All you need to do is take a quick look at Moneypuck’s power rankings to realize their model is a hot dump of shit. They got us all the way down at #15 with the fucking Wild ahead of us in 7th place. I can’t even begin to try and explain that one.
Moral of the story is just use them for individual advanced stats and apply context to them yourself. Don’t use their models for determining team power rankings, playoff odds, etc. The stats themselves are good and fairly reliable. It’s when they try combining a bunch of their stuff to create their models that shit starts getting out of whack.
aLostBattlefield
I really couldn’t care less about betting or odds.
Never-serious
I think this one of those times you can bet on your favorite team and make bank.
esp211
Moneypucks is very flawed. It does not account for defense or goaltending, which are 2/3 of the equation.
kvnklly
I think its because our points come a lot from special teams, and technically it should not be sustainable
InevitableHome343
If it means we’ll get less penisey, I’m here for it
kmisca92
Moneypuck’s model is (garbage) weighted off recent play. Last night the currently not in the playoffs NJD had a 13.8% chance to win the cup. This morning they have a 7.%. NYR last night was 3% and today is 6.9%. Imagine thinking 1 game in December could change increase/decrease your odds by 50%.
Eatabulletnow
The only thing that matters, as of right now the sports books have the Rangers as favs to win the cup (along with Colorado).
TooManySaws
I really don’t get why people pay any attention to Moneyfuck.. they’re terrible. I wish people would stop posting/linking to/supporting them.
SteveyPugs2020
I read them today and was just disgusted with there bullshit.
NYM32
let’s look at basic team stats for a second. Here are NYR’s 5v5 stats, league-wide ranked:
71 goals (22nd among teams in NHL)
2.5 goals per 60 (20th)
70 goals against (16th)
2.46 goals against per 60 (15th)
838 shots (18th)
816 shots against (18th)
they’re in the bottom half of NHL teams in even the most basic 5v5 statistics — we haven’t even gotten into advanced analytics. NYR has exactly 1 line doing anything 5v5, and only 1 player in NYR’s entire bottom 6 has at least 10 points. they are a +1 in goal differential at 5v5, and 5v5 is literally 80% of their TOI. why else do you think MoneyPuck, or *any* model for that matter, has NYR as a mediocre team at 5v5?
the team-wide advanced analytics don’t factor in goaltending because they’re derived solely from what the skaters in front of them are giving up/contributing on the offensive end. this makes sense anyway, because a goaltender standing on his head on his way to a win doesn’t mean that the *team* is playing well — could’ve sworn we just spent 15 years watching Lundqvist do that and they not only didn’t win a cup but only made the finals once
it’s fantastic that NYR is in first place (albeit we’re not even halfway through the season), but models don’t use winning or losing games to calculate if the team is playing well. The only model that ranks teams by wins and points is the…standings. it’s worth mentioning this sub has this exact same conversation every single year about how the team is winning games, the models say that they aren’t playing well, people then claiming the models are bad, and then NYR ultimately loses in the playoffs due to not playing well (I also look forward to someone trying to tell me that i’m not a fan/want them to lose, solely for pointing out this basic information as the team is in 1st place). it’s also worth mentioning that MoneyPuck right now lists NYR as controlling play at 5v5, which is not only better than what other models have them at, but it’s better than they did last year, *and* MoneyPuck uses rolling averages in their calculations, so odds get better/worse as the season progresses.
it’s just amazing to me that NYR can literally be middle of the pack in the most basic stats imaginable, and then people are upset that the models show that NYR is…middle of the pack. whether you think the models are good or bad, it doesn’t take much to put 2 and 2 together as to why the models are showing what they show.
ExperienceNo7751
I’ll spoil all of hockey stats right here, and I’ll probably delete this in a few days because it flies in the face of a bunch of good people’s jobs.
Advanced stats did not revolutionize the Premier Leagues or more specifically Chelsea FC. They spent hundreds of millions. You know what they said behind closed doors when asked why their analysis hadn’t produced anything close to their forecasting?
They said it was because of the inaccuracy of reporting data. The reason for that is unlike baseball, and to a lesser extent NFL, there is an inherent lack of consistency in repeatable results. Even with puck and player tracking, the differences in game-condition is so wide, the data becomes too narrow in sample size once filtered to a reasonable degree.
That said, individual player performance can be judged against their team mates, quite effectively and at a low cost. DM me.
SufficientQuiet130
I looked yesterday, Oilers were a 6.7% chance to win cup, rangers 3%. Lmaooooo
14 Comments
Reminds me of EPA in the NFL.
All you need to do is take a quick look at Moneypuck’s power rankings to realize their model is a hot dump of shit. They got us all the way down at #15 with the fucking Wild ahead of us in 7th place. I can’t even begin to try and explain that one.
Moral of the story is just use them for individual advanced stats and apply context to them yourself. Don’t use their models for determining team power rankings, playoff odds, etc. The stats themselves are good and fairly reliable. It’s when they try combining a bunch of their stuff to create their models that shit starts getting out of whack.
I really couldn’t care less about betting or odds.
I think this one of those times you can bet on your favorite team and make bank.
Moneypucks is very flawed. It does not account for defense or goaltending, which are 2/3 of the equation.
I think its because our points come a lot from special teams, and technically it should not be sustainable
If it means we’ll get less penisey, I’m here for it
Moneypuck’s model is (garbage) weighted off recent play. Last night the currently not in the playoffs NJD had a 13.8% chance to win the cup. This morning they have a 7.%. NYR last night was 3% and today is 6.9%. Imagine thinking 1 game in December could change increase/decrease your odds by 50%.
The only thing that matters, as of right now the sports books have the Rangers as favs to win the cup (along with Colorado).
I really don’t get why people pay any attention to Moneyfuck.. they’re terrible. I wish people would stop posting/linking to/supporting them.
I read them today and was just disgusted with there bullshit.
let’s look at basic team stats for a second. Here are NYR’s 5v5 stats, league-wide ranked:
71 goals (22nd among teams in NHL)
2.5 goals per 60 (20th)
70 goals against (16th)
2.46 goals against per 60 (15th)
838 shots (18th)
816 shots against (18th)
they’re in the bottom half of NHL teams in even the most basic 5v5 statistics — we haven’t even gotten into advanced analytics. NYR has exactly 1 line doing anything 5v5, and only 1 player in NYR’s entire bottom 6 has at least 10 points. they are a +1 in goal differential at 5v5, and 5v5 is literally 80% of their TOI. why else do you think MoneyPuck, or *any* model for that matter, has NYR as a mediocre team at 5v5?
the team-wide advanced analytics don’t factor in goaltending because they’re derived solely from what the skaters in front of them are giving up/contributing on the offensive end. this makes sense anyway, because a goaltender standing on his head on his way to a win doesn’t mean that the *team* is playing well — could’ve sworn we just spent 15 years watching Lundqvist do that and they not only didn’t win a cup but only made the finals once
it’s fantastic that NYR is in first place (albeit we’re not even halfway through the season), but models don’t use winning or losing games to calculate if the team is playing well. The only model that ranks teams by wins and points is the…standings. it’s worth mentioning this sub has this exact same conversation every single year about how the team is winning games, the models say that they aren’t playing well, people then claiming the models are bad, and then NYR ultimately loses in the playoffs due to not playing well (I also look forward to someone trying to tell me that i’m not a fan/want them to lose, solely for pointing out this basic information as the team is in 1st place). it’s also worth mentioning that MoneyPuck right now lists NYR as controlling play at 5v5, which is not only better than what other models have them at, but it’s better than they did last year, *and* MoneyPuck uses rolling averages in their calculations, so odds get better/worse as the season progresses.
it’s just amazing to me that NYR can literally be middle of the pack in the most basic stats imaginable, and then people are upset that the models show that NYR is…middle of the pack. whether you think the models are good or bad, it doesn’t take much to put 2 and 2 together as to why the models are showing what they show.
I’ll spoil all of hockey stats right here, and I’ll probably delete this in a few days because it flies in the face of a bunch of good people’s jobs.
Advanced stats did not revolutionize the Premier Leagues or more specifically Chelsea FC. They spent hundreds of millions. You know what they said behind closed doors when asked why their analysis hadn’t produced anything close to their forecasting?
They said it was because of the inaccuracy of reporting data. The reason for that is unlike baseball, and to a lesser extent NFL, there is an inherent lack of consistency in repeatable results. Even with puck and player tracking, the differences in game-condition is so wide, the data becomes too narrow in sample size once filtered to a reasonable degree.
That said, individual player performance can be judged against their team mates, quite effectively and at a low cost. DM me.
I looked yesterday, Oilers were a 6.7% chance to win cup, rangers 3%. Lmaooooo