S97 defense part 0: introduction

[VattghernCZ 1,114]

A few seasons ago, I posted a multi-week media spot about some goaltending stats that aren't available to us, but are commonly tracked in the NHL - stuff like Goals Saved Above Average, Quality Starts or Really Bad Starts. It was tedious, because it made me manually go through all the game logs and compile the data. Still, I enjoyed doing it.

 

 

 

Last offseason, I got my feet wet in using Python (thanks you, AI!) to create a few programs to collect the data for me, and I put together a big piece on hitting, where I made up a few stats - successful hits (hits that separate the puck carrier from the puck), effective hits (hits that lead to a turnover) and ineffective hits (hits that separated the puck carrier from the puck, but didn't lead to a turnover).

 

 

This offseason, my focus is defense. Once again, I will be using Python programs to collect data from all game logs and compile it. I wanted to post this article this week, but I didn't find the time to run the programs (courtesy of Kingdom Come Deliverance 2 release) but still need the 6 capped, so I'm writing this introduction that would have been a part of the article anyway. In this 0th part, I will introduce the stats I will be tracking, and the method by which I am collecting them.

 

Successful Hits (SH)

Physical play will always be a part of defense. I already used this made-up stat in last offseason's article. The reason I will be using Successful Hits instead of Effective Hits is simple: in said article about hitting, there was a big difference in the ratio of effective to ineffective hits between forwards and defenders - forwards were generally much more efficient in hitting. The reason is simple, as @Pifferfish pointed out; Defense STHS attribute is the most important attribute for puck retrievals, and defenders typically have this attribute higher than forwards. So if a forward hits the opposing puck carrier, there are 2 forwards and 2 defenders available to retrieve the puck, as opposed to 3 forwards and 1 defender in case a defender lays the hit. I will only use Successful Hits to eliminate this discrepancy.

 

The Python program will count a Successful Hit for a player every time it finds the phrase "is hit by [player name] and loses puck." in the game logs.

 

Puck Retrievals (PR)

Every time there is a loose puck, either after a hit, a rebound or a blocked/missed shot, you want your team to be the first one to get to the puck. It's especially important in your defensive zone.

 

The Python program will count a Puck Retrieval for a player every time it finds the phrase "Puck retrieved by [player name]" in the game logs.

 

Pass Interceptions (PI)

You don't want your opponent to settle in your defensive zone and get the cycle going, or to send a breakout pass from their defensive zone. The premier two-way forwards are proficient in cutting away passing lanes and intercepting the passes between the opponents to start a quick counterattack.

 

The Python program will count a Pass Interception for a player every time it finds the phrase "Intercepted by [player name]" in the game logs.

 

Some stats I would like to compile, but can't be a**ed to think about what formulation to use, because it's not as straightforward as those stats I'm tracking: rebounds clearing, breakout passes, defensive zone exits. Not to mention any stats including expected goals, as these are literally impossible to track without shot maps.

 

Hopefully, the article is coming next week. Unless the programs take too long to run. I'll probably ditch the project if that's the case.

Edited February 9 by VattghernCZ