[GDT] Kraken @ Flames 7pm MT

[topfiverecords]

Originally Posted by Bingo

Not sure pointing out they average 7 shoots outs a season qualifies as a data review.

Some serious touchiness on stats these days.

Let's say you're chatting with a friend about the game and say something like "I'd probably put someone like Kerins in the lineup over Kirkland"

and they jokingly say "But we need Kirkland for the shootout though".

and then you respond with "Do we though? Hold on a sec I'm gonna look up a bunch of stats about that."

It's not touchiness.

[Hockey-and_stuff]

Originally Posted by Macho0978

Here is my theory why there is a disconnect between the stats and the people who watched the game and thinking Bahl and Parekh played well.



Seattle did not dress their best lineup. The one player/line that stands out from their lineup last night is Catton or Catton, Tolvanen and Marchment line.



Bahl and Parekh played the most minutes against Catton, they played very well against Catton and he is the guy that people would say is the biggest threat from last night's line up to score against the Flames.



Bahl and Parekh struggled more against the no names dressed last night



If you watch the game and you are out against some no name plug or AHL plug for a scoring chance against, it probably gets disregarded more than if a star player gets open. The plug likely did nothing with the chance.

Honestly I'm not even sure what is being counted on these graphs. But I'll also admit I'm not always paying attention to or remembering every play each player makes. Sometimes I'm trying to watch someone else.

To my eyes, Parekh seemed ok defensively (he had a couple bad reads) and great offensively. Bahl to my eyes had a very rough outing.

Ultimately, exactly one player ended the night with a negative game score (Bahl) which is usually indicative of the entire team having a great night, and maybe the plays Bahl and Parekh had didn't have as much of an impact relative to the rest of the team.

Just my thoughts on how I'm interpreting the graphic. Pretty fun to break it down in my head!

Sent from my SM-S931W using Tapatalk

[Goriders]

Originally Posted by Macho0978

Been saying it all summer that he needs to go before the season starts.

Brzustewicz looking good so far and could push to make the team.

I still think Kuznetsov deserves a crack.

We can sit Bean, Pachal or Hanley but my concern is Ras ice should dip from his 24 a night. His value isn't going up IMO.

But I have been told “it’s not a distraction”

[Goriders]

I watched the last two games Parekh was involved in.

I have a feeling he might be an offensive defenseman.

This is not confirmed yet. So don’t hold me to it.

[topfiverecords]

Originally Posted by Goriders

I watched the last two games Parekh was involved in.

I have a feeling he might be an offensive defenseman.

This is not confirmed yet. So don’t hold me to it.

107 pts last year in 61 games and he has scored in 100% of his NHL regular season games so far.

[NegativeSpace]

Originally Posted by topfiverecords

107 pts last year in 61 games and he has scored in 100% of his NHL regular season games so far.

Yeah, those are just stats - what do your eyes tell you?

[Enoch Root]

Actually, some stats ARE garbage. Some are flawed in their design. Others (and the ones in this discussion fall into this category) suffer terribly with small sample sizes.

Just because something is quantified, that doesn't automatically make it valid.

[PepsiFree]

Originally Posted by Enoch Root

Actually, some stats ARE garbage. Some are flawed in their design. Others (and the ones in this discussion fall into this category) suffer terribly with small sample sizes.

Just because something is quantified, that doesn't automatically make it valid.

OK, if these stats are garbage then provide some deeper insight beyond “I seen it” to make your case, because that’d be worth discussing.

Are they invalid logically? or practically? How did you come to that conclusion? How do they “suffer terribly” with small sample sizes and what sample size would be more acceptable to measure what happens in a game? Hell, what are the specific “stats like these” you’re referring to?

I’m no expert myself but I think this stuff is pretty interesting. On the other hand, an argument to dismiss stats without any concrete/objective reason why is just incredibly boring.

[Enoch Root]

Originally Posted by PepsiFree

OK, if these stats are garbage then provide some deeper insight beyond “I seen it” to make your case, because that’d be worth discussing.

Are they invalid logically? or practically? How did you come to that conclusion? How do they “suffer terribly” with small sample sizes and what sample size would be more acceptable to measure what happens in a game? Hell, what are the specific “stats like these” you’re referring to?

I’m no expert myself but I think this stuff is pretty interesting. On the other hand, an argument to dismiss stats without any concrete/objective reason why is just incredibly boring.

I have outlined many times some of the problems I think these stats face. And people shouldn't need to be told that small sample sizes can produce wildly unrepresentative data - I would hope everyone knows that.

Also, I am not going to write - and no one would read - a thorough explanation of what can be wrong with them. If you genuinely want to learn more (which I doubt), I am confident that you know how to use the internet to go about it.

[PepsiFree]

Originally Posted by Enoch Root

I have outlined many times some of the problems I think these stats face. And people shouldn't need to be told that small sample sizes can produce wildly unrepresentative data - I would hope everyone knows that.

Also, I am not going to write - and no one would read - a thorough explanation of what can be wrong with them. If you genuinely want to learn more (which I doubt), I am confident that you know how to use the internet to go about it.

When you refer to “these stats,” I’m curious which ones you’re specifically referring to. Can you be more specific?

As far as sample size goes, you actually do need to tell people why stats taken from a single game are actually unrepresentative of what happened in that single game, since everyone actually knows the opposite to be true. I’m interested to see your theory here.

And, to be clear, I’m not asking for a thorough explanation of what can be wrong with them that I could google. I’m asking you for your opinion on what you think is wrong with the stats behind that card. I assume it doesn’t take more than a few sentences to provide an explanation beyond “they’re garbage” or “some stats are invalid.” Doesn’t need to be a novel. If you can’t or haven’t actually thought about it and you’re just doing the bluster thing that’s OK.

[Enoch Root]

Post 281

And no, I don't need to do anything.

[PepsiFree]

Originally Posted by Enoch Root

Post 281

And no, I don't need to do anything.

So you don’t actually have an explanation for why single game stats are unrepresentative of a single game?

Boring. Oh well.

[Enoch Root]

Originally Posted by PepsiFree

So you don’t actually have an explanation for why single game stats are unrepresentative of a single game?

Boring. Oh well.

That's the response I would expect from a 10 year old.

Your ongoing attempts to score points against me are weird and kind of creepy. More importantly, they are boring as #### - no one cares. So do CP a favour and put me on ignore.

[PepsiFree]

Originally Posted by Enoch Root

That's the response I would expect from a 10 year old.

Your ongoing attempts to score points against me are weird and kind of creepy. More importantly, they are boring as #### - no one cares. So do CP a favour and put me on ignore.

Sorry, what else am I supposed to gather from your response? That you do have an explanation but you’re not telling because it’s a secret and so complex and uninteresting that nobody could possibly be interested even though you’re literally being asked for that explanation? lol

Asking you questions about your opinion is not attempting to score points against you. If you don’t have answers and don’t know or haven’t thought about things that’s fine, we’re just a bunch of hockey fans shooting the ####. I literally do not know how a stats from a single game cannot be representative of that single game. That’s why I asked.

But playing super coy about your thoughts and labelling questions about them “weird” or “creepy” or pretending to speak for the whole board in an attempt to shut it down is not a normal way to have a conversation, Enoch. If you’re going to act that way, why bother participate at all? Who is that for?

I say things all the time I haven’t thought through all the way. It’s OK to just go on vibes and be wrong or not know stuff. It’s not like there’s a job waiting for the best pretend hockey expert on CP, so just relax once in a while. No need to put on a show to impress anybody.

[Canada 02]

Wrong thread

[Reggie Dunlop]

Good game.

[Bingo]

Originally Posted by topfiverecords

Let's say you're chatting with a friend about the game and say something like "I'd probably put someone like Kerins in the lineup over Kirkland"

and they jokingly say "But we need Kirkland for the shootout though".

and then you respond with "Do we though? Hold on a sec I'm gonna look up a bunch of stats about that."

It's not touchiness.

"bunch of stats ..."

Lets check the list

1. Number of shootouts per season

Maybe green text then? It's not like we don't have that established.

Works well because then I don't think you actually mean that he should make the team because of shoot outs so I don't feel the need to point out that not many games make it to a shoot out.

[GirlySports]

Originally Posted by Enoch Root

Actually, some stats ARE garbage. Some are flawed in their design. Others (and the ones in this discussion fall into this category) suffer terribly with small sample sizes.

Just because something is quantified, that doesn't automatically make it valid.

Stats are like mini-skirts. They give you some ideas but hide the important things.

[Enoch Root]

Originally Posted by PepsiFree

Sorry, what else am I supposed to gather from your response? That you do have an explanation but you’re not telling because it’s a secret and so complex and uninteresting that nobody could possibly be interested even though you’re literally being asked for that explanation? lol

Asking you questions about your opinion is not attempting to score points against you. If you don’t have answers and don’t know or haven’t thought about things that’s fine, we’re just a bunch of hockey fans shooting the ####. I literally do not know how a stats from a single game cannot be representative of that single game. That’s why I asked.

But playing super coy about your thoughts and labelling questions about them “weird” or “creepy” or pretending to speak for the whole board in an attempt to shut it down is not a normal way to have a conversation, Enoch. If you’re going to act that way, why bother participate at all? Who is that for?

I say things all the time I haven’t thought through all the way. It’s OK to just go on vibes and be wrong or not know stuff. It’s not like there’s a job waiting for the best pretend hockey expert on CP, so just relax once in a while. No need to put on a show to impress anybody.

Wow. This is all in your head. I am not being coy, or pretending anything, or trying to have secrets. I don't get upset over comments on CP (as you have tried to imply at other times). And I'm not acting in any way or trying to put on a show - I simply commented on some stats because I had an opinion to share. And I don't owe you or anyone else an explanation for my comments.

If you in fact don't know how small sample sizes can be wildly misrepresentative, and you are interested, I would suggest you look into it. Stats and statistical analysis are, IMO, an incredibly interesting and useful field (though I fully understand that most people don't agree, LOL). And as I have said on here a hundred times, most people do not know how to read and interpret them, never mind understanding how dependent models are on the inputs chosen, and how useless the results actually are, if the inputs are chosen poorly (and the problem with hockey is that events are all muddied together and players are co-dependent on each other, so events can't be fully isolated (like baseball for example), and thus most of the inputs are pretty useless, unfortunately).

So learning a little more about it can be interesting and helpful.

But I have also found that bringing these things up is completely pointless and frustrating, and the rebuttals usually involve comments about eye tests and personal bias (which have some validity, but also completely miss the point), while the next post will re-quote the same stat again, and make the same mistake with its interpretation.

So I comment and I move on.

[Bingo]

Stats are an important tool to help eliminate bias in eye test.

Sample size is key for sure.

Also context ... stat selection.

What kind of player is it? If it's a fourth line player don't compare xGF% because they don't generate much xGF60 in their role. But how do they compare in xGA60 against their teammates knowing they have a sheltered role? How do they look against other fourth liners?

With the right context a stat isn't wrong. It's just important or not important.

Can you use one game sample sizes? Of course you can. If player x was on the ice for 8 HD chances against and no other player was on the ice for more than 2 it was likely a bad game.

The nature of the stat doesn't pin it on the single player though. So that's why longer data sets make more sense. If he's on the ice for more HD against over 40 games than any other player? Probably an issue.

We should try and think these things through, and hopefully land on something more constructive than "the stats are BS, I'll trust my eye test thanks!"