Compared with last season...

All advertisments are hidden for logged in members, why not log in/register?


is Mr Grafikhaus related to Lewis or Duckworth

football is simple , 2 points per game means going up

Apart from last season, where 92 points would have still had us finishing third. 81 would have been more effective, it was enough for Udders. 93 was what we really needed......
 
The problem is that the X axis labels do not begin at 0.

Grafikhuas sent me his spreadsheet earlier, and i present it here to you in what i believe to be the correct from (This is after a a few whiskys though....:))




Screen Shot 2012-10-25 at 01.10.18.png
 
The problem is that the X axis labels do not begin at 0.

Grafikhuas sent me his spreadsheet earlier, and i present it here to you in what i believe to be the correct from (This is after a a few whiskys though....:))




View attachment 5359

Did the FA feel sorry for us and donate 2 points for the "trend line" at the start of the season?:)
 
OK, Iwas on the wife beater last night, but surely he trendline is a regression line, a line of best fit. It's aim is to get the minimum error between itself and the data and then extrapolate forward, not take the average and extrapolate forward. 2 different methods hence 2 different results?

UTB
 
OK, Iwas on the wife beater last night, but surely he trendline is a regression line, a line of best fit. It's aim is to get the minimum error between itself and the data and then extrapolate forward, not take the average and extrapolate forward. 2 different methods hence 2 different results?

UTB

Can I just suggest that this is all very simple.

To get the "trend line" all you have to do is divide points gained by games played and multiply by 46 - that will tell you how many points you will get at the end of the season if the trend continues.

This wholel thread is a wonderful example of the ability of technology to make a straightfoward thing very complicated....
 
Surely the graph should start at zero points and zero games this one starts at zero points but at game #1

No it shouldn't/doesn't. The last version (as Post #86 and below) shows at Match 1, we'd got 3 points. We know we had null points after no games were played, like we had 90 points at Game 46 last season. What would be the point of saying we had 90 points after a (non-existant) 47 games?

is Mr Grafikhaus related to Lewis or Duckworth

football is simple , 2 points per game means going up

Yes, but this model illustrates progress compared to last season. For example we are - unbelievably - ahead of last season at the moment. But look what happened between games 17 and 23 last season - 18 points in six games.

Let's see how we compare by the turn of the year.

Oh, and thanks for your input, Highbury.

win14.gif
 
Can I just suggest that this is all very simple.

To get the "trend line" all you have to do is divide points gained by games played and multiply by 46 - that will tell you how many points you will get at the end of the season if the trend continues.

This wholel thread is a wonderful example of the ability of technology to make a straightfoward thing very complicated....

Forecasting often uses things like three or four point moving averages to get a trend - smoothing out the spikes in a short set of data - so if you are really into making things a little less simple then have a look at that.
 
Can I just suggest that this is all very simple.

To get the "trend line" all you have to do is divide points gained by games played and multiply by 46 - that will tell you how many points you will get at the end of the season if the trend continues.

This wholel thread is a wonderful example of the ability of technology to make a straightfoward thing very complicated....


I totally agree. But there seems to be a fair bit of debate going into why the 2 things are different, when they're measuring 2 different things.........Simples.

:)

UTB
 

That's because the games played figures are represented by the lines not the columns: games played are absolutes. There should be 46 vertical lines.
Quite simple really, if you are building a fence with 46 panels you need 47 fence poles trust a university graduate to make a program that complicates something a 5yr old could work out in his head.
 
Quite simple really, if you are building a fence with 46 panels you need 47 fence poles trust a university graduate to make a program that complicates something a 5yr old could work out in his head.

The poles are the games. On grafikhaus' chart we've either played a game or we haven't.

The panels are immaterial...a bit like my status as a graduate in this case.
 
Oh my christ football just got far more complicated than it really needed to be. Heres one for you that might straighten things out:-

Get more points than anyone else then we get promoted.
Get 2nd most points and we get promoted.
Nothing else matters because if we finish outside that we will fail.
 
Get more points than anyone else then we get promoted.
Get 2nd most points and we get promoted.
Nothing else matters because if we finish outside that we will fail.

Are you seriously suggesting that, should we finish outside the top two places, you are less than confident we will be victorious in the play offs? :eek:
 
Are you seriously suggesting that, should we finish outside the top two places, you are less than confident we will be victorious in the play offs? :eek:

We have been in 4 play off finals and lost all of them. Assuming a 50-50 chance of winning each one, the chances of losing all 4 are 1 in 16. That must mean we will win the next one.... doesn't it?
 
Show it in a pie chart.

Here you go

World-Most-Accurate-Pie-Chart.jpg
 

All advertisments are hidden for logged in members, why not log in/register?

All advertisments are hidden for logged in members, why not log in/register?

Back
Top Bottom