Last week, The Athletic featured a fun Jack Lang article looking at a hypothetical Premier League expansion draft.
Mirroring the Seattle Kraken’s upcoming draft as the NHL’s 32nd and newest team, the 21st Premier League team would select their squad by cherrypicking from other squads – although each existing team can protect a selection of their best players.
Jack Lang’s article makes use of The Athletic’s network of journalists to recreate which players existing teams would protect and which would be begrudgingly made available.
FC Python, with a somewhat smaller network, will use Transfermarkt’s player values to select which players are protected by their current teams and which remaining players get selected to the Premier League’s 21st team.
In this article, we’ll go through the rules of our mock draft, a process to run our version of it and building the best squad from the available players.
Expansion Rules
To achieve the NHL’s dual aims of creating a team capable of competing with their best rivals immediately, while maintaining fairness with existing teams, there are a number of rules as to who can be picked.
The rules are deeper than we need to get into here. Boiling them down to use as rules for our draft, we’re going to go with the following:
- Existing teams can protect 1 goalkeeper from selection
- Existing teams can protect 11 outfield players
- Existing teams must make available 3 outfield players with significant playing time – we’ll call that 10 90s played last season.
- Promoted teams will be exempt from losing players.
We’ll then find the most valuable Transfermarkt starting XI and see what we come out with.
Data Process
The process to scrape and process our data is quite simple. Firstly, we work through each team on Transfermarkt’s Premier League page and save each player, their value, team and position to a data frame – along with a bit of tidying of the fiddly format of the values.
You can find the code to create this table here.
90s played figures are added from another source, before splitting the dataframe up by team to calculate their protected and available players.
Goalkeepers are easy – the highest valued stopper is protected.
Outfield is slightly more complicated. Firstly, the three lowest-valued players that have played more than 10 90s are made available. Following this, the 11 outfield players with the highest values are blocked from being selected.
Notable exposed players
Let’s start by looking at some of the players that we have available to us. Starting with some of the players key to their teams, these are the 5 players with the most minutes played in the 19/20 season:
1. Jonny Evans
2. Chris Basham
3. João Moutinho
4. Dan Burn
5. Matt Doherty (alright, this is a bit unfair)
Turning to $$$, our top 5 players by Transfermarkt value:
1. Henderson (TM value – €28m)
2. Zinchenko (€24m)
3. Marcos Alonso (€18m)
4. Fred (€17.5m)
5. Harry Wilson (€17.5m)
Final Team
Now then, let’s crack on and build the highest rated team by TM figures.
Our three most valued goalkeepers are Dean Henderson (€20m), Emi Martinez (€8m) & Fraser Forster (€8m). Henderson starts in nets, and I think we are off to a solid start.
As for outfield, we find ourselves with a fairly strong back line, and some options for strength in depth. Taking a flat back four by the highest values, we find ourselves with Mendy – Aké – Christensen – Azpilicueta. If any of those don’t do it for you, help yourself to Zinchenko, Bailly or Reece James instead.
You’ll notice that we still have options from Champions League qualifiers here, showing their strength in depth at the back and sheer number of more expensive options ahead of them. This likely also hints at a possible bias of the TM values towards attackers.
If you insist on mid-table representation, Boly, Hoedt, Hause & Valery give you other options.
Moving up to the midfield 3, our options start to drift off. We could go for some players crucial to their teams & racking up major minutes-played numbers: Noble, Moutinho or Fernandinho.
But on highest values and somewhat fitting into a functional 3, we are taking Fred, Henderson & Loftus-Cheek. Plenty of options that do not seem too far off their quality though. Barkley, Özil & Milner might find themselves on the bench, with Willock and Choudhury offering decent younger options too.
A front three is really where we might struggle. With any good attackers earning themselves a high enough value to leap into their team’s protected basket, we are left with the relative dregs – as most fantasy players will be well-acquainted with.
Occupying the lead of the three attackers, we have options of Origi, Iheanacho, Tosun & Giroud. With none of these playing more than 11 90s last year, this might be a long season.
Supporting them, Harry Wilson and Lamela are our most valuable options. Although Bernard, Shaqiri and Walcott aren’t too far off them.
Throwing all of this together, I think we get our strongest line-up as follows:
Looking for feedback, the 21st team get mixed reviews from an esteemed panel of experts:
- “Zero goals in that team”
- “Probably would finish 14th”
- “No danger of relegation at all”
It doesn’t quite match the performance of the last NHL expansion team. The Las Vegas Golden Knights made it all the way to the final of the 2018 Stanley Cup playoffs in their inaugural season.
If you think there are better teams available, a selection of the available players are below. Let us know what you come up with on Twitter, or if you fancy a bit more on scraping, check out the tutorials!
Goalkeepers
Defenders
Midfielders
Forwards
