Has Kanté’s new Chelsea role improved his Fantasy potential?

The only time I ever use stats is if I’m torn between 2 similar players, other than that it’s all down to the eye test 😀

I feel like following stats constantly kinda takes the fun out the game, who’s picking your players? You? Or an Algarithim

I think xG and xA are over rated stats. There is almost as much randomness in the quality of chances created as there is in the goal being scored.

I've got young two kids and a missus who dislikes football. As the eye test becomes an increasingly less viable option my game has become more stats driven and if I'm honest the difference has been negligible.

Gotta moneyball... I would say there will always be anomalies but overall the underlying statistics prevail. Shakiri was a prime example of the long shooting last year, banging shots from 30 yards every 20 minutes or so... But if they keep flying in the stats are there for all to see. I think what has changed is points in relation to teams, it's not enough to pick a player based on the team performance as much as it was a few years back. i.e. in the past having a Utd/chelsea Def for clean sheet points in years gone by, you could get away with one of and it didn't really matter which, now you have to have the right player because of the level of attacking returns we're seeing in Def in the prem over the last few years. All in all, it'll probably ruin this for everyone, a brief look at stats will tell you must of what you need to know... Without actually ever watching a football game... And that's we're the eye test comes in... Because is that not the entire point!

I love watching as much football as possible and seeing not only a player, but a team performance. I pretty much use the eye test to narrow down my options and then do a little stats research to pick between them. And even then I don’t do stats too much. For instance... looking at stats I would’ve sold Mané on my wildcard, but having watched him and Liverpool each week I’m definitely keeping him.

Heat maps are hugely overrated.

Better stats give naturally more info than before. But margins are also small and I started last year playing first two months without eye test. Result was terrible. Everything changed when I started following PL. In future stats will be better, but now there no stats available (or are there?) about quality of missed shots for example.

For example for me Mané seems to be in form now and more clinical than last year and even though Dilva had decent xG in last game, he also looked dangerous. So I expect both of them to beat their xG:s. And as a hobby it is very nice to see something that all others won't notice. And there are no stats for attitude also. In future stats will give even more info (and they are important already), but I will want to enjoy the fun of watching games and trying to be more clever than the algorithm 😉

I only use stats to compare two players or get some basic information. Watching the player gives a much better idea and the stats help to confirm your choice

- Xg is like the ‘emperors new clothes’ (check how it’s complied 😉 )

I tend to watch a lot of games and eye test is therefore one important thing, but i also enjoy watching football in general so to only watch from a FPL perspective would be limiting my joy. But i try to do it in some degree, and it´s useful. I think i did the exactly same thing with Coutinho, no returns but if you saw the games it was inevitably going to be returns if he kept doing and he was eventually. Stats and numbers could be important but they could also be misleading, in for example, a player have many touches in the box doesn´t always tell the whole picture.

Eye test is the most important thing. And the only that matters. Someone who’s watching the games (esspecialy if he used to play football) will always know more than stats-guy.
Football is not a sport where stats is important. If you like numbers try with baseball, tennis, basketball...

https://i.imgur.com/oJ59jnL.gif

My wife managed 2 top 10k finishes in a row with very little help from me.

She did two things...

She perm-captained the best player; one year it was RVP and the other other year it was Suarez. She didn't care who the opponent was.

She brought in the best player from watching MOTD every week. She wasn't interested in under-lying stats; form; fixtures. If a player from Watford banged in 2 goals. In he came. She brought them in on the Saturday.

The transfer value rocketed. It was 100% the eye-test. It worked, twice.

I play a lot more with statistics in mind and my record is good, but not as good as hers. I enjoy the statistical approach. But there is something to be said for captaining the best player every week and picking the week's best scoring player. However, much we like to call it the casual approach. It can work.

Does 'Eye Test' mean watching football?

Barely watch football. Was trying to avoid it completely this season but was at a friends house when they had the Chelsea-Newcastle match on in the background.

Why use my opinion when I can harvest the opinions of thousands on here? Add in some stats. Job done.

This is a really interesting post on a number of levels WhatTheHeck!. The problem with "the eye test" or any kind of subjective analysis is that it is subject to cognitive bias. For example, how many times do you hear on commentary something along the lines of "if a Brazillian did that we'd be going mad over how brilliant it is"? And there's something to this. I listen to commentators go on about the Brazil team and roll my eyes. They already have in their heads visions of Pele's Brazil in 1970 and ignore that Brazil became more functional and European in style back in the early 90s. Even then, you watch one of their players perform endless step overs and go nowhere and know that if an English player did that, the MOTD team would be lambasting him for doing nothing with the ball. We are all subject to these biases which is why I never trust "the eye test", especially when players good on the eye don't always produce FPL returns.

On the flip side, there is a reason that so many have tried and failed to use "Money Ball" statistical analysis in fluid sports like football and NFL (Damien Comolli at Liverpool with the backing of FSG, who embraced Billy Beane's methods at Boston after Beane's success with Oakland A's comes to mind). It's because the statistical technology is not there and useful applied conclusions not readily available like with static games such as baseball. The limits of statistical analysis is one reason I get irritated by sweeping conclusions on here by people "analysing" by taking conclusions from previous games etc when variables are all so different and each "trial" not independent thus invalidating any conclusion.

I think, personally, you have to combine both together. It's the same with the stuff like XG Delta. You never hear it on here but there are always going to be players who have a positive delta i.e. they score more than XG says they should without having to "mean revert". XG is the "average" player and a number derived from all shots taken from a similar position, good and bad shooters. A Sergio Aguero or Harry Kane is not an average player and you would expect them to outperform the average i.e. XG. But you only really know that from looking at them. But a Richarlison, you know from watching games going back over a whole season, he's not so great at shooting so his positive XG Delta right now probably will revert.

'Eye test' over stats any day for me but...You need a lot of spare time to watch enough football to trust the eye test. I tend to have the time so I rarely use stats!
I can totally understand the alternative though as watching the goal highlights on MOTD isn't very useful in isolation.

Think of it this way: there should be a range of characteristics that predict how well each player will perform relative to each other. Stats aim to capture these characteristics, from which you should be able to pick all the best players to own (price determines feasible combinations).

If you had access to perfect data and were trained in sophisticated statistical analysis, then you could optimise your team quantitatively. But few people will have a powerful model for predicting points, the data probably has errors and it's unlikely that all characteristics are captured in the stats. The eye test can try to capture missing elements and you would have to adjust your stats analysis to reflect that. I'm sure the bookies won't be putting their fingers in the air when they set their odds!!!

So put simply, I think stats and the eye test can help, but if you want to do it well you need to understand and manipulate data better than other people.

Yesterday I started doing analysis of previous seasons to see which stats predict points. The next thing i want to analyse is the goal threat of skilful creators like Hazard and Silva, who score a gair ampunt of goald when they play but don't look good in the (non-creative) numbers. By the end of the season I should know if I should stick to the eye test!!

Eye test > all.

There are too many people around who base there decisions solely on statistical analysis but they can be construed in many ways. Stats can clearly assist and the xG is very interesting, but all of last season we were told Mee or Duffy were bound to score because of the amount of headed chances they were getting or Richarlison was shooting 20 times a game that he would eventually rediscover his form. Didn't turn out great in the end.

For me, stats can help if it's between two players are there is not much to separate them generally (eye test, fixtures, price, form) but generally I will base my decisions from actually watching a player and making a verdict.

For me, the eye test is fine, but who has the time to watch all of the games? You've got to go to the stats so you don't miss players. Otherwise you're likely to be overly influenced by what you've seen. But yes, watching definitely helps, as others have mentioned. seeing the confidence/composure of a player you can often tell whether they are lucky (e.g. shots look wayward, they just happen to be on target) or unlucky (getting in good positions, hitting the woodwork or foiled by top class saves).

Just like to add that Simon March won FPL a few yeras back while in Singapore with virtually zero eye test access. I know this because I was pretty close to him all season.

Personally I like eye test and current form as the keys. But I think eye test for many just means MOTD which is not enough I feel, watching 90 mins in the flesh is a big advantage. Stats are fine but can be used in too many ways to suit opinion for me.