Professional Documents
Culture Documents
Mills, Katie
5/24/2017
INTRODUCTION
Our group is comparing statistics in hockey. We comparing goals scored with other data.
This paper explores the relationship between goals scored and average time on ice per game of
the top 30 centres in the 2016-2017 NHL season. I chose this topic to learn more about a
subject I know very little on. By researching statistics in hockey I hoped to learn more about the
BACKGROUND
In order to be more precise we kept our data as uniform as possible. We made sure to
only use centres so that the difference in position would not affect the amount of goals scored.
We also made sure that we all used the same top 30 players from the same season. This is
important because it allows us to compare our data with each other. This paper focuses on the
average time on ice per game per player compared to the amount of goals scored in the
season. Average time on ice refers to the amount of time the players are on the ice for. In
hockey, players play in shifts to allow them to rest and different players to take over. Generally
shifts tend to be about 45 seconds long, but defense players often have longer shifts. Usually
the longer the average time on ice is, the better the player is. With this in mind I compared the
amount for goals scored with the average time on ice to look for a correlation.
The data for NHL statistics is located on the NHL official website. Statisticians have
already compiled data for number of goals scored, -/+, average time on ice per game and other
data. Before making my graphs I had to convert the average time on ice per game from minutes
and seconds to just seconds. This involved multiplying every minute by 60 then adding the
remaining seconds. The NHL website also let us sort by many different variables, including
position of the player, what teams they play for, what season we want the data from, and
whether a player was a veteran or a rookie. Using this information we were able to decide as a
group under what restrictions we wanted to view our data. We were able to sort the
information according to position played and chose the top 30 centres. This way our group not
only had the same data to work with, but our conclusions from the data would be more
accurate.
TREND ANALYSIS
15 y = 0.0009x + 26.417
10 R² = 0.0002
y = 26.535e1E-05x
5
R² = 4E-05
0
0 500 1000 1500
The trends presented in this graph are very weak and slightly positive. There appears to
be little to no correlation between the number of goals scored and the amount of time on ice.
There appears to be a couple outliers in the group with more goals scored than any other
players but their time on ice is in the middle range. The time on ice per player also has a large
Total
7
6
5
Frequency
4
3
2
1
0
The most frequent range is 1050-1079 seconds. There is a left skew to the graph but
then a sudden drop off after the max frequency value. Other than the most frequent range the
rest of the ranges seem to be around the same frequency. One of the higher outliers form the
previous graph falls into the largest frequency range of 1050-1079 seconds, while the other falls
under one of the less frequent ranges, 1230-1259 seconds. This indicates that the average time
Time on Ice
Series3
Series4
Series5
0 500 1000 1500
CONCLUSION
In conclusion there is no correlation between average time on ice per game and the
amount of goals scored in a season. This goes against what we first assumed about the data.
Usually the amount of time on ice is linked to how useful a player is to the team, if they aren’t
as good they won’t play as often. What this data shows us is that the time on the ice is not
related to how many goals the player may score. This could mean that the value of the player
could be measured in different ways then how many goals scored. It could be related to how
many assists are made, or their endurance. If a player is very good but has low endurance they
would have less time on the ice then another player even though they are more valuable. This
data points to other factors that affect the time on ice and the goals scored in a season, but it
Official Site of the National Hockey League. (n.d.). Retrieved May 24, 2017, from
https://www.nhl.com/