Mass-Analysis of Game Events?

[Thirty One]

Could anyone explain how it's done?

For example, how does Extra Skater turn this:

Into this:

Any help would be appreciated.

Click to read more...

 

[MarkGio]

I'm not sure what the "C" and "F" stand for. Corsi and Fenwick?

The first spreadsheet looks like data collection of shots. It will go on for the duration of the game. The second spreadsheet summarizes the information from the first spreadsheet.

I can see from the first spreadsheet that at 11 seconds into the game, Kreider takes a wrist shot from 53 feet and misses the net. Sounds like he dumped the puck to chase it. That is provided on the first line of the spreadsheet. I can see roughly 20 lines down the first spreadsheet that Kreider takes another shot 15 minutes into the game.

So the second spreadsheet shows Kredier having 6 shots for (SF) at the end of the game. While he was on the ice, which is a sum of all his shifts, the opposition shot the puck against Lundquist 5 times, leaving Kreider with 5 Shots against (SA)

And so forth.

 

[MarkGio]

The percentage of shots for (SF%) is a ratio of the SF to the shots pertaining to a given player (SF + SA). In other words, SF% = SF/(SF + SA)

As you can see in the second spreadsheet, Brian Boyle had 3 shots for (SF) and 3 shots against (SA) during the 6.9 minutes he was on the ice for the whole game. 50% of all shots involving Brian Boyle were his Shots for [3/(3+3) = 0.50].

Is 50% a good percent? I dunno, let's compare it to the average. We do this by comparing Boyle's SF% to the rest of his teammates SF%, resulting in Relative (to Rangers players) shooting for percentage, or called "Rel SF%" in the spreadsheet.

I see Brian is 13.6% less than the average, making the average 63.6% (Brian's 50% + 13.5%)

 

[Thirty One]

Sorry I don't think I explained myself very well. I understand the calculations. I'm wondering how they mechanically input that mass of data and spit out those outputs. I'm sure they're not doing it by hand. Is there some add-on in excel?

 

[MarkGio]

Some data I don't understand. Like why is Lunquist on the ice for only 47.3 minutes in a 60 minute game. He wasn't pulled?

Plus I don't understand Lunquist's Shots against. According to the box score, Lundquist faced 21 shots


I think the shooting percentage and save percentage is different than the conventional NHL formula (shots/goals). I think its a +/- sort of stat, except in percentage. Some sort of any/all shots on goal when a player is on the ice relevant to the goals.

GF is basically +/- as far as I can tell

 

[MarkGio]

Sorry I don't think I explained myself very well. I understand the calculations. I'm wondering how they mechanically input that mass of data and spit out those outputs. I'm sure they're not doing it by hand. Is there some add-on in excel?

Oh, I see. I have no idea. Copy and paste the data from the box scores perhaps? I believe NHL.com has better data too


Edit: FYI, you most certainly did explain it poorly. That was a waste of my time

 

[Cunneen]

Could anyone explain how it's done?

For example, how does Extra Skater turn this:

Into this:

Any help would be appreciated.

You need to be able to code. You write code that scraps the data from NHL.com, and then manipulates the data into certain stats, and then displays them on a site.