I explain the simple, yet potentially helpful, FPL Captain Metric I developed a while back in the hope of making the decision each week a little easier
First of all, I feel it’s my responsibility to let you all know that I don’t profess to have worked out some magic formula for picking a successful captain week to week – that, I’m afraid, does not exist.
Football is an unpredictable game and as a result of that, as is FPL, of course, so there are always going to be surprising outcomes, despite what this metric tells us, but what I believe it can do, is help us in those weeks where things are extremely tight between 2 or even 3 players, not to mention the relief of not having to agonise about it all week.
For example, many times in a season, you could flip a coin between player X and player Y for the captaincy, as most of the FPL experts will probably tell you and in complete truth, it is very close, no matter how you look at it, but how do we differentiate between two players when it becomes difficult?
How can we get to that definitive point where we say, “Yes, you’re my captain”?
We all like to play FPL differently. Some of us use our emotions and gut-feelings to tell us where to go, others, like to purely consult the stats and use logic and rationale to make their decisions, or you might use a blend of both.
As the name and featured picture suggest, this probably favours the stat geeks out there, but even for those that do consult the ‘gut-feeling’, what happens if you just don’t have a strong feeling either way?
Well I believe this can help when faced with that scenario.
So how does it work?
I understand and accept that this metric is designed by myself and the stats and parameters that I have selected to use as such, are subject to my own personal interpretation of what I believe to be the most relevant in assessing a captain candidate.
Now that’s out the way, let’s get to it…
Basically, we’ll take the 3 key captain candidates that we identify in our poll, sometimes 4 depending on the week, for example, for Gameweek 1 it would have been Salah, Agüero and Zaha (usually it’s 3), and we cross reference certain relevant statistics in a 3-way battle and whoever has the best ‘Metric score’, wins.
Metric score is calculated using a simple key that we have provided in each screenshot of the results of our metric. If a player has the highest value in a certain statistic, then it will be highlighted in green, if a player comes second out of the 3 for that statistic, then it will be white, if a player has the lowest value for a certain statistic, then it’s red.
Green = 3 points
White = 1 point
Red = 0 points
We add up the points and whoever has the highest total, is the best captaincy option according to our top 3 selection and the statistics we use.
In special cases where there is 4 viable captain options, it will be a 4-way battle with the points being distributed as follows:
Green = 3 points
Blue – 2 points
White = 1 point
Red = 0 points
Which stats/factors are used?
So think about it yourself, what factors come into play when assessing your captain candidates?
I’d imagine the first 3 everyone came up with, but we have another couple that we feel, statistically, are relevant;
- Player Form – defined as how many points the player has returned in the last 5 gameweeks where the player has been involved in the match.
- Team Form – defined as how many ‘big chances’ (where big chance is defined by OPTA as a chance in which the player is expected to score) the team a candidate plays for has created over the last 5 gameweeks.
- Fixture Difficulty – defined as how many ‘big chances’ the team a candidate is playing against has conceded over the last 5 gameweeks.
- Anytime Goalscorer Odds – defined as the likelihood of the player to score based on the calculated odds of bookmakers.
- xGI (expected goal involvement) – defined as the calculated expected goal involvement of the candidate over the last 5 gameweeks.
- Home/Away Goal Conversion – defined as the goal conversion rate for a player at home or away from home depending on where the fixture takes place for a candidate.
- Reliability % – defined as the % the player has returned (goals/assists) per game. For example, Salah returned goals/assists in 23 games out of the 38 he played in the Premier League last season, meaning he was 60.5% reliable to return you a goal/assist per game.
- Explosivity % – defined as the % the player has returned 10pts or more (double figures) per game. For example, Salah hit 8 double-figure hauls last season, so was 21.05% likely to return a double-figure haul per game.
Of course there are other variables such as, how nailed-on to start the player is, set-piece duties, or the fact he’s a midfielder and gets an extra point for scoring a goal as well as clean sheet bonus, or even extreme circumstances like when we knew Stoke had no defence against Chelsea a couple of seasons back etc, but these things are subjective, and are extremely difficult to quantify and this is purely a statistical aid.
Typically speaking, over the course of a season, those 8 variables should give us a solid spine of data to consult week to week in order to help make our decisions.