Read on Twitter
#FPL Some statistical and scientific bias to challenge the point chasing moves.
Football games are highly highly variable events. In scientific terms we would describe it as destructive testing: after measurement of an event to gather data, the event no longer exists.
Football games are highly highly variable events. In scientific terms we would describe it as destructive testing: after measurement of an event to gather data, the event no longer exists.
This varies from non-destructive data such as let’s say voting patterns.
If the same game of football was to be played out 10 times over 10 days with the same people, you would get 10 different events with a range of results, goalscorers, clean sheets etc.
If the same game of football was to be played out 10 times over 10 days with the same people, you would get 10 different events with a range of results, goalscorers, clean sheets etc.
If you were to perform an election poll over 10 days with the same people the results would be much more comparable, or ‘stable and predictable’. We call it repeatable: you can measure the same thing 10 times and get the same result
There are less factors of variation involved which is the key difference.
So how does this affect FPL? The key point here with the 10 football games is still that there will be some general and underlying patterns. These could be due to form, ability, tactics, etc.
So how does this affect FPL? The key point here with the 10 football games is still that there will be some general and underlying patterns. These could be due to form, ability, tactics, etc.
When you pick a player, you are looking at these underlying stats. The risk is that the many events that make up the underlying stats are affected by the variation. One game 10 shots could result in 5 on target but no goals, and the next game it could be 3 on target and 3 goals.
Take a same shot at goal for example 10 times: there is the cleanliness of the strike of the ball, the position of the GK, the placement of the ball at the time of stone, possibly the position of defenders and so on.
There is more variation in each of these single events (shots, tackles, passes, decisions, fitness) than the variation in the predictable aspects like form, tactics, confidence, talent, and so on.
Over time the sheer quantity of the single events makes the overall result more predictable, but the outcome of the individual events still has the variation.
If you base decisions on underlying trend you will improve your OR in FPL. If you gamble & try to catch the variables...
If you base decisions on underlying trend you will improve your OR in FPL. If you gamble & try to catch the variables...
It becomes risk and reward. The variation shows more often than not you will get it wrong. This is why sometimes the underdog wins, or someone gets a hat trick.
If you gamble each time and hit the right captain/transfer/formation etc then that is the key to winning the game. It’s like a massive game of heads of tails - with 7m+ players there are some people that would correctly predict it week after week.
In summary: Chase the predictable patterns and trends, and not the single variable events.
I hope this makes some kind of sense?!
I hope this makes some kind of sense?!
