GW 32 Transfers and Team Selection

Please welcome to the statistics on team selection and transfer activity for GW 32. Full stats are available in this google spreadsheet.

TRANSFERS AND TEAM VALUE

Random Sample Top 10K
Number of Managers 20,000 10,000
Average Transfers Made 0.46 0.83
Point Hits Taken 7.9% 12.3%
Average Points Deducted (Incl. No Hits) 0.54 0.56
Wildcards Played 0.3% 0.3%
Wildcard Status
– played 49.8% 93.7%
– available 50.2% 6.3%
– active 0.0% 0.0%
Average Team Value ₤100.4m ₤109.3m
Average Money in the Bank ₤0.7m ₤0.9m

TRANSFERS MADE

Number Random Sample Top 10K
0 74.5% 36.7%
1 12.2% 45.6%
2 8.9% 15.2%
3 2.5% 1.8%
4 and more 1.5% 0.3%
Wildcard 0.3% 0.3%

POINT HITS TAKEN

Point Hit Random Sample Top 10K
0 points 92.1% 87.8%
4 points 5.1% 10.7%
8 points 1.6% 1.2%
12 points 0.6% 0.2%
16 points and more 0.6% 0.1%

TRANSFERS IN

Random Sample Top 10K
1 Barkley 3.1% 1 Barkley 10.9%
2 Mutch 2.8% 2 Azpilicueta 5.2%
3 Lukaku 1.8% 3 Mutch 4.8%
4 Yaya Touré 1.7% 4 Cahill 3.6%
5 Bony 1.6% 5 Lukaku 3.3%
303 different players bought 207 different players bought

TRANSFERS OUT

Random Sample Top 10K
1 Koscielny 3.6% 1 Koscielny 12.8%
2 Adebayor 2.2% 2 Nolan 7.0%
3 Mertesacker 2.1% 3 Sterling 5.2%
4 Giroud 1.8% 4 Mertesacker 4.8%
5 Sterling 1.8% 5 Szczesny 3.2%
376 different players sold 266 different players sold

FORMATIONS

Formation Random Sample Top 10K
‘3-4-3’ 42.3% 77.3%
‘3-5-2’ 7.3% 4.2%
‘4-3-3’ 15.3% 14.1%
‘4-4-2’ 31.5% 3.6%
‘4-5-1’ 0.9% 0.1%
‘5-2-3’ 0.7% 0.4%
‘5-3-2’ 1.8% 0.3%
‘5-4-1’ 0.4% 0.0%

CAPTAINCY

Random Sample Top 10K
1 Suárez 39.4% 1 Suárez 92.5%
2 van Persie 9.7% 2 Hazard 2.3%
3 Sturridge 6.8% 3 Rooney 2.1%
4 Hazard 5.2% 4 Sturridge 2.0%
5 Rooney 4.7% 5 Lukaku 0.5%
250 different captain choices 20 different captain choices

STARTING GOALKEEPERS

Random Sample Top 10K
1 Mignolet 25.6% 1 Mannone 31.1%
2 Cech 10.9% 2 Boruc 13.6%
3 Szczesny 9.3% 3 Cech 9.7%
4 De Gea 7.0% 4 Mignolet 8.3%
5 Howard 6.7% 5 Adrián 8.3%
62 different goalkeepers 31 different goalkeepers

STARTING DEFENDERS

Random Sample Top 10K
1 Coleman 36.5% 1 Coleman 75.7%
2 Mertesacker 21.9% 2 Ivanovic 25.7%
3 Zabaleta 21.0% 3 Zabaleta 25.0%
4 Ivanovic 19.0% 4 Cahill 23.8%
5 Fonte 18.6% 5 Azpilicueta 22.6%
196 different defenders 94 different defenders

STARTING MIDFIELDERS

Random Sample Top 10K
1 Yaya Touré 44.5% 1 Yaya Touré 92.4%
2 Hazard 38.8% 2 Hazard 84.6%
3 Lallana 30.3% 3 Lallana 41.4%
4 Barkley 15.9% 4 Gerrard 41.3%
5 Gerrard 13.7% 5 Silva 22.9%
235 different midfielders 93 different midfielders

STARTING FORWARDS

Random Sample Top 10K
1 Suárez 50.3% 1 Suárez 99.2%
2 Sturridge 33.8% 2 Sturridge 82.2%
3 Giroud 20.0% 3 Lukaku 50.0%
4 Lukaku 19.5% 4 Rooney 17.3%
5 van Persie 13.6% 5 Dzeko 13.6%
114 different forwards 37 different forwards

BENCHED PLAYERS

Random Sample Top 10K
1 Kelvin Davis 12.5% 1 Ward(CRY) 23.0%
2 Whittaker 10.1% 2 Kelvin Davis 18.1%
3 Baker 7.9% 3 Adrián 11.4%
4 Gabbidon 7.0% 4 Collins 10.4%
5 Harper 6.4% 5 Sterling 9.9%
626 different players on the bench 398 different players on the bench

OWNERSHIP AND CAPTAINCY DISTRIBUTION FOR MOST CAPTAINED PLAYERS

This slideshow requires JavaScript.

RAW DATA FOR TOP 10K

An Excel file with all the entries for the top 10K can be downloaded here.

About This Post

In this post, I take a look at 2 samples: randomly selected 20,000 FPL teams and the top 10,000 FPL teams. Because of the law of large numbers, we can make inferences about the overall FPL league based on the statistics for the random sample. Interval estimates for most statistics for the whole FPL game are also available. If you’re familiar with the concept of confidence intervals, you can point at a specific number characterising the random sample to see 99% confidence intervals for the respective number characterising all the 3 million FPL managers.

Advertisements

15 comments on “GW 32 Transfers and Team Selection

  1. Love the work you’re doing! Wondered whether, if on the first week of next season, it’s possible to get data from all of this season’s top 10k based on the gameweek history page?

    • I think it’s possible theoretically, but impossible practically. All teams will have other IDs next season, so I’ll need to run a script for all the teams in FPL to identify those who were in the top 10K a season before. The speed of my scripts is roughly 30-40K FPL pages per hour, so I’d need 3M / 30K = 100 hours to go through all the teams and check if they were in the top 10K this season. This doesn’t sound like something that I’m ready to do 🙂

  2. These posts are really amazing. What i think would be good is for the most selected players, to show the 10 most owned in each position. While 5 is a solid indicator of ownership, I think 10 would be more interesting.

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