Stanley Cups relative to expectations (1985-2019)

[Hockey Outsider]

Purpose - This thread is inspired by a recent observation from @overpass - namely, that we can assess how many Stanley Cups a player won, relative to the expectations of his team's strength, by taking into account the pre-season odds.

Method - I took the pre-season Stanley Cup odds from hockey-reference.com. They have a complete listing going back to 1985. I converted the odds each year into percentages. I adjusted the percentages so that they sum to 100% each year. (Without this adjustment, the percentages would add to over 100%, as there's a profit element built in. Interestingly, the profit element has decreased steadily over time - perhaps this is due to more online competition that drives the spreads downwards - I don't gamble on sports so maybe someone who's done that can confirm if that's correct).

As I usually try to do, I'll offer some self-criticisms of this approach:

• This is a fairly crude analysis. It only looks at whether a team wins the Stanley Cup or not. It doesn't take into account whether they lost in game 7 of the Stanley Cup finals, or if they were swept in the first round. An obvious way to improve this analysis would be to get the odds series by series.
• Although I do some individual analysis, it's questionable if it's truly appropriate to apply this to individual players. It's likely not appropriate to ascribe all of the differences to the players I study (i.e., a team might lose due to poor goaltending, which likely isn't the top scorer's fault).
• The analysis is based upon pre-season odds. It would be more instructive to use odds right before the playoffs began, as that would take into account significant injuries, mid-year trades, etc.
• The purpose of gambling organizations is to maximize their profit (not necessarily to assess the strength of teams). To the extent that there's an efficient market, these would give similar evaluations. If you believe that gamblers exploit particularly optimistic fanbases, the results may be skewed.

Click to read more...

 

[Hockey Outsider]

TOP FIFTEEN PRE-SEASON STANLEY CUP FAVOURITES

Rank	Year	Team	Odds	Result
1	1988	Edmonton Oilers	25.9%	Won Stanley Cup
2	1987	Edmonton Oilers	23.9%	Won Stanley Cup
3	1986	Edmonton Oilers	22.0%	Lost in 2nd round
4	1985	Edmonton Oilers	21.9%	Won Stanley Cup
5	1989	Calgary Flames	19.7%	Won Stanley Cup
6	1990	Calgary Flames	18.7%	Lost in 1st round
7	1996	Detroit Red Wings	16.5%	Lost in 3rd round
8	1991	Calgary Flames	16.3%	Lost in 1st round
9	1986	Philadelphia Flyers	16.1%	Lost in 1st round
10	2000	Dallas Stars	15.9%	Lost in SC finals
11	2000	Detroit Red Wings	15.9%	Lost in 2nd round
12	2003	Detroit Red Wings	15.2%	Lost in 1st round
13	1997	Detroit Red Wings	15.0%	Won Stanley Cup
14	1988	Philadelphia Flyers	15.0%	Lost in 1st round
15	1985	New York Islanders	14.6%	Lost in 2nd round

[TBODY] [/TBODY]

Of the top fifteen pre-season Stanley Cup favourites, five went on to win the Stanley Cup (including four of the top five teams). This is better than expected, as the odds implied that these teams would have combined to win approximately 2.7 Stanley Cups. Surprisingly, five of these teams lost in the first round - the early 1990's Flames and mid/late-80's Flyers both make two appearances on that list.

The biggest surprise is the 1998 and 2002 Red Wings didn't make the list. Note, as I mentioned before, that these are based on preseason expectations. So, the historically dominant 2019 Lightning have relatively modest 9.1% odds (which would surely be higher if the odds were recalculated in early April).

TOP FIFTEEN STANLEY CUP UNDERDOGS

Rank	Year	Team	Odds	Result
15	2000	Tampa Bay Lightning	0.2%	No playoffs
14	2000	Atlanta Thrashers	0.1%	No playoffs
13	2001	Atlanta Thrashers	0.1%	No playoffs
12	2001	Columbus Blue Jackets	0.1%	No playoffs
11	2001	Minnesota Wild	0.1%	No playoffs
10	1996	Ottawa Senators	0.1%	No playoffs
9	1994	San Jose Sharks	0.1%	Lost in 2nd round
8	1994	Tampa Bay Lightning	0.1%	No playoffs
7	1995	Ottawa Senators	0.1%	No playoffs
6	1994	Florida Panthers	0.1%	No playoffs
5	1994	Mighty Ducks of Anaheim	0.1%	No playoffs
4	1994	Ottawa Senators	0.1%	No playoffs
3	1993	San Jose Sharks	0.1%	No playoffs
2	1993	Tampa Bay Lightning	0.1%	No playoffs
1	1993	Ottawa Senators	0.0%	No playoffs

[TBODY] [/TBODY]

Not a terribly surprising list, with a lot of recent expansion teams. All but one of these teams failed to qualify for the postseason. It's something of a minor miracle that the 1994 Sharks qualified for the playoffs at all, let alone knocking out a 100-point team.

Also note that, on both of these lists, there aren't any teams from 2006 onwards. This is a result of the salary cap - there's more parity, so it's harder to identify the favourites and the underdogs, especially at the start of each season. Your mileage may vary as to whether this is better or worse for the league.

 

[Hockey Outsider]

FRANCHISE SUMMARY

Franchise	Expected	Actual	Difference
Pittsburgh Penguins	2.02	5.00	2.98
Edmonton Oilers	1.75	4.00	2.25
New Jersey Devils	1.39	3.00	1.61
Chicago Blackhawks	1.44	3.00	1.56
Detroit Red Wings	2.52	4.00	1.48
Los Angeles Kings	1.11	2.00	0.89
Colorado Avalanche	1.15	2.00	0.85
Carolina Hurricanes	0.37	1.00	0.63
Montreal Canadiens	1.55	2.00	0.45
Anaheim Ducks	0.65	1.00	0.35
Tampa Bay Lightning	0.70	1.00	0.30
Arizona Coyotes	0.05	-	(0.05)
Vegas Golden Knights	0.06	-	(0.06)
Atlanta Thrashers	0.11	-	(0.11)
Dallas Stars	1.15	1.00	(0.15)
Minnesota North Stars	0.19	-	(0.19)
Mighty Ducks of Anaheim	0.20	-	(0.20)
Columbus Blue Jackets	0.23	-	(0.23)
St. Louis Blues	1.24	1.00	(0.24)
Hartford Whalers	0.27	-	(0.27)
Phoenix Coyotes	0.31	-	(0.31)
Quebec Nordiques	0.33	-	(0.33)
Florida Panthers	0.38	-	(0.38)
Minnesota Wild	0.40	-	(0.40)
Winnipeg Jets	0.42	-	(0.42)
New York Rangers	1.42	1.00	(0.42)
Nashville Predators	0.49	-	(0.49)
Calgary Flames	1.49	1.00	(0.49)
Washington Capitals	1.58	1.00	(0.58)
Boston Bruins	1.67	1.00	(0.67)
Ottawa Senators	0.72	-	(0.72)
New York Islanders	0.81	-	(0.81)
San Jose Sharks	0.90	-	(0.90)
Toronto Maple Leafs	0.94	-	(0.94)
Buffalo Sabres	0.96	-	(0.96)
Vancouver Canucks	1.04	-	(1.04)
Philadelphia Flyers	1.98	-	(1.98)
TOTAL	34.00	34.00	-

[TBODY] [/TBODY]

I don't think the results are terribly surprising overall. The Penguins won nearly three more Stanley Cups than expected over the past 35 years. They were expected to win almost the same number of Cups as their cross-state rivals, but they won five while the Flyers won zero. Any team with even two Stanley Cups has slightly out-performed expectations. On the other hand, the Sharks, Leafs, Sabres and Canucks are near the bottom of the list, with the aforementioned Flyers in a class of their own.

 

[Hockey Outsider]

PLAYER ANALYSIS - GOALIES AND DEFENSEMEN

PLAYER	EXPECTED	ACTUAL	DIFFERENCE
Patrick Roy	1.59	4.00	2.41

[TBODY] [/TBODY]

Roy won significantly more Stanley Cups than expected over his career, given the strength of his teams. As I mentioned before, it's probably inappropriate to ascribe all of the variances to a single player, but given that he won three Conn Smythes, and was likely the runner-up for a fourth, it seems reasonable in this case. He had close to an even split - he was expected to win 0.86 Cups in Montreal and 0.73 Cups in Colorado (winning two in both cities).

PLAYER	EXPECTED	ACTUAL	DIFFERENCE
Dominik Hasek	0.56	1.00	0.44

[TBODY] [/TBODY]

It's absolutely true that Hasek, on average, played on weaker teams than Roy, and was expected to win few Stanley Cups. But even after taking the strength of their teams into account, Roy still won more Stanley Cups (both in terms of the raw number, and as a percentage of expectations). Note that I've included Hasek's 1994 to 2001 seasons in Buffalo, his 2002 and 2007 seasons in Detroit, and his 2006 season in Ottawa. I've excluded his 2008 season in Detroit, where he and Osgood split the regular season minutes nearly evenly, while Hasek played four games in the first round. (Detroit won the Cup that spring, but I don't think it was due to Hasek's 2-2 record and sub 89.0% save percentage).

PLAYER	EXPECTED	ACTUAL	DIFFERENCE
Martin Brodeur	1.15	3.00	1.85

[TBODY] [/TBODY]

It's true that Brodeur sometimes gets downplayed due to the strength of his team - but three Stanley Cups in this era is impressive. The Devils were only expected to win 1.15 during that period. (I'm looking at 1994 to 2014 - excluding his short stints in 1992 and 2015).

PLAYER	EXPECTED	ACTUAL	DIFFERENCE
Henrik Lundqvist	0.48	-	(0.48)

[TBODY] [/TBODY]

I was curious to see Lundqvist's results - he's a good example of a player who's had a long career on generally mediocre teams. I believe that he's roughly met (or perhaps slightly exceeded) expectations given the strength of his team and length of his career (with one loss in the Stanley Cup finals, and two losses in the conference finals).

PLAYER	EXPECTED	ACTUAL	DIFFERENCE
Ed Belfour	1.11	1.00	(0.11)

[TBODY] [/TBODY]

I excluded 1989 and 1990, when he was still trying to become a regular. Belfour roughly met expectations for his career. He was expected to win 0.39 Cups in Chicago, 0.57 Cups in Dallas, 0.14 Cups in Toronto, and let's not forget the 0.02 Cups in Florida.

PLAYER	EXPECTED	ACTUAL	DIFFERENCE
Nicklas Lidstrom	2.11	4.00	1.89

[TBODY] [/TBODY]

It's hard to imagine a player who spent his career in a more favourable environment. (No, the Red Wings were never as dominant as Gretzky's Oilers, but they remained competitive much longer). Lidstrom, personally, was expected to win more Stanley Cups than any franchise, aside from the Red Wings themselves, over the course of the past 35 years - truly remarkable. Still, Lidstrom was a key reason that Detroit won nearly twice as many Cups as expected during his career.

PLAYER	EXPECTED	ACTUAL	DIFFERENCE
Raymond Bourque	1.05	1.00	0.05

[TBODY] [/TBODY]

From 1985 to 2001, Bourque essentially met expectations, in terms of the number of Stanley Cups won. If we looked at his entire career (since I only have numbers back to 1980), he'd likely fall slightly short. It's hard to imagine a better ending to someone's career (with Joe Sakic getting the greatest "assist" of his career during the Cup presentation).

PLAYER	EXPECTED	ACTUAL	DIFFERENCE
Paul Coffey	1.66	3.00	1.34

[TBODY] [/TBODY]

Substantially all of Coffey's out-performance relates to his time in Edmonton (keep in mind I'm only going back to 1985 - so his fourth Cup in 1984 is excluded here). He almost precisely met expectations the rest of the way (he won 1 Stanley Cup, versus expectations of 0.98, in Pittsburgh, LA, Detroit, Philadelphia, Carolina, and Boston). If you drop the results from 1997 onwards (when he clearly no longer was the player he used to be - which I don't think he always realized), he's at +1.62.

PLAYER	EXPECTED	ACTUAL	DIFFERENCE
Chris Chelios	2.19	3.00	0.81

[TBODY] [/TBODY]

Like Lidstrom, Chelios played on very strong teams over the course of his career (and, like Lidstrom, he was obviously an important part of their success). Chelios was +0.46 in Montreal, -0.50 in Chicago, and +0.85 in Detroit. I excluded his seven-game stint in Atlanta. If you ignore his 2008 and 2009 seasons as well (when he was getting very limited minutes in the postseason at ages 45 and 46 (!)), he almost exactly meets expectations for his career, with 2 Cups relative to expectations of 1.98.

PLAYER	EXPECTED	ACTUAL	DIFFERENCE
Chris Pronger	0.91	1.00	0.09

[TBODY] [/TBODY]

Over the course of his career, Pronger won pretty much what was expected. But I think this analysis underrates him. He would have been a perfectly valid choice for the Conn Smythe in 2006 and 2007, and maybe even in 2010 as well. The only negative I'll mention is people have largely forgotten that Pronger took heat for the Blues' disappointing performance before the 2004 lockout.

PLAYER	EXPECTED	ACTUAL	DIFFERENCE
Zdeno Chara	0.84	1.00	0.16

[TBODY] [/TBODY]

Chara is another player who would probably look better if we dug into the results of individual series. He slightly exceeded expectations for his career by winning one Cup, but he lost in the Stanley Cup finals twice, and also lost game 7 of the cofnerence finals.

PLAYER	EXPECTED	ACTUAL	DIFFERENCE
Al MacInnis	1.47	1.00	(0.47)

[TBODY] [/TBODY]

My general impression is MacInnis was a strong playoff performer, but he also played on some consistently underwhelming teams. After the winning the Cup in 1989, the Flames were really disappointing (losing in the first round rour times over the next five years, and missing the playoffs that fifth year - and this is a team that won two division titles during that span). As I mentioned with Pronger, the Blues disappointed as well. I'm not saying this is entirely (or evenly mostly) MacInnis's fault, but it should be considered.

 

[Hockey Outsider]

PLAYER ANALYSIS - FORWARDS

PLAYER	EXPECTED	ACTUAL	DIFFERENCE
Wayne Gretzky	1.58	3.00	1.42

[TBODY] [/TBODY]

Again, remember that I only have results going back to 1985. The Great One outperformed expectations, but that was entirely due to his team on the Oilers. By franchise, Gretzky was +2.06 in Edmonton (1985-1988 only), -0.40 in LA, -0.07 in St. Louis, and -0.18 in New York. I estimate that, looking at his entire career (making a ballpark estimate of 1980 to 1984), he was probably between +1.60 and +1.80.

PLAYER	EXPECTED	ACTUAL	DIFFERENCE
Mario Lemieux	0.86	2.00	1.14

[TBODY] [/TBODY]

Lemieux's teams outperformed expectations over the span of his career. But it's important to emphasize how much the strength of his supporting cast improved over time. The 1993 Penguins' Stanley Cup odds were nearly as much as the odds for the 1985, 1986, 1986, 1988, 1989 and 1990 teams combined.

PLAYER	EXPECTED	ACTUAL	DIFFERENCE
Jaromir Jagr	1.12	2.00	0.88

[TBODY] [/TBODY]

There's a lot to digest for Jagr. On the plus side, he won about one extra Stanley Cup than expected over the course of his career - that's good. On the other hand, those wins were in his first two seasons. He was a big contributor in 1992 (4th on the team in scoring), but that wasn't the case in 1991. Then again, the numbers again include everything except for his last, partial season in Calgary (at age 45). I don't know if anyone truly expected Jagr to be a difference-maker in the playoffs during his post-KHL comeback (age 39 onwards). But he generally wasn't on very good teams, so that only "cost" him -0.19 Cups relative to expectations. If you drop 1991 and everything from 2012 onwards, Jagr essentially met expectations (1 Cup versus 0.90 expected).

PLAYER	EXPECTED	ACTUAL	DIFFERENCE
Sidney Crosby	1.09	3.00	1.91

[TBODY] [/TBODY]

Crosby's results are excellent. The Penguins were only expected to win about one Stanley Cup, but earned three championships. There are some legitimate weaknesses in Crosby's playoff resume (his scoring in his four trips to the Stanley Cup finals has been disappointing), but he's almost certainly the best playoff performer of his generation.

Malkin is slightly better at +1.96 (3 wins versus 1.04 expected), since he missed the Penguins' awful 2006 campaign (when they had a very small, but non-zero, expectation of winning the Cup).

PLAYER	EXPECTED	ACTUAL	DIFFERENCE
Alexander Ovechkin	0.64	1.00	0.36

[TBODY] [/TBODY]

Ovechkin has outperformed expectations when looking at Stanley Cups won versus expectations. But I wonder how he'd look if we examine this round-by-round. The Capitals have only been out of the second round once. That's disappointing for a team with nine division titles (excluding 2020 - this article was written prior to the playoffs being completed) and three President's trophies.

PLAYER	EXPECTED	ACTUAL	DIFFERENCE
Mark Messier	1.81	5.00	3.19

[TBODY] [/TBODY]

Messier has become underrated by fans who overrate the importance of regular season stats and/or over-emphasize his debacle in Vancouver. Messier was an integral part of many teams that outperformed expectations. Keep in mind that this excludes his first five years, so his career number is probably somewhere around +3.40 to +3.60. Even if you ignore my ballpark estimate for 1980 through 1984, Messier easily has the best result of any player here.

PLAYER	EXPECTED	ACTUAL	DIFFERENCE
Joe Sakic	1.11	2.00	0.89

[TBODY] [/TBODY]

Like Lemieux, Sakic is a player who played for horrifically bad teams, and also some of the best teams in the league. The expectations were highest in 2004 (which turned out to be a hugely disappointing season) - but his team's Cup odds that year alone were nearly as high as the combined Cup odds for his 1989 through 1995 teams combined. Overall, Sakic outperformed expectations for his career by nearly a full Cup (but with much less tricky context than Jagr, who had virtually the same result). Note that I excluded his 2009 season here (damn that snowblower).

Forsberg's results are nearly identical to Sakic's. They overlap from 1995 to 2004, then we add in the Flyers in 2006 and Predators in 2007. Forsberg ends up at +0.98 (2 wins versus 1.02 expected). This assumes we give him full credit for 2001, where he missed the final two rounds.

PLAYER	EXPECTED	ACTUAL	DIFFERENCE
Steve Yzerman	1.83	3.00	1.17

[TBODY] [/TBODY]

We all know the story about Yzerman. He started his career on the awful "Dead Things". Detroit steadily improved, and Yzerman got the reputation of being choker (especially after they lost in 1996 after a 131 point season). Then Detroit won three of the next six Stanley Cups, Yzerman won a Conn Smythe, and he was nearly universally revered as a playoff performer and captain. I think there's some validity to the criticism up to 1996 - the Red Wings probably should have won a Cup by then (0.70 expected). But, as good as the Red Wings were, three Cups in the next six years vastly exceeded expectations.

Fedorov's results are better than Yzerman's. They overlap from 1991 to 2003. Then Fedorov spent five years on some teams that generally had low expectations. He ends up at +1.44 (3 wins versus 1.56 expected) - because him not winning in Anaheim and Columbus came at a lower "cost" than Yzerman spending a few more years not winning in Detroit.

PLAYER	EXPECTED	ACTUAL	DIFFERENCE
Patrick Kane	0.74	3.00	2.26

[TBODY] [/TBODY]

An excellent result - better than Crosby, and actually better than Messier if you're looking at this on a percentage basis. A remarkably consistent performer in the spring.

PLAYER	EXPECTED	ACTUAL	DIFFERENCE
Teemu Selanne	0.77	1.00	0.23

[TBODY] [/TBODY]

Selanne played on some bad teams for most his prime - worse, I think, than most people remember. From 1993 to 2001 (nine years), his teams were expected to win just 0.17 Stanley Cups. He finally won a championship during his second stint with the Ducks. He did play on some strong teams that disappointed in the playoffs though - 2004 really jumps out (though he was badly struggling with a knee injury - the 2005 lockout saved his career).

PLAYER	EXPECTED	ACTUAL	DIFFERENCE
Brett Hull	1.16	2.00	0.84

[TBODY] [/TBODY]

I'm looking at Hull's career from 1988 to 2004. Like Pronger and MacInnis, he was a significant contributor to the underwhelming Blues for many years (-0.40), but he redeemed himself in Dallas (+0.62) and Detroit (+0.62). Hull remains the only player in NHL history with 90+ playoff goals who didn't play on Gretzky's Oilers.

PLAYER	EXPECTED	ACTUAL	DIFFERENCE
Joe Thornton	0.89	-	(0.89)

[TBODY] [/TBODY]

Finally, let's look at a player who has a reputation for being disappointing in the playoffs. Thornton has had a long career on generally strong teams. They were rarely among the top three or five contenders, but by playing on teams that were this competitive, for this long, Thornton probably should have won a Cup by now. As I've said a few times, it's probably not appropriate to ascribe all of a team's successes and failures to one player, but there's no denying that Thornton's scoring has dropped significantly in the postseason.

 

[plusandminus]

PLAYER ANALYSIS - IN PROGRESS - FORWARDS

PLAYER	EXPECTED	ACTUAL	DIFFERENCE
Wayne Gretzky	1.58	3.00	1.42

[TBODY] [/TBODY]

Again, remember that I only have results going back to 1985. The Great One outperformed expectations, but that was entirely due to his team on the Oilers. By franchise, Gretzky was +2.06 in Edmonton (1985-1988 only), -0.40 in LA, -0.07 in St. Louis, and -0.18 in New York. I estimate that, looking at his entire career, he was probably between +1.60 and +1.80.

PLAYER	EXPECTED	ACTUAL	DIFFERENCE
Mario Lemieux	0.86	2.00	1.14

[TBODY] [/TBODY]

Lemieux's teams outperformed expectations over the span of his career. But it's important to emphasize how much the strength of his supporting cast improved over time. The 1993 Penguins' Stanley Cup odds were nearly as much as the odds for the 1985, 1986, 1986, 1988, 1989 and 1990 teams combined.

As usual, you contribute great posts and stats.

However, in this case, I wonder if the stats doesn't tend to be a bit "over analyzed".
Also, as you write in your threadstarter, trades during season isn't considered. I'm no trades expert, but on at least one occasion Pittsburgh made some trades during season which increased their chances of winning the Stanley Cup. That would alter the already vulnerable stats even more.
So to me, this is more like a curiosity thread that I don't think we can make a lot of conclusions from.

But if you (or someone else) would focus on regular season win percentages vs Stanley Cup success, I would be more interested.
Or if we focused on odds from between end of regular season and start of the playoffs.

 

[Hockey Outsider]

SUMMARY TABLE

Player	Expected	Actual	Difference	Percentage
Messier	1.81	5	3.19	176%
Roy	1.59	4	2.41	152%
Kane	0.74	3	2.26	305%
Malkin	1.04	3	1.96	188%
Crosby	1.09	3	1.91	175%
Lidstrom	2.11	4	1.89	90%
Brodeur	1.15	3	1.85	161%
Fedorov	1.56	3	1.44	92%
Gretzky	1.58	3	1.42	90%
Coffey	1.66	3	1.34	81%
Yzerman	1.83	3	1.17	64%
Lemieux	0.86	2	1.14	133%
Forsberg	1.02	2	0.98	96%
Sakic	1.11	2	0.89	80%
Jagr	1.12	2	0.88	79%
Hull	1.16	2	0.84	72%
Chelios	2.19	3	0.81	37%
Hasek	0.56	1	0.44	79%
Ovechkin	0.64	1	0.36	56%
Selanne	0.77	1	0.23	30%
Chara	0.84	1	0.16	19%
Pronger	0.91	1	0.09	10%
Bourque	1.05	1	-0.05	-5%
Belfour	1.11	1	-0.11	-10%
MacInnis	1.47	1	-0.47	-32%
Lundqvist	0.48	0	-0.48	-100%
Thornton	0.89	0	-0.89	-100%

[TBODY] [/TBODY]

I was debating whether to post a summary table. There's a lot of context that should be considered, and I don't want people to glance at this chart and walk away with any definitive conclusions. But I think, on balance, this is more helpful than harmful. I think people are smart enough here not to argue that Sakic is better than Jagr because 0.89 > 0.88.

I can post specific players if requested.

 

[tarheelhockey]

There are a lot of interesting stats here.

Perhaps the most depressing, among several good candidates, is Arizona having an Expected Cups of 0.05

 

[The Panther]

TOP FIFTEEN PRE-SEASON STANLEY CUP FAVOURITES

Rank	Year	Team	Odds	Result
1	1988	Edmonton Oilers	25.9%	Won Stanley Cup

[TBODY] [/TBODY]

As you note, a lot of things can change from pre-season to spring. Regarding the 1987-88 Oilers, I would guess their actual expectation to win went something like this:
Sept. 1987 - 25.9%
April 1988 - 10%

The only noisy media person I recall predicting they would win was Don Cherry, to give him his credit.

 

[Hockey Outsider]

As usual, you contribute great posts and stats.

However, in this case, I wonder if the stats doesn't tend to be a bit "over analyzed".
Also, as you write in your threadstarter, trades during season isn't considered. I'm no trades expert, but on at least one occasion Pittsburgh made some trades during season which increased their chances of winning the Stanley Cup. That would alter the already vulnerable stats even more.
So to me, this is more like a curiosity thread that I don't think we can make a lot of conclusions from.

But if you (or someone else) would focus on regular season win percentages vs Stanley Cup success, I would be more interested.
Or if we focused on odds from between end of regular season and start of the playoffs.

Those are fair/valid comments. I'd imagine that any variations would average out over the course of a 15-25 year career (ie some years a team struggles with injuries and their preseason odds overrate their chances, while other years a team makes some good trades and their preseason odds underrate their chances).

So, I wouldn't argue that one team is better than another because one had a 10% chance of winning the Cup before the start of the year, while another only had a 5% chance. There's too much variation. But I think it's probably appropriate to say that someone who was expected to with around 0.8 Cups over the course of their career (ie Selanne) played in an environment less conducive to winning than someone who was expected to win 1.5 Cups (ie Fedorov). That doesn't, of course, mean that Selanne is better than Fedorov (who's clearly the better playoff performer) - but it allows us to have more depth to the conversation than 3 > 1.

Looking at the numbers (in terms of expected Cups, and the difference between actual and expectations), the results generally look "about right" to me. I've tried to provide some wording about any relevant context. But if there's anything that looks suspicious/wrong, let me know.

(The other point I'll make is - this type of analysis probably only works with star players in crucial roles. Chris Kunitz's result is probably something like +2.5 over the course of his career. Obviously he's not anywhere close to Roy as a playoff performer - so I've focused on a fairly small group of star players so that depth players aren't getting credit for the results driven by the team's goalie and top three or four skaters).

 

[ted2019]

Where can I find the thread that overpasses observation is in?

 

[Kyle McMahon]

Purpose - This thread is inspired by a recent observation from @overpass - namely, that we can assess how many Stanley Cups a player won, relative to the expectations of his team's strength, by taking into account the pre-season odds.

Method - I took the pre-season Stanley Cup odds from hockey-reference.com. They have a complete listing going back to 1985. I converted the odds each year into percentages. I adjusted the percentages so that they sum to 100% each year. (Without this adjustment, the percentages would add to over 100%, as there's a profit element built in. Interestingly, the profit element has decreased steadily over time - perhaps this is due to more online competition that drives the spreads downwards - I don't gamble on sports so maybe someone who's done that can confirm if that's correct).

As I usually try to do, I'll offer some self-criticisms of this approach:

• This is a fairly crude analysis. It only looks at whether a team wins the Stanley Cup or not. It doesn't take into account whether they lost in game 7 of the Stanley Cup finals, or if they were swept in the first round. An obvious way to improve this analysis would be to get the odds series by series.
• Although I do some individual analysis, it's questionable if it's truly appropriate to apply this to individual players. It's likely not appropriate to ascribe all of the differences to the players I study (i.e., a team might lose due to poor goaltending, which likely isn't the top scorer's fault).
• The analysis is based upon pre-season odds. It would be more instructive to use odds right before the playoffs began, as that would take into account significant injuries, mid-year trades, etc.
• The purpose of gambling organizations is to maximize their profit (not necessarily to assess the strength of teams). To the extent that there's an efficient market, these would give similar evaluations. If you believe that gamblers exploit particularly optimistic fanbases, the results may be skewed.

To the bolded, I'd say this is almost certainly the case. Until pretty recent history, you had to physically go to Las Vegas or Atlantic City to place a bet. Obviously black market bookies were everywhere, but I don't think there was much of a market for futures bets (are you going to leave a substantial amount of money with a back alley bookie for several months?)

The odds data from the last 15 years or so is likely more informed than from before that time period, probably significantly so. There are all sorts of advanced predictive models being built by gamblers with the computer programming know-how to sort the mountains of data that are now available. Of course, Vegas has all this info too. There's little doubt in my mind that market efficiency (accuracy of betting lines) has increased over the last 10 to 15 years.

That said, gambling becoming so easily accessible online and so mainstream in the last ten years means there are now dumptrucks full of square money being pumped into the market by Joe Six Pack who just wants to bet on his favourite team. This will depress the odds for popular teams to an extent (as you say, taking advantage of over-optimistic fan bases).

A good example of this was the Maple Leafs this past pre-season, who tended to be listed as around the third or fourth favourite to win the 2020 Stanley Cup, which wasn't really indicative of reality. Even if you felt Toronto was indeed the 3rd-best team in the NHL, Tampa Bay and Boston (1st and 2nd according to those same odds) both standing in their way should have logically skewed their odds upwards.
Recent editions of the Blackhawks have tended to be consistently over-valued as well, as a popular team living off their reputation rather than actual recent results.

The biggest surprise to me from your study, which is fantastic as usual, was Crosby-era Pittsburgh's somewhat low expected Cups number (or at least, lower than I anticipated). Evidently their pre-season expectations have been consistently lower than their pre-playoff expectations for some reason. Going by memory, Pittsburgh was either a favourite or close second-favourite to come out of the East at the start of the playoffs in most seasons since 2009 (except maybe 2015, and 2011 due to injury-related reasons).

 

[Theokritos]

Where can I find the thread that overpasses observation is in?

Pierre Turgeon in the playoffs, specifically this post.

 

[ted2019]

Pierre Turgeon in the playoffs, specifically this post.

Thank for very much. I always have lots of interest in Overpass posts/threads.

 

[MadLuke]

The biggest surprise to me from your study, which is fantastic as usual, was Crosby-era Pittsburgh's somewhat low expected Cups number (or at least, lower than I anticipated).

It is hard to judge, I would say being so much bigger than the Caps (the only exact comparable I could see above) was a bit surprising to me, 1.09 vs .64, considering how good they have been during that time frame.

I imagine once you start winning you have that ability to get bets even with giving little in returns relative to the actual team year-year quality/performance, a bit like the Hawks.

When you are Wayne Gretzky, you probably skew cup expectation by yourself quite a bit (specially if you play in popular better markets), making the data quite different than how many cup a regular NHLer in their situation would have likely won.

 

[Kyle McMahon]

It is hard to judge, I would say being so much bigger than the Caps (the only exact comparable I could see above) was a bit surprising to me, 1.09 vs .64, considering how good they have been during that time frame.

I imagine once you start winning you have that ability to get bets even with giving little in returns relative to the actual team year-year quality/performance, a bit like the Hawks.

When you are Wayne Gretzky, you probably skew cup expectation by yourself quite a bit (specially if you play in popular better markets), making the data quite different than how many cup a regular NHLer in their situation would have likely won.

You're correct on both those points, which is why I figured Crosby's expectation would be a little higher. Once the Penguins reached the final in 2008 and especially after they won in 2009, they had reputation going for them. And with Crosby, they also had the mega-star effect which will attract casual bettors as well.

I suppose a 1.09 expected Cups doesn't sound as high as it truly is mathematically though. Pittsburgh was probably averaging around 0.08 expected Cups per year from 2009 until now. 1 in 12 odds are still pretty short, I guess in my mind I figured they'd have been around 1 in 10. Not an insignificant difference, but not huge either.

 

[Hockey Outsider]

I'll throw the bat signal out to @pnep (or anyone else who might have this) -

Is there any source that has historical series-by-series betting odds? For example, if I wanted to look up what the implied odds were for Detroit vs Anaheim (first round series in 2003) - where could I find that?

 

[pnep]

I'll throw the bat signal out to @pnep (or anyone else who might have this) -

Is there any source that has historical series-by-series betting odds? For example, if I wanted to look up what the implied odds were for Detroit vs Anaheim (first round series in 2003) - where could I find that?

NHL playoffs series |SportsOddsHistory.com

?

 

[Hockey Outsider]

NHL playoffs series |SportsOddsHistory.com

?

This is perfect - thanks.

 

[LightningStorm]

Interesting stuff as always, @Hockey Outsider. After seeing your 2 top 15 lists for Stanley Cup favorites and underdogs, I'm suddenly curious about something: who are the top 15 underdogs to win the Stanley Cup?