Paul

The reason why strictly following an FPL model is so overpowered this season is not only because we have more ways to score points and some of them are almost unclassifiable by humans, but also because there is a lot of garbage data on social media. FPL models work with pretty basic non-contextualised data, so why are they naturally better than "data-driven" managers? Because models are consistent and unbiased. Social media gives us this fake perception of data being user-friendly and easy for ranking players, when in reality most feeds are biased and require tons of context. And even when we have this context, our room for mistakes is directly proportional to the room for interpretation. The FPL model minimises that risk because even with the most basic data and interpretation, it's still going to be consistent and unbiased, so it simply doesn't have as much room for mistakes. Advanced data requires a lot of work and as you see on your timeline, people mostly use it for simple answers that such data isn't even able to give by its nature. And the moment you see the payoff in points after using any garbage data, outcome bias and fake analytical evidence suck you in and give you this false understanding that the data has a clear correlation with the result you got. Good advanced data requires serious involvement, work, and as sad as it might be to hear, a lot of time. And to see results from working with it in FPL, you need consistency. It is objectively time-consuming and most importantly it requires consistently successful interpretations. You have to be exceptional with your planning every time and be a thousand times more involved than the person strictly following a solver, while constantly battling time, biases, and all the other psychological factors. The model will not. It's not me telling you to completely outsource your FPL team, but it's definitely something you have to keep in perspective when playing, and maybe it's also something that would help you appreciate yourself a bit more in this game.