Q&A with Jason Nightengale

[Brawndo]

https://twitter.com/llysowski/status/1289169697594683392?s=21

[North Buffalo]

Tweet deleted... what did it say?

[steveoath]

Is it a link to this article?

https://buffalonews.com/sports/sabres/how-the-sabres-use-analytics-q-a-with-jason-nightingale-assistant-director-of-scouting/article_8b1189d4-d1b5-11ea-b9e1-9f13799724e3.html

 

Edited July 31, 2020 by steveoath
Gist: stats are good m'kay?!

[LGR4GM]

On the upcoming NHL draft:

Quote

We have access to data, to things that will tell us about player tendencies, than we’ve ever had and we do have time now to really implement a data-driven player evaluation process, specifically over the last month I’ve really enjoyed working with Jeremiah and our scouting staff. They’ve been working hard in revisiting the 2020 draft and trying to incorporate more information, to be open-minded when it comes to evaluating players... - Jason Nightingale

Edited July 31, 2020 by LGR4GM

[LGR4GM]

Lol there is a line in there that they used analytics and data to help drive drafting over the last 2 years... 

[Weave]

What motivation does he have to lie about it?

[LGR4GM]

Just now, Weave said:

What motivation does he have to lie about it?

None, but it reinforces something I have been told, by actually a few ppl now, that the Sabres used analytics to justify their decisions not make decisions. 

[Marvin]

8 minutes ago, LGR4GM said:

None, but it reinforces something I have been told, by actually a few ppl now, that the Sabres used analytics to justify their decisions not make decisions. 

I have heard that they would make a decision and then cherry pick data to justify the decision (or lack thereof) ex post facto.

[Stoner]

So now the Sabres are using DATA and VIDEO?! What's next, the hockey department getting their own email addresses?

[MakeSabresGrr8Again]

1 hour ago, LGR4GM said:

None, but it reinforces something I have been told, by actually a few ppl now, that the Sabres used analytics to justify their decisions not make decisions. 

This doesn't make sense to me.

Let's say I make a decision to go with Ryan Johnson. I then check the analytics and justify that pick. Or, I go to analytics to figure out which player to pick, shouldn't those results show the same player as Johnson being the best pick?

My conclusion would suggest that something else in the decision making process is where they're possibly going astray....or maybe analytics don't work as well as some think.

It's a risk no matter what and sometimes you make the right decision, sometimes you don't. I'm sure we all know someone who wins all the time on scratch-off lotteries while we seem to get all the losers.

 

Edited July 31, 2020 by MakeSabresGrr8Again

[MakeSabresGrr8Again]

34 minutes ago, PASabreFan said:

So now the Sabres are using DATA and VIDEO?! What's next, the hockey department getting their own email addresses?

Nope, the next thing is having their Twitter account suspended for 2 months.

[LGR4GM]

7 minutes ago, MakeSabresGrr8Again said:

This doesn't make sense to me.

Let's say I make a decision to go with Ryan Johnson. I then check the analytics and justify that pick. Or, I go to analytics to figure out which player to pick, shouldn't those results show the same player as Johnson being the best pick?

My conclusion would suggest that something else in the decision making process is where they're possibly going astray....or maybe analytics don't work as well as some think.

It's a risk no matter what and sometimes you make the right decision, sometimes you don't. I'm sure we all know someone who wins all the time on scratch-off lotteries while we seem to get all the losers.

 

No. You are creating facts to suit theories instead of theories to suit facts. 

I don't know if this happened but from what I have heard (the facts available) they would focus in on a player and then ask the analytics team to give them numbers to justify something they were going to do anyways. 

[dudacek]

It can be as simple as this:

Jason Botterill believes zone exits and entries is the best measure of a good defenceman, that defencemen are more important that wingers, that skating is the most important asset a player can have, that defensive laziness is a huge red flag, and that college-bound players have a better chance for NHL success.

He asks Nightengale to analyze entries and exits. Nightengale does and discovers Johnson was elite at entries and exits.

The old school scouts rate Johnson’s skating as superb, and the college and defenceman boxes check off. According to those things the organization values, Johnson ranks really high.

Nightengale did his job and the analytics supplied exactly the info Botterill was looking for. However, they don’t address the fact that Botterill may have been wrong to give so much weight to the factors he did, and if more value had been put on goals, and the CHL and less on “character,” Kaliyev might have come out ranked higher.

The data is only as useful as the context its used in.

Edited July 31, 2020 by dudacek

[MakeSabresGrr8Again]

32 minutes ago, LGR4GM said:

No. You are creating facts to suit theories instead of theories to suit facts. 

I don't know if this happened but from what I have heard (the facts available) they would focus in on a player and then ask the analytics team to give them numbers to justify something they were going to do anyways. 

So you're saying that analytics had nothing to do with their decision to pick a player. That's why I said there's something other than analytics in the decision making process that's causing the failure in drafting. The analytics results should be the same either way and only the interpretation of them could be a problem. 

[Thorner]

2 hours ago, PASabreFan said:

So now the Sabres are using DATA and VIDEO?! What's next, the hockey department getting their own email addresses?

They still get their internet via mailbox CD

[Thorner]

1 hour ago, dudacek said:

It can be as simple as this:

Jason Botterill believes zone exits and entries is the best measure of a good defenceman, that defencemen are more important that wingers, that skating is the most important asset a player can have, that defensive laziness is a huge red flag, and that college-bound players have a better chance for NHL success.

He asks Nightengale to analyze entries and exits. Nightengale does and discovers Johnson was elite at entries and exits.

The old school scouts rate Johnson’s skating as superb, and the college and defenceman boxes check off. According to those things the organization values, Johnson ranks really high.

Nightengale did his job and the analytics supplied exactly the info Botterill was looking for. However, they don’t address the fact that Botterill may have been wrong to give so much weight to the factors he did, and if more value had been put on goals, and the CHL and less on “character,” Kaliyev might have come out ranked higher.

The data is only as useful as the context its used in.

I’d say this is pretty much exactly my reading too.

I like this type of player, go find me guys that have strong metrics in those specific areas. The Botterill auto-filter when sorting. 

Edited July 31, 2020 by Thorny

[kas23]

The other, more ominous scenario was that Botterill plans on drafting Johnson, for whatever non-analytical reason. He asks Nightengale for his analytics. Botterill then goes through and confirms his suspicions based on all the positive analytical data, while ignoring the negatives (because, after all, it’s just analytics). 

[Marvin]

4 hours ago, MakeSabresGrr8Again said:

This doesn't make sense to me.

Let's say I make a decision to go with Ryan Johnson. I then check the analytics and justify that pick. Or, I go to analytics to figure out which player to pick, shouldn't those results show the same player as Johnson being the best pick?

My conclusion would suggest that something else in the decision making process is where they're possibly going astray....or maybe analytics don't work as well as some think.

It's a risk no matter what and sometimes you make the right decision, sometimes you don't. I'm sure we all know someone who wins all the time on scratch-off lotteries while we seem to get all the losers.

From what I heard, Botterill had several problems with analytics:

1. For the statistics that are generally regarded as useful, he over-valued some and under-valued others.

2. He had his own extra analyses that he wanted done.  Some may have been useful whereas others were head-scratchers.

3.If he zeroed in on someone, he would over-emphasise the positive analytics for that player and under-emphasis the negative analytics.

If any one of these is true (IMHO, #1 is immediately obvious given the imbalance in skills among the defencemen and the lack of centres; #3 is immediately obvious given the Frolik and Simmonds trades and that he allegedly pushed for Tage Thompson over better prospects in the ROR trade), then we have a misuse of statistics.  I think I might put this into the Analytics group.  A thread on how teams and people misuse statistics would be useful.

Also, unless you are a numbers person, it is easy to get snowed in by a blizzard of digits, decimal places, and plus and minus signs.  You can fall into the numbers verifying preconceived notions of player quality versus learning what the numbers are telling you.

[Weave]

Sounds like most every human on Earth, really.  We all have our biases and seek out data that supports it.