.

However, since Jagr (a) scored points in a higher percentage of his team's goals, and (b) played all the games for one team, I think he should have won the Hart.

 

[Hockey Outsider]

Alan Ryder's Player Contribution statistic - an all-in-one stat which was like a better version of Point Shares, but was still limited by the quality of the underlying statistics - rated Sergei Zubov as the most valuable player for the 2005-06 season.

A surprising result, considering Zubov was only 3rd in Norris voting with no first place votes, and didn't receive a single vote for the Hart trophy. But consider his team context and you can see how the numbers would point this way. Dallas finished with 112 points and the third best record in the league. And Zubov played almost 500 more minutes than the #2 Dallas skater, led the team in TOI in all situations, and was only 6 points behind Modano for the team lead in points.

Zubov's shootout success put him over the top in the Player Contribution formula. Dallas dominated the shootout in its first season with a 12-1 record, led by Zubov and Jussi Jokinen. Zubov scored on 7 of 12 shots and led the team with 4 game deciding goals in the shootout. He was particularly impressive relative to his blueline peers, as his 7 shootout goals almost matched the combined season total for all other defencemen in the league (10), and his 4 game deciding goals were more than all other defencemen combined (3, including Marek Malik's legendary goal). Dallas took 24 points out of 13 games that went to shootouts, and if you give Zubov full credit for his contribution to those points, it's quite a bit of value.

But we've had the shootout for almost 20 years now and I don't know that anyone actually looks at shootout success when voting for the Hart trophy. I'm generally OK with ignoring it as a regular season gimmick with no value in the playoffs. But on the other hand, if you are voting on a regular season award, shouldn't you consider value added in the shootout as well?

Of the other Hart trophy contenders listed here, Ovechkin was the only skater to add value in the shootout.

Ovechkin - 6/13
Crosby - 2/6
Jagr - 2/8
Alfredsson - 2/8
Staal - 1/4
Thornton - 1/5
Lidstrom - 0/0
Niedermayer - 0/1

Looking at the goalies, Brodeur and Lundqvist were both above average in the shootout this season. Kiprusoff was pretty bad, and if you count this against him it probably takes him out of Hart contention.

Brodeur - 8-3, saved 29/38 (0.763)
Kiprusoff - 1-7, saved 11/23 (0.478)
Lundqvist - 4-3, saved 28/37 (0.757)

In the end I don't think I would have put full weight on the shootout results, and I would have voted Jagr and Thornton as the top 2. But it's interesting that there is a statistical argument for a player who didn't receive a single vote.

I dug up Ryder's article for 2006. A few observations:

You're right, Zubov got a ton of credit for his success in the shootout. Excluding the shootout, Lidstrom was the top skater. (A tangent - I hardly ever hear people talk about the shootout in making the case for the Hart. I hate the shootout, and obviously it has no carryover value to the playoffs, but it seems odd for people - myself included - to simply ignore it, when it impacts the regular season results).

Ryder's system is harsh towards Thornton. It ranks him as only the 12th most valuable forward (still only 8th if we exclude shootouts). It rates Cheechoo as significantly more valuable (123 vs 101 PC "points"). The same thing happened with Forsberg and Hejduk in 2003. The systems treats goals as more valuable than assists (which is true in general - but certainly not in the case of an all-time great playmaker helping their linemate reach a career high in goals). We also see that, to a lesser extent, with Gagne vs Forsberg.

It's very tough to measure defense in a comprehensive way. Ryder's system reads Pronger as the best (most valuable?) defensive player in the league. Probably a reasonable conclusion. Lidstrom is second.

 

[MadLuke]

I dug up Ryder's article for 2006. A few observations:

You're right, Zubov got a ton of credit for his success in the shootout.

Do you remember what was considered a shootout game winning goal (I imagine it is impossible to distinguish the value for any goal scored by the winning team, does any scorer get a lot of wins points ? and do they get full credit or just how much better than their average replacement would have)

It is indeed strange how little value we come with (except indirectly has we punish a lot a player for missing the playoff) shootout and hart, even for a goaltender were it should be relatively direct, it is a lot of points changing hands.

On the other hands was there not strong evidence that there almost 0 link about being good at it one season and being good at it the next season, despite hart candidate rarely shifting team, so maybe a single player cannot do much.

If a mediocre (among those who get a lot of shoots) shootout player is a 30% and all time good one is at 40% and that having the chance of taking 10 shootouts shots a year is a lot, top of the league amount, how much does an elite shooter versus your average stars can really do here, by scoring one more goal that would win possibly (maybe not) to an extra points ? 2 goals if they shoot 50% having a peak year a la Toews.

I fear if it get a giant amount of value that it does not take well how many goals the replacement would have scored...

 

[overpass]

I dug up Ryder's article for 2006. A few observations:

You're right, Zubov got a ton of credit for his success in the shootout. Excluding the shootout, Lidstrom was the top skater.

Yeah, Lidstrom was just a bit ahead of Zubov in all the other stats, so that makes sense that he scored higher excluding the shootout.

I don't agree with all the inputs and assumptions to Ryder's model. For example, I think the Goals Created offensive method undervalues assists and overvalues goals. Ryder admitted himself in a later article that his system gave too much credit to Cheechoo and not enough to Thornton. And there are problems with his shot quality model. I believe it had Boston as a poor defensive team in 2010-11, and he called Chara the most overrated player. I don't agree with those assessments at all (not to get into the 2010-11 thread debate here).

Do you remember what was considered a shootout game winning goal (I imagine it is impossible to distinguish the value for any goal scored by the winning team, does any scorer get a lot of wins points ? and do they get full credit or just how much better than their average replacement would have)

I don't think Ryder gave extra credit to shootout game winning goals. I listed them because they are listed on nhl.com in the shootout report and are easily available there. There isn't a definition on how to define them.

NHL Stats

The official source for NHL Stats including skaters, goalies, teams stats and more.

www.nhl.com

I didn't find the 2005-06 season details, but here's Ryder's article on how to allocate value from the shootout.

The Shootout « Hockey Analytics

hockeyanalytics.com

He allocates all the team points to players, based on the player's marginal contribution above a certain replacement threshold.

For shooters, marginal shootout goals are calculated as shootout goals scored less a threshold level of goals, given the number of shootout chances (for goaltenders the calculation is similar, based on saves). I won’t go into how I determined these thresholds, but they are representative of ‘zero value added’ performance.

And more on Zubov's shootout performance.

There was a great deal at stake in the shootout (145 points) last season. And a limited number of players had a very significant impact. Of Dallas’ 12 shootout points, I attribute about 3.5 to Jokinen and 2.3 to Zubov (I allocate team shootout points amongst players based on marginal shootout goals). Both of these performances were statistically significant (unlikely to be simply lucky) and therefore representative of a repeatable skill.

To put these contributions into perspective, over an 82 game season, an average skater contributes about 4 points to his team’s success while an all-star skater typically contributes between 10 and 13 points (goalies can contribute more). Jokinen did an average season’s work in just 13 trips to the net. In the case of Zubov, this performance was the determining factor in making the Dallas defenseman, according to my analysis, the league’s most valuable skater last season (note that Joe Thornton was invisible in the shootout). Five other shooters contributed at least two points to their team’s shootout success:

 

[MadLuke]

Sound good, I forgot just how less shootout the nhl has now (3v3) versus back then, would have to look at it.

Dallas got 12 extra points that year (league record) on 13 try, mediocre team would have got around 6.5 pts

There 5.5 pts above mediocrity, 3.5 to Jokinen, 2.3 to Zubov, 2.2 to Turco even if ytou give 0 to everyone else you are already at 8 here, I imagine there a lot of value of reaching mediocrity in the nhl, specially in we are sending our best goaltender and a top 3 skaters at something challenge (being mediocre in that small of a group is extremely good).

What would make it a bit boring to consider them is the case that the Oilers went into 13 SO that year and the Sens-Sharks only 8....

 

[buffalowing88]

I remember there being a lot of pushback in the media against a perceived "East Coast bias" for MVP voting. It does seem like it skewed to that conference from 90-06 pretty significantly and I think voters pushed back against that notion and it hurt Jagr, who was deserving.

 

[Hockey Stathead]

Sound good, I forgot just how less shootout the nhl has now (3v3) versus back then, would have to look at it.

Some data for this:

 

[SillyRabbit]

Funny thing about Jagr is this isn't even his closest call to me - I thought he deserved the Harts in 2000 and 1995 even more than this one

It's funny how with Jagr - he only has 1 hart, but changing very little to history he could so easily have had 5 or 6. Really shows how much competition and timing of competition can impact trophy counting.

Definitely feels like Jagr got the short end of the stick when it came to Hart voting more than almost any other player I can think of.

 

[EpochLink]

I honestly thought Jagr would win the Hart this year, not surprised Thornton won it.