Speculation: Jordan Binnington to Toronto?

[Shane Diesel]

So, my full sample of playoff games is deceptive, but your smaller portion of that sample isn't? Who is the one that is deceptively cherry picking stats to try and make their argument?

What are you talking about? Binnington has eight games of play against the Avs in the playoffs. Those are the games I provided a screenshot above and have been discussing. A seven game sample size (plus three shots against in an eighth game) is a tiny sample size. That's why I took a look at the games individually.

The 2 options are either, Binnington as a whole has played the Avs well in the playoffs, or that playoff stats should always be taken with a grain of salt because of small samples, which also invalidates your argument.

My argument has been and remains Binnington is a mixed bag against the Avs in the playoffs and I can find as many bad games as good. It's really simple.

I know you want to look at the numbers in aggregate because it allows you to throw out the bad games whole cloth, but with a sample size of seven it's completely fair to look at all games individually. If we were talking about hundreds of games I agree that would be cheery picking, but it's not.

Either way, for all of the arguments you could make against Binnington, you are choosing one of the dumbest.

Looking at the individual games against one specific opponent is a dumb argument? How so? It's literally all we have to go on.

 

[bleedblue1223]

I know you want to look at the numbers in aggregate because it allows you to throw out the bad games whole cloth, but with a sample size of seven it's completely fair to look at all games individually. If we were talking about hundreds of games I agree that would be cheery picking, but it's not.

Not sure you understand the meaning of aggregate or cherry-picking.

 

[Shane Diesel]

Not sure you understand the meaning of aggregate or cherry-picking.

Says the poster arguing a sample size of seven is conclusive, but please enlighten me. What am I missing?

 

[bleedblue1223]

Literally never said that, I even gave you the proper counter a few posts prior. But, since you want to argue your position with certainity, that point would invalidate your entire argument.

 

[Shane Diesel]

Literally never said that, I even gave you the proper counter a few posts prior. But, since you want to argue your position with certainity, that point would invalidate your entire argument.

Against the Avs, Binnington is a mixed-bag in the playoffs. That's the argument. And the results of those seven games bear that out. This is extremely simple.

Whatever nonsense you're arguing about now is meaningless.

 

[bleedblue1223]

No one evaluates goalies like that, that's why what you are doing is nonsensical. Saying I am throwing out his bad games by using his entire sample, and you saying you aren't cherry picking by doing the exact definition of cherry picking.

With any other goalie, you would not consider at .925 sv% to be a mixed bag.

 

[Thallis]

I'm arguing his .925 is highly deceptive because you are looking at a sample size of only seven games. Of those seven games three were shit. Four were good. And because of this tiny sample size the good games are enough to bring up his SV% to a decent number.

You think playing three out of seven playoff (half game away from an even split) games at .906 or below is playing well? That's what you're saying?


A whole 67 minutes of actual game time. Wow.

.906 in a loss vs a much better team is not a bad goaltending performance.

 

[Shane Diesel]

.906 in a loss vs a much better team is not a bad goaltending performance.

The league average for SV% in 2022 was .907. So actually it was slightly below average for that season. That's why I included it.

 

[HockeyFan100]

10/10 self own

I don't think you know what a "self own" is.



Oh well Blues fans can live in their own world. I just know when ol Binner is in net there's a good chance my favourite team is gonna pot 5 goals on him. But hey it was 1-1 series that one time in 2022. I guess that's way better than getting swept the previous year.

 

[Golden_Jet]

No one evaluates goalies like that, that's why what you are doing is nonsensical. Saying I am throwing out his bad games by using his entire sample, and you saying you aren't cherry picking by doing the exact definition of cherry picking.

With any other goalie, you would not consider at .925 sv% to be a mixed bag.

925 save percentage is a good average, other poster is arguing for the sake of it.
Most teams would take that any day.

 

[Zonk]

With Woll, I wonder if they rushed his development a bit too fast. Another year with the Marlies, would have done Woll a lot of good.

It is extremely unlikely that Woll would have cleared waivers, so he either played for the Leafs or they lost him for nothing. (He was also 25 years old last season, so doubt that he was rushed.)

 

[Shane Diesel]

There are people that understand sample sizes and then there's HF.

 

[Shane Diesel]

No one evaluates goalies like that, that's why what you are doing is nonsensical. Saying I am throwing out his bad games by using his entire sample, and you saying you aren't cherry picking by doing the exact definition of cherry picking.

With any other goalie, you would not consider at .925 sv% to be a mixed bag.

Sample size. Where it begins and ends with this discussion. And what you continually ignore.

If the sample size was 80+ games then pulling out three games would indeed be cherry picking. With a sample size of seven games pointing out half his games were bad is not cherry picking. This isn't hard. What is three divided by seven? Nearly a majority in the context of this specific conversation.

Small sample sizes can skew numbers and this is a perfect example. How anyone can say a goaltender is great when literally half the games we're discussing are slightly below average to horrendous is beyond me. Your ignorance of this point of how this works and why it's important to the context of analysis isn't negated by you repeating the same line over and over that .925 is great.

The Disadvantages of a Small Sample Size

Sampling errors can significantly affect the precision and interpretation of the results of surveys and empirical research.

sciencing.com

Read up on the topic from actual experts. You want to discuss seven gane stretches I can make a fourth line scrub look like Gretzky.

 

[bleedblue1223]

Sample size. Where it begins and ends with this discussion. And what you continually ignore.

If the sample size was 80+ games then pulling out three games would indeed be cherry picking. With a sample size of seven games pointing out half his games were bad is not cherry picking. This isn't hard. What is seven divided by three? Nearly a majority in the context of this specific conversation.

Small sample sizes can skew numbers and this is a perfect example. How anyone can say a goaltender is great when literally half the games we're discussing are slightly below average to horrendous is beyond me. Your ignorance of this point of how this works and why it's important to the context of analysis isn't negated by you repeating the same line over and over that .925 is great.

The Dangers of Small Samples and Insufficient Methodological Detail

This brief commentary is to speak concerns about the presentation of data in figure 1 in the editorial written by Collins and colleagues1 titled “The Role

academic.oup.com

Read up on the topic from actual experts.

And now you are shifting the goal posts. Original argument was Binnington isn't good against the Avs. A .925 sv% is good. I was the one that brought up that a better counter is simply saying that all playoff sample sizes are going to be flawed because even a large playoff sample size is just a collection of small sample sizes. Instead of saying that, you decided to go down the route of saying a few of his games were really bad, and that outweighs his few really good performances, even though as a whole, they all equate to a really good average.

Now, we can go down the route of a quality start %, which tries to take into account the amount of bad and good starts to get some level of consistency, and yes, 4/7 is above average when evaluating quality starts, so that wouldn't even fit your argument.

 

[rumrokh]

Post it then. It's not my job to defend your assertion.

I don't care. It's a fact. Not posting it doesn't make it less of a fact.

It's incredibly easy to look at team win and shooting percentages over time spans.

 

[Shane Diesel]

And now you are shifting the goal posts. Original argument was Binnington isn't good against the Avs. A .925 sv% is good.

I have done no such. I laughed at a poster claiming Binnington was good against the Avs and then laid out an argument that he's the definition of hit or miss and that a sample size of seven is too few games to make any sort of declaration either way.

I was the one that brought up that a better counter is simply saying that all playoff sample sizes are going to be flawed because even a large playoff sample size is just a collection of small sample sizes. Instead of saying that, you decided to go down the route of saying a few of his games were really bad, and that outweighs his few really good performances, even though as a whole, they all equate to a really good average.

No, not at all. My point was to show he's had almost as many bad games as he's had good against the Avs and declaring him great is silly with only seven relevant games. You're either confused or have assumed I'm making arguments I'm not.

Now, we can go down the route of a quality start %, which tries to take into account the amount of bad and good starts to get some level of consistency, and yes, 4/7 is above average when evaluating quality starts, so that wouldn't even fit your argument.

And the other three? What about those? They certainly weren't great. Which is part of my larger point about sample size. And looking at things holistically when we don't have a proper sample.

He's hit or miss in his games against the Avs, it has been and remains the point.

I don't care. It's a fact. Not posting it doesn't make it less of a fact.

It's incredibly easy to look at team win and shooting percentages over time spans.

"Trust me, bro."

 

[bleedblue1223]

I have done no such. I laughed at a poster claiming Binnington was good against the Avs and then laid out an argument that he's the definition of hit or miss and that a sample size of seven is too few games to make any sort of declaration either way.


No, not at all. My point was to show he's had almost as many bad games as he's had good against the Avs and declaring him great is silly with only seven relevant games. You're either confused or have assumed I'm making arguments I'm not.


And the other three? What about those? They certainly weren't great. Which is part of my larger point about sample size. And looking at things holistically when we don't have a proper sample.

He's hit or miss in his games against the Avs, it has been and remains the point.


So you say.

If a goalie has a very good start 57% of the time in a large sample, would you still consider them a very hit or miss goalie?

 

[Shane Diesel]

If a goalie has a very good start 57% of the time in a large sample, would you still consider them a very hit or miss goalie?

In a large sample sure, this isn't it.

 

[bleedblue1223]

In a large sample sure, this isn't it.

That doesn't matter when we are evaluating whether he's performed well in the actual sample that exists. Sample size matters when you try to extrapolate or project forward. Like if we argued that Binnington will always be good against the Avs in the playoffs and then only point to that small sample.

In the sample we have, he's performed strongly against the Avs in the playoffs. It's not that complicated.

 

[Shane Diesel]

That doesn't matter when we are evaluating whether he's performed well in the actual sample that exists.

I seriously question whether you have any idea what you're even saying at this point. There are seven relevant games in the playoffs against the Avs. Four were good, three were below average to horrendous.

Averages in large sample sizes are useful, averages in small samples are not.

Sample size matters when you try to extrapolate or project forward.

We can do that too. You can look at his numbers during the regular season against Colorao if you want a larger sample to extrapolate. Those numbers are even worse there. Strange, I wonder why his playoff numbers are so much better.

Like if we argued that Binnington will always be good against the Avs in the playoffs and then only point to that small sample.

Taken as a whole he's not played the Avs very well at all.

 

[bleedblue1223]

I seriously question whether you have any idea what you're even saying at this point. There are seven relevant games in the playoffs against the Avs. Four were good, three were below average to horrendous.

We can do that too. You can look at his numbers during the regular season against Colorao if you want a larger sample to extrapolate. Those numbers are even worse there. Strange, I wonder why his playoff numbers are so much better.


Taken as a whole he's not played the Avs very well at all.

Again, you are essentially arguing that a .925 sv% is not good. You can't even make the argument that Binnington has been good against the Avs in the playoffs and only slightly above average in the regular season. That's why I can't take you seriously.

 

[BlueOil]

If the sample size was 80+ games then pulling out three games would indeed be cherry picking. With a sample size of seven games pointing out half his games were bad is not cherry picking. This isn't hard. What is three divided by seven? Nearly a majority in the context of this specific conversation.

it's not half of the games, it's objectively less. do the math you asked everyone else to do before jumping to your conclusions.

Small sample sizes can skew numbers and this is a perfect example. How anyone can say a goaltender is great when literally half the games we're discussing are slightly below average to horrendous is beyond me. Your ignorance of this point of how this works and why it's important to the context of analysis isn't negated by you repeating the same line over and over that .925 is great.

anyone can say it because in 4 of the 7 games he was objectively great with a 93.9 SV% or higher. it's not half and half like you're rounding down to and being a small sample size doesn't invalidate the outcomes seen in the majority of the games. sure, you can suggest that because it's a small sample size you think that trend may be unreliable with more games played but that's simply your personal assertion, not objective evidence he's not great. the trend of him putting up more great than bad games against the Avs could easily continue and the trend would be solidified or could even improve with a larger sample size. you aren't wrong to mention small sample sizes are unreliable, but you're using it in a warped way to fit your argument, which seems entirely to be you simply don't like binnington despite the stats.

 

[miscs75]

Binnington would probably throw a blocker at Marner causing Mitch to go into the player assistance program. I’ll allow it.

 

[Reality Czech]

I don't think you know what a "self own" is.



Oh well Blues fans can live in their own world. I just know when ol Binner is in net there's a good chance my favourite team is gonna pot 5 goals on him. But hey it was 1-1 series that one time in 2022. I guess that's way better than getting swept the previous year.

Typical classy Avs fan right here. He's better than Four-giev, that's for sure.

 

[sfvega]

The league average for SV% in 2022 was .907. So actually it was slightly below average for that season. That's why I included it.

The league-average is against ALL teams. So you're again moving the goalposts. The regular season average isn't a sample size played against a team that was 1st in scoring in 2021 regular season and 4th in 2022. It's not exclusively played against a team that had the highest goals/game in 2021 and 2022 playoffs. And the highest PP% in both playoffs. How are you going to call .907 against a mixed bag of sub and above-.500 teams "average", but .906 against the best offense in tbe league over that span "playing like shit"?