Match Thread: vs Portsmouth (a)

[richinspain]

9 minutes ago, DRBee said:

This is not down to "sampling variability" . It is the system not conforming to the model, not due to chance, but due to imperfections in the model being used.

 

 

And there was me thinking that it was because last year we were rubbish and that this year we aren't.

[FlyBritishMidland]

1 hour ago, ap04 said:

No, what I'm saying is that I can use the metric I want for Derby but have no time or interest to watch 11 other games a week so I have to go by the next best (models).

 

It works both ways though, remember at Northampton everyone saying how we were a disgrace and deserved nothing from the game, guess what that was nonsense too.

First point, what you’re saying is you’ll go by commonly used statistics for the rest of the league but make your own up for Derby’s matches to suit what outcome you’re trying to prove.  So you’ve proved my original point.

Second point, we lost against Northampton.  I get criticism when we’ve lost and not played particularly well.  You seem intent on making things up to prove we shouldn’t have had a positive outcome when we have.

[DavesaRam]

Shakespeare got it about right when he said "Xg or not Xg? Stat is the question"!

[Caerphilly Ram]

1 hour ago, Ghost of Clough said:

Anyone thinking a goal should count as an xG of 1.0, "becuase it was a goal" doesn't havea firm grasp of statistics. 

Imagine flipping a coin. It has an xH (expected heads) of 0.5 and xT (expected tails) of 0.5. This particular poster thinks after flipping the coin and it landing on heads, the probability should be counted as landing on heads 100% of the time.

The rest of what the poster says isn't much better than that

I’ve conceded defeat in that particular exchange. If someone is repeatedly more interested in arguing as to why we don’t deserve to be second in the league than enjoying the fact we are that’s their issue and I’m content to leave them to it.

 

[ap04]

2 hours ago, Srg said:

You use xG to prove when we are crap, and when it doesn't, you say it isn't good and invoke 'sampling variance' instead.

I never look at xG when I've watched the game(s) as that would be idiotic so the above is simply not true (or if it is it's only a coincidence).

But sorry if you think that's boring or off-putting, I'll try and post less - not that easy when everyone disagrees eh

 

2 hours ago, Ghost of Clough said:

Anyone thinking a goal should count as an xG of 1.0, "becuase it was a goal" doesn't havea firm grasp of statistics.

Except I never said that did I? I said it's not the perfect analysis tool because it will never do that.

 

1 hour ago, Chester40 said:

So if you score 2 early goals -  from a goal keeping error from a corner and a deflected shot from distance,  and you spend the rest of the game defending behind the ball,  launching it out, running down the clock, letting the keeper make 5 or 6 routine-ish saves.

No fan would think you had been lucky ...but the stats would suggest you were?

The luck element would be getting rewarded with 2 goals from only 2 clear chances (that would normally take 7.4), which then allowed you to sit back. But the win would be just about fair as long as the opposition created nothing afterwards.

 

1 hour ago, richinspain said:

And there was me thinking that it was because last year we were rubbish and that this year we aren't.

It all points to: Slightly worse this year (remaining schedule caveat) - lucky this year (with points/results not necessarily position) - unlucky last year - one maybe two stronger rivals last year.

 

30 minutes ago, Caerphilly Ram said:

If someone is repeatedly more interested in arguing as to why we don’t deserve to be second in the league

Another thing I haven't said (yet)

[ap04]

1 hour ago, DRBee said:

Analyses show that some teams show great discrepancy in  their  goals scored compared to the rating of their goal scoring chances and it can happen over lengthy periods of time. This is not down to "sampling variability" . It is the system not conforming to the model, not due to chance, but due to imperfections in the model being used.

Goals are randomly distirbuted (Poisson). The number of goals per near-goal events (the accurate measure of quality in an ideal model, NOT chances) is in the long run the same for every team, Derby or Man City, being off that in any given period is chance. The discrepancies you mention are mainly not due to the imperfections of the models (and they are imperfect) but the even bigger imperfections of the goal system as a metric (binary, most events ignored, no specificity).

A simpler way to show this: Assume two equal teams, playing at exactly the same level all the time. If goals weren't random they would draw every single time. More often than not they won't.

[Srg]

Really enjoy reading a thesis of why 1 person is right and the rest of the fanbase isn’t 😂

[FlyBritishMidland]

1 hour ago, ap04 said:

Another thing I haven't said (yet)

We’ve got that to look forward to?  Can’t wait!!

[Crewton]

8 minutes ago, FlyBritishMidland said:

We’ve got that to look forward to?  Can’t wait!!

I was hoping it might be "goodbye".

[PuP]

1 hour ago, Caerphilly Ram said:

I’ve conceded defeat in that particular exchange. If someone is repeatedly more interested in arguing as to why we don’t deserve to be second in the league than enjoying the fact we are that’s their issue and I’m content to leave them to it.

 

I am old school I'm afraid. I think that where any team is in the league at this stage of the season is the real accurate description of where they should be. A team can be top after 10 games and be in a false position, like us last season, but to maintain that over the season is the real measure. So if your team is top end of the league at this stage its because you deserve to be there. Conversely those at the bottom prove they are the worst teams. I think the only statisric that I pay much attention to is attempts on target, and the actual goals scored obviously. 

[Comrade 86]

3 minutes ago, PuP said:

I am old school I'm afraid. I think that where any team is in the league at this stage of the season is the real accurate description of where they should be. A team can be top after 10 games and be in a false position, like us last season, but to maintain that over the season is the real measure. So if your team is top end of the league at this stage its because you deserve to be there. Conversely those at the bottom prove they are the worst teams. I think the only statisric that I pay much attention to is attempts on target, and the actual goals scored obviously. 

xG indicates trends not outcomes IMO and some of what he says is actually at least partially correct. The problem with matey boy is he says one thing one week, then the opposite the nest and thinks nobody notices. 100% not a Ram in any case IMO and even if he were, he'd be one that trolls his fellow supporters, so either way, he's just another for the 'ignore' list. 

[McArthur Park]

3 hours ago, ap04 said:

Goals are randomly distirbuted (Poisson). The number of goals per near-goal events (the accurate measure of quality in an ideal model, NOT chances) is in the long run the same for every team, Derby or Man City, being off that in any given period is chance. The discrepancies you mention are mainly not due to the imperfections of the models (and they are imperfect) but the even bigger imperfections of the goal system as a metric (binary, most events ignored, no specificity).

A simpler way to show this: Assume two equal teams, playing at exactly the same level all the time. If goals weren't random they would draw every single time. More often than not they won't.

Is it just me, or does anybody else have no idea what any of this means? 

[Caerphilly Ram]

2 minutes ago, McArthur Park said:

Is it just me, or does anybody else have no idea what any of this means? 

I know that “Poisson” is French for fish 🤷🏻‍♂️ 

[Foreveram]

3 minutes ago, McArthur Park said:

Is it just me, or does anybody else have no idea what any of this means? 

Or care🤷🏻‍♂️

[Premier ram]

4 hours ago, ap04 said:

Goals are randomly distirbuted (Poisson). The number of goals per near-goal events (the accurate measure of quality in an ideal model, NOT chances) is in the long run the same for every team, Derby or Man City, being off that in any given period is chance. The discrepancies you mention are mainly not due to the imperfections of the models (and they are imperfect) but the even bigger imperfections of the goal system as a metric (binary, most events ignored, no specificity).

A simpler way to show this: Assume two equal teams, playing at exactly the same level all the time. If goals weren't random they would draw every single time. More often than not they won't.

 

                           

[kevinhectoring]

7 hours ago, ap04 said:

A simpler way to show this: Assume two equal teams, playing at exactly the same level all the time. If goals weren't random they would draw every single time. More often than not they won't.

Your use of language doesn’t help you.  I suspect you are saying not that goals are random; but that chance plays a part in whether or not a goal is scored from a particular situation. 

True of course.  But chance ultimately obeys the laws of probability. So of course your two hypothetical teams won’t draw all the time. But if they play an infinite number of times, each is likely to score the same number of goals 

 

Edited April 4 by kevinhectoring

[Archied]

7 hours ago, kevinhectoring said:

Your use of language doesn’t help you.  I suspect you are saying not that goals are random; but that chance plays a part in whether or not a goal is scored from a particular situation. 

True of course.  But chance ultimately obeys the laws of probability. So of course your two hypothetical teams won’t draw all the time. But if they play an infinite number of times, each is likely to score the same number of goals 

 

Thanks , you’ve cleared my mind for me with chance following the laws of probability,,, 

if derby win/ do well.    Chances are we will probably have wums appear on the match day thread👍

[Comrade 86]

12 hours ago, Caerphilly Ram said:

I know that “Poisson” is French for fish 🤷🏻‍♂️ 

A Poisson distribution is a discrete probability distribution. It gives the probability of an event happening a certain number of times (k) within a given interval of time or space. The Poisson distribution has only one parameter, λ (lambda), which is the mean number of events.

No need to thank me 🤪

[The Scarlet Pimpernel]

11 minutes ago, Comrade 86 said:

A Poisson distribution is a discrete probability distribution. It gives the probability of an event happening a certain number of times (k) within a given interval of time or space. The Poisson distribution has only one parameter, λ (lambda), which is the mean number of events.

No need to thank me 🤪

Yeah but why is it called "poisson" that's what I need to know.... 

[MR-RAM]

ap04 is to statistics what Dryhorce is to transfers………

or could they be the same person? 🤔