Publications by Part Time Analyst
Using R and Poissance Distribution to Forecast the F1 World Championship
The excitement of Formula 1 racing lies in the intense competition between drivers and teams as they navigate the twists and turns of each race to claim the coveted championship title. In recent years, data-driven predictions have become increasingly popular in sports, and F1 is no exception. Inspired by FiveThirtyEight’s comprehensive soccer pre...
3728 sym Python (7934 sym/4 pcs) 6 img
Kaggle Playground Series – Tidymodels
Hello readers, we are entering another Kaggle playground competition, so get your Yorkshire tea ready and enjoy the process of joining. This month the competition I entered is this one https://www.kaggle.com/competitions/playground-series-s3e7It’seiew It’s looks like looks are canncellations from hotels and spoiler alert – I had a lot of fun...
3891 sym R (9316 sym/8 pcs) 20 img
Cricket Weighted Batting Average in R
Hello, I hope you have your Yorkshire tea ready as today I am going to be exploring weighted averages using R. bb3_sum = bb1 %>% mutate(wick = if_else(filt == "bowled", "bowled", if_else(filt == "caught", "caught", if_else(filt == "caught and bowled", "caught and bowled", ...
3913 sym R (7552 sym/4 pcs) 16 img
Kaggle January Playground Series – Tidymodels
Hello, hope you have your Yorkshire tea ready this is going to be a new series on the blog in which each month I am going to be tackling Kaggles monthly playground series. Find the link to Januarys below feel free https://www.kaggle.com/competitions/playground-series-s3e1 So let’s get started EDA Above is the structure of the training datase...
4260 sym R (3127 sym/4 pcs) 18 img
Australian GP FP2
Hello, this is a new series which I will be analysing the data from Friday practice at the Grand Prix. The focus is going to be on strategy so key to a Formula 1 race. This is the first version of this analysis, lookout for more information on future races Tyre Degradation Relatively low degradation across the 3 compounds despite the soft bein...
1805 sym 6 img
Predicting Twenty 20 Cricket Result with Tidy Models
Hello, hope you have your Yorkshire tea to hand and sitting comfortably ready to read today’s blog. In it I am going to be doing some machine learning with tidymodels to predict the outcome of some twenty20 cricket matches. I am using the data from cricsheet as used in this blog and using the win probability added metric to create variables fo...
4355 sym R (6032 sym/5 pcs) 12 img
1st Ashes Test — My Team
Hello welcome to today’s blog looking at something a bit different. The first ashes test starts on 1st August and England will imminently name their squad. In this piece I look at what the team for the first game might look like. First things first this is the team that played the last test match England played in February in the West Indies Je...
6963 sym
Predicting Qualifying — 2
In the last blog I outlined creating a model which predicts the fastest time for each driver in F1 qualifying. theparttimeanalyst.com/2019/07/10/predicting-f1-qualifying/ Today I am going to be dissecting the model to understands its strengths and weaknesses and to look if their is any bias within the model. First lets look at the importance matr...
2997 sym 12 img
Finding Undervalued Air Bnb’s
Hello, today I am going to do an EDA (exploratory data analysis) on AirBnB in the New York area. This data set is available here on Kaggle https://www.kaggle.com/dgomonov/new-york-city-airbnb-open-data Lets read the data into R and take a look of it So I can see there are 17 columns and over 48000 records with information covering the price and ...
5526 sym 36 img
F1 Drivers Rated
Hello, welcome to today’s blog and in it I’m going to be developing methods to evaluate F1 drivers. Currently there is no real way to tell if an F1 driver is any good. It seems sort of arbitrary how a racing driver is decided if they are good or not. Being a data fan I think there is much more that can be done to rate racing drivers and F1 dr...
5278 sym 14 img