Solving our Defense Problem, with DATA

rainbow

Long time, no see.

I got back from UMass about a week ago, and have been attempting to think of something to post.

While at UMass, we had multiple lessons from different fields of sport, and I was really excited for the Analytics day. I had plenty of questions to ask him, mostly pertaining to hockey, of course.

I asked him my first question, and it turned out, he was more of a Football and Baseball guy.

OF COURSE HE WAS.

One of my questions was “How could the Bruins fill the Hamilton void?”. I wanted his pure opinion. He knew the basic stuff about hockey, but nothing too technical, so I really didn’t get much help. So, I decided that I would make my project for the two weeks I was there, to find the right defenseman for our team.

While I was at UMass, I was still checking twitter, and I picked two players that were rumored to be in talks with the Bruins, and those players were; Cody Franson and Marek Zidlicky.

To find this out, I used 14 points of comparison, with 4 of these being points you may or may not be familar with.

iCorsi (Individual Corsi): shots + shot attempts that were blocked or missed

iFen (Individual Fenwick): shots + shots that missed the net.

GF60/GA60: Goals scored for/against when the player is on ice.

*I really want to point out that both Marek Zidlicky and Cody Franson were both traded at the Trade Deadline. Cody went from Toronto to Nashville, and Marek went from New Jersey to Detroit, which may or may not cause some differences in the data.*

For visualizing my data, I will be using Spotfire. If you are an analytic and graph/chart junkie, I highly recommend this tool.

This is all the Data I gathered in an excel document.

hamfranzid

(I gathered my data from the NHL website and Puckalytics.)

The most red marks (lowest value) come from Zidlicky.

Hamilton has the most bold (highest value), but that was expected.

Our goal is to not find an exact copy of Dougie, but to find someone who can fill that huge void we are missing with due to the trade. Remember, we also acquired Matt Irwin & Colin Miller, and Zach Trotman is expected to be with the big club this season, so the void may also shrink a little with that.

Since I can use only 2 points of comparison per graph, I wanted to make only one chart as the main comparison.

The two points of data I chose were GA60 & GF%.

GF% is the percent of goals scored while player is on the ice that were scored by his team.

Mathmatically, the formula is:  Goals For / (Goals For + Goals Against)

After doing all of my math, why did I chose these?

I picked GA60 because I really think that if a defenseman is really good, then there shouldn’t be a million goals if they are on the ice.

I picked GF%, because I feel like a D-Man should have some sort of offensive upside. I feel like plus-minus and GF% are in the same boat, in a way. I kinda want to consider this the “moneyball” statistic of hockey, in my opinion.

Both of these are the data for 5v5 Even Strength. This also applies for Fenwick, Corsi, and GF/GA60. That’s why the Goals and Points I have on the data don’t match with the other data.

hamfranzid graph

For Reference: Green=Hamilton, Blue=Franson, and Red=Zidlicky.

All circles are sized by GF60.

When visualized…

All of them have up sides.

For Zidlicky, he has a really low GA60, and a high GF%, but offensively, there is hardly anything. He is the lowest of the 3 when it comes to Points/60 and GF60.

Franson has the highest GF60 on the graph, and is more consistant compared to Zidlicky. When looking at the data chart on the excel document, he only has 2 red data points, while Zidlicky has a whooping 7. Also, since Franson is still pretty young (27y/o) and Zidlicky is older (38y/o!), Franson does have more room to grow.

Both Franson and Zidlicky began the season on teams that struggled offensively and defensively. The teams they began the season with, Toronto Maple Leafs and New Jersey Devils Respectively, didn’t make it to the playoffs, and some of these stats do show for that. However, that is the main reason I compared each player individually (iCorsi, iFen), instead of using the Team Corsi/Fenwick.

To Answer the question…

When looking at these stats, I feel like the winner here is Cody Franson. His stats aren’t as glowing compared to Dougie’s, but he is more consistant with his stats compared to Marek Zidlicky. The Bruins are looking for a long term D-Man, to maybe even fill in for Big Z when he retires. Cody Franson could be what we need. He is young, at 27 years old, has experience, and is looking for a long term contract. I would prefer we lock Franson in now, because if we sign him short term, and he does progess nicely with the B’s system, he could be demanding more money sooner than later.

If we were wanting to sign Zidlicky, we should do something Peter Chiarelli didn’t like. That is to sign him short term. More particularly, to a one year contract. Since he is older, there is a risk of him declining. His goals and points have decreased this past season (12G/42P in 2014, 7G/27P in 2015), and putting too much money in at long term could cause cap issues, and the Bruins have struggled with that in the past.

BONUS: Contracts I would offer.

I added this section because we learned about contract offers and GM type stuff on the last day of my program, and I really liked it, so I wanted in incorporate it.

For Franson: 4 year, $17.4 million. Limited NTC 2015-16,2016-17.

For Zidlicky: 1 year, $1.4 million. (Short term, and to the point.)

You can open you eyes now, no more Math!

Actually, math is good.

Any who, the UMass Summer Pre College for Sport Management helped me discover that there is more to sports than I believed, and I hope to be able to show what I learned on future posts, and in my future.

For any questions, tweet me (@bruinsatgahden), or let me know down in the comments below!

To play around on spotfire with my data, click here.

see you in october.

^AHC

Advertisements

5 thoughts on “Solving our Defense Problem, with DATA

  1. Pingback: NHL concussion suit update; MacLellan talks MoJo; Travis Roy, 20 years later (Puck Headlines) | Jeffery Lipton - freelance writer

  2. Pingback: NHL concussion suit update; MacLellan talks MoJo; Travis Roy, 20 years later (Puck Headlines)

  3. Pingback: NHL concussion suit update; MacLellan talks MoJo; Travis Roy, 20 years later (Puck Headlines) | Wolrd Blogs

  4. Pingback: playing the trade game (first edition) | bruins at the gahden

  5. Pingback: Things to look out for in the Bruins 2015-16 season | bruins at the gahden

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s