El Gato
Villarato!
Impeccable understanding of a metric on display.
Premier League 2018/19
- Thread starter El Flaco
- Start date
Impeccable understanding of a metric on display.
Nice new phrase for "I can't come up with a proper counter-argument so I'll just resort to subtle insults".
I didn't say they were wrong. I said they were "flawed". You're analyzing predictive stats in isolation. Tell me. Why does xG data not translate over to real world results a lot of the time?
El Gato
Villarato!
Subtle? Nothing subtle about it.
You said something about it being flawed, got challenged to demonstrate why it is flawed, responded by making a vague statement about 'lots of external factors' which is irrelevant to the question you got challenged on and you blabber out something about a counter-argument, only to finish off by asking me to show you how it works after you've stated it is flawed!
There is no argument thus far. I'm not arguing anything. I'm waiting for you to present a shred of understanding of the metric and tell me how it is flawed like you said it is.
You said something about it being flawed, got challenged to demonstrate why it is flawed, responded by making a vague statement about 'lots of external factors' which is irrelevant to the question you got challenged on and you blabber out something about a counter-argument, only to finish off by asking me to show you how it works after you've stated it is flawed!
There is no argument thus far. I'm not arguing anything. I'm waiting for you to present a shred of understanding of the metric and tell me how it is flawed like you said it is.
The Observer
New member
Xstats are not bad. It doesn't factor in the psychological aspects though. According to xstats we would have lost 1-2 at Anfield for example. When you shit the bed though even a corner turns into a penalty against you. On the other hand, when it's your moment, every half chance goes in. Liverpool have their moment now.
khaled_a_d
Senior Member
Can you explain why it is reliable? I mean that should be the way not the opposite.
Subtle? Nothing subtle about it.
You said something about it being flawed, got challenged to demonstrate why it is flawed, responded by making a vague statement about 'lots of external factors' which is irrelevant to the question you got challenged on and you blabber out something about a counter-argument, only to finish off by asking me to show you how it works after you've stated it is flawed!
There is no argument thus far. I'm not arguing anything. I'm waiting for you to present a shred of understanding of the metric and tell me how it is flawed like you said it is.
Put in some more smiley faces instead of answering my question. That really helps your case.
El Gato
Villarato!
Reliable for what? What should be the way? Someone coming to you to spoonfeed you the content on how a metric works? Explanations are out there a couple of clicks away.
I'm not the one that made the statement something's 'flawed' underhandedly stripping it of value utterly without understanding and justification. This was my issue with the poster who continues to make a clown of himself by ignoring the simple question - "what exactly is flawed about it".
There are plenty of resources telling you exactly what xG is good for, everyone has Youtube, if you want, go do a bit of digging and make your own mind up. If I recall correctly Tifo made a video and/or a podcast about it detailing what it does and doesn't show, and how it ought to be used. My issue isn't with calling the metric shit, many stats are, particularly when misused - it's the doing so without knowing what it shows. Which this guy clearly doesn't. But instead of asking "how is xG actually good/bad" or requesting sources for this kind of information like a normal person without knowledge, he made an unsupported statement and then doubles down on his stance instead of admitting he made a dumb claim. And I'm not here to do the homework for him.
You really do love your strawman arguments, don't you? I said "Expected goals don't count for shit in a sport where many many external factors can influence the actual outcome of each game". Why did you think I said that? What was the first post I was replying to? Hmm? It was about randomness nullifying the xG metric of City's expected points tally this season. What was my post about that? "xG metrics are incredibly flawed. Using them, United should also be scoring a lot more goals than this. There's a huge difference between predictions made from cherry-picked stats and reality. Fans of the clubs do it to make themselves feel better about their team's poor results.". I was explaining why "expected <insert category>" does not explain real world outcome as a whole. I didn't say the metrics were false full stop. You took one sentence from my post out of context and decided to ignore everything else. And BTW, if you've seen enough Tifo videos, you'd also know that the disparity between the "expected" metrics and real world results are often explained then and there as well. But yes, keep playing the ostrich to save face. Here, I'll even insert extra smileys for you.
Jenks
Senior Member
All the xG metric tells you is how many goals a team should score based on the quality of chances they create. A disparity in the goals they're expected to score/concede and the ones they actually do doesn't suggest randomness. You still have to factor in how clinical your forwards prove to be and whether your defenders and goalkeeper can stop goals in situations where they statistically should.
El Gato
Villarato!
Good. Again, that'd be silly. Because metrics literally can't be 'false'.
Falling back on randomness.. and I'm the one trying to save face? Why don't you prove randomness is the statistical reason for the disparity first? Again - make a claim, provide evidence.
You seem obsessed with smileys too. Might just do the KingLeo approach.
Good. Again, that'd be silly. Because metrics literally can't be 'false'.
Falling back on randomness.. and I'm the one trying to save face? Why don't you prove randomness is the statistical reason for the disparity first? Again - make a claim, provide evidence.
You seem obsessed with smileys too. Might just do the KingLeo approach.
Putting words in my mouth again, eh? I was replying to a post that said randomness affected xG metrics for City. You didn't bother reading the original post carefully. No surprise there. I was saying that there are external factors affecting the outcome of results, and therefore, relying solely on xG metrics is "flawed" when analyzing actual results. Why don't you tell me why the actual table is different from the one showed by the "expected" metric? Hmm? There's your answer. That was the context of the conversation before you decided to cut in and reveal your amazing knowledge of xG metrics.
FinBarcelonafan
Well-known member
It's not 100% accurate but I think those standings match my eye test. It's good indicator not perfect.
El Gato
Villarato!
Nope. You didn't say "relying solely on xG metrics is flawed", that implies a fault of the user, which is correct. You said "xG metrics are incredibly flawed". Here, just to remind you since you seem to have forgotten (or you just don't care how imprecise you're being):
Nor did I reveal any knowledge of the xG. I only asked you how the metric itself is flawed. That you still did not explain. I'd hope you got that already from Jenks et al who gave you the explanation on the silver platter, but then you say this:
.. which shows you clearly still don't really understand what the number does and doesn't show and where it should or shouldn't be used, while expecting somebody else to tell you.
Give yo head a wobble buddy.
KingLeo10
Senior Member
You seem obsessed with smileys too. Might just do the KingLeo approach.
I'm keeping this in mind if we win the 18th Clasico, dawg.
Home of Barca Fans
