Show me the parents of a good young hockey player & I'll tell you if pro hockey could be in their future-
Witnessed a hockey dad angrily berate & get physical with his ~10 yr old son last night at spring hockey game. Dad didn't realize dressing room was connected to ours with doors open (have reported & will follow up)
In my 20+ year career as NHL agent, I've seen WAY more kids ruined by over zealous parents alone than bad coaching, not enough ice time & all other reasons imaginable combined
Don't try to coach them, dont try to be their agent, don't push them, just be their supportive Mom/Dad & let them have fun. Otherwise you'll be the reason they don't reach their full potential
Viewers of the Pens/Flyers series have questions. This metric is a great idea but it's completely useless with the data currently available for it.
In last nights game, this model assigned an expected goal value of .04% on the first Flyers goal and .19% for the second. Thats saying an average NHL goalie stops the first goal around 24 of 25 times and the 2nd goal around 4 of 5 times. Now go look at those goals with that in mind. For that matter, look at all the shots Skinner faced in the 2 games and compare with this chart. You will have questions, as he's been excellent. But this paints a very different and inaccurate picture
This is junk but many people think it has value in assessing goalie play. It could have value but only with better data, so let's work towards getting that before we use incomplete data to improperly assess NHL talent
@Versa_Jae Sure, but it's also somewhat irresponsible to post lists of "best goalies" and "worst goalies" using the current metrics available. It's not complete enough to make those assertions
Screens, own deflections, shooter time and space, and puck trajectory (where is it going in the net) are all not tracked (by most models). Those items combined are way more important than what is currently tracked. Those aren't easy to quantify which is why it hasn't been done. That's also why it's not complete and accurate
So what’s the value in tracking this if many smart people can’t reach a consensus? We only have limited data for this right now, why not wait until we have more and better data, rather than relying on such a small data sample and then publishing “best” and “worst” goalie lists based on an incomplete picture of shot quality? It risks undermining the credibility of the analytics community. Being early doesn’t matter as much as being right, accurate and complete.
Labeling a goal as “good” or “bad” without accounting for shot placement (where's the puck going toward the net, not just location where it was released) relative to the goalie’s position and movement, and without considering factors like screens, own deflections, time, back pressure or broken plays renders the process largely ineffective. The fact that different models produce widely varying results only reinforces that point.
Currently, the assigned shot/save value is literally the same for a double screen shot deflected 4 times by a goalie's teammates that goes post and in as a shot from the same location that is unscreened and hits the goalie in the chest. That is ridiculous and you can't change my mind.
By the way, I value goalie analytics, I was the first NHL agent to use advanced stats in contract negotiations for goalies. But let's get it right
@RayPetkau@GavalierR The performance of private vs public models has been tracked. Over the course of a season they disagree with each other as much as they disagree with MoneyPuck.
I agree that sample size is extremely important for most metrics. And for sure private models are getting closer.
But the way this is tracked for public models is still mostly useless even in large quantities because shot placement is not tracked in any way, even paid models. Neither are own deflections and screens. A post and in shot/goal has the same value in these models as a shot right to the goalies chest. One would be a great save, and the other is impossible to let in. Yet same assigned shot value based on shot release.... that makes it completely misleading and inaccurate
James Reimer posted his 32nd career shutout and helped the @Senators move within three points of the idle Bruins for the second Wild Card spot in the Eastern Conference.
#NHLStats: https://t.co/44107zswJy
Connor Hellebuyck achieved one lifelong dream, and is ready to chase another 🥇🏆
Hear from the Olympic Gold medalist on the bonds he built with fellow Olympic athletes, bringing a Stanley Cup to Winnipeg, and much more.
While a lot of the focus appears to be on which goalie starts for Edmonton, curious if Oilers can stop bleeding defensively in Game 2 after giving up more high % chances (14) and highest XG against (4.69) of any playoff game so far in Game 1. All 6 goals were on high % chances
Love him or hate him, Donald Trump grew the game of hockey today, tremendously and biggly😂
No amount of money could buy that kind of publicity for the NHL