Time For A Rethink In Rolling Averages?

[Brooksy 120]

If you read my post, the average, as a function, is set in stone.

What I am saying is that the more measuremets you take, the more accurate an average is.

 

Now, this little example will demonstrate the point, and the one you didn't grasp a couple of weeks back:

 

Let's take two measurements: 5.8, 7.6.

Average = 6.7

 

Let's take 10 measurements: 5.8, 7.6, 5.8, 7.6, 5.8, 7.6, 5.8, 7.6.

Average = 7.2

 

So, as you can see, the number of measurements (or rides, in this case) is crucial.

 

 

This point is to prove that 97.98274928374% of all statistics are made up.

 

Where does the 7.2 come from?? I still make it 6.7.

 

 

Sorry, just worked out your last line. Thought you were trying to make a serious point. Should've known better.

Edited September 13, 2012 by Barney Rebel

[Fromafar 10,360]

If you read my post, the average, as a function, is set in stone.

What I am saying is that the more measuremets you take, the more accurate an average is.

 

Now, this little example will demonstrate the point, and the one you didn't grasp a couple of weeks back:

 

Let's take two measurements: 5.8, 7.6.

Average = 6.7

 

Let's take 10 measurements: 5.8, 7.6, 5.8, 7.6, 5.8, 7.6, 5.8, 7.6.

Average = 7.2

 

So, as you can see, the number of measurements (or rides, in this case) is crucial.

 

 

This point is to prove that 97.98274928374% of all statistics are made up.

I can only see 8 measurements, is this a mistake in my counting or am I still missing the point you are trying to make. Still an average is an average

[dontforgetthefueltapsbruv 10,706]

 

 

 

One thing everybody misses about rolling averages is that the longer they are used, and the more meetings a rider completes, the more accurate they become.

Surprised this has not been mentioned.

Not completely the case thats why.

 

If the scores from the start of the data collection are old they then become irrelevant to the assessment of the result.

 

For rider averages the aim is to gauge form/ability.

 

If you have a larger sample size you begin to include irrelevant data - ie scores from 2 years ago.

[Fromafar 10,360]

Where does the 7.2 come from?? I still make it 6.7.

 

 

Sorry, just worked out your last line. Thought you were trying to make a serious point. Should've known better.

Where does the 7.2 come from?? I still make it 6.7.

 

 

Sorry, just worked out your last line. Thought you were trying to make a serious point. Should've known better.

Nobody takes him seriously ,he is tap dancing all the same

[bigeddiechek 1,094]

Not completely the case thats why.

 

If the scores from the start of the data collection are old they then become irrelevant to the assessment of the result.

 

For rider averages the aim is to gauge form/ability.

 

If you have a larger sample size you begin to include irrelevant data - ie scores from 2 years ago.

 

Yes and no, for example, a smaller data set may include outlying data due to something like mechanical problems.

Actually one of the uses of an average is to "smooth" out that type of data, but it only really becomes useful as the data set increases.

As you mention, gauging of form is time sensitive, therefore it makes sense that your data is collected across and apposite time-scale.

 

Also, as the data set increases, data from several years ago will have less of an impact on the current calculation.

 

With respect to Jim and Barney, I was maybe a bit cheeky, but my point is that average is not really a very helpful statistic: in any case (or at least that I can think of).

Jim, it is clear that you didn't listen to me as I said 5 times (with witness) that my point is that you aren't comparing apples with apples.

 

You did, however, say several times that an average is an average.

 

The point you were making is that you thought this system is better: and I agreed, just for different reasons.

 

Here is a final - serious - question.

 

Would it be more meaningful to have a league average that is calculated from all riders, and all rides: then each individual is given a standard deviation rather than an average?

[Fromafar 10,360]

Yes and no, for example, a smaller data set may include outlying data due to something like mechanical problems.

Actually one of the uses of an average is to "smooth" out that type of data, but it only really becomes useful as the data set increases.

As you mention, gauging of form is time sensitive, therefore it makes sense that your data is collected across and apposite time-scale.

 

Also, as the data set increases, data from several years ago will have less of an impact on the current calculation.

 

With respect to Jim and Barney, I was maybe a bit cheeky, but my point is that average is not really a very helpful statistic: in any case (or at least that I can think of).

Jim, it is clear that you didn't listen to me as I said 5 times (with witness) that my point is that you aren't comparing apples with apples.

 

You did, however, say several times that an average is an average.

 

The point you were making is that you thought this system is better: and I agreed, just for different reasons.

 

Here is a final - serious - question.

 

Would it be more meaningful to have a league average that is calculated from all riders, and all rides: then each individual is given a standard deviation rather than an average?

Lets clear one thing up Big Ed ,the point I was making a couple of weeks ago was the the averages are unfair regarding guests IMO where perhaps the currentt season average should be applied rather than the rolling average(the example I gave was Craig Cook who is averaging around 10pts THIS season but due to rolling averages Edinburgh could only get a guest around 8.5 silghtly unfair IMO,my second point was that Dyer and Aspregen were still on good rolling averages as they had not yet rode 38 matches and I considered them to be a good signing with that point in mind).YOU were the one that said an average is an average and were mocking my opinion(no problem there all in a Saturday nights entertainment) listen to the whole story before trying to shoot me down.Better go and dig up some more bait now .

[bigeddiechek 1,094]

Lets clear one thing up Big Ed ,the point I was making a couple of weeks ago was the the averages are unfair regarding guests IMO where perhaps the currentt season average should be applied rather than the rolling average(the example I gave was Craig Cook who is averaging around 10pts THIS season but due to rolling averages Edinburgh could only get a guest around 8.5 silghtly unfair IMO,my second point was that Dyer and Aspregen were still on good rolling averages as they had not yet rode 38 matches and I considered them to be a good signing with that point in mind).YOU were the one that said an average is an average and were mocking my opinion(no problem there all in a Saturday nights entertainment) listen to the whole story before trying to shoot me down.Better go and dig up some more bait now .

 

I was never mocking your opinion, if you remember I was agreeing with you, in general. I was saying you can't compare to completely different methods though.

Your point does prove the nature of the calculation being time-sensitive when one uses it as a measure of form.

 

What do you think about my idea of using standard deviation as a measure?

 

 

 

I'd never, ever mock you.

I enjoy our blethers.

Edited September 13, 2012 by bigeddiechek

[Fromafar 10,360]

I was never mocking your opinion, if you remember I was agreeing with you, in general. I was saying you can't compare to completely different methods though.

Your point does prove the nature of the calculation being time-sensitive when one uses it as a measure of form.

 

What do you think about my idea of using standard deviation as a measure?

 

 

 

I'd never, ever mock you.

I enjoy our blethers.

I will believe you regarding mocking (i suppose).

I think your idea is a wind-up Your brain is still frazzled from all these average calculations.

More important issue is will the track be as slick as Ipswich's last visit.If so what will the result be.Although injuries and sackings mean it is not the same riders on show.

[bigeddiechek 1,094]

The idea of using standard deviation isn't a wind-up, I can assure you.

 

I'm interested to here what others think, even if it is to shoot me down.

There must be some statisticians on here who can give us an informed opinion.

Edited September 13, 2012 by bigeddiechek

[Fromafar 10,360]

The idea of using standard deviation isn't a wind-up, I can assure you.

 

I'm interested to here what others think, even if it is to shoot me down.

There must be some statisticians on here who can give us an informed opinion.

I wouldn't hang by your thumbs waiting for a reply on this one .

[dontforgetthefueltapsbruv 10,706]

 

 

What do you think about my idea of using standard deviation as a measure?

 

 

 

Presumably with all teams building to 0 the next season although some tolerance would be needed or it would be impossible.

 

This would require no change in team numbers and for all riders to be the same from one season to the next to work in its purest form.

 

Any deviation in either factor would render the figure inaccurate.

[bigeddiechek 1,094]

Presumably with all teams building to 0 the next season although some tolerance would be needed or it would be impossible.

 

This would require no change in team numbers and for all riders to be the same from one season to the next to work in its purest form.

 

Any deviation in either factor would render the figure inaccurate.

 

OK, so you are saying there would need to be the same number of teams, and the same number of riders, for it to be meaningful.

Actually, I think you are also saying it must be the same pool of riders too.

 

Would it actually need to be the case though? I'm sure you know there are variations in ways of calculating the standard deviation for a sample.

I'm not sure about the answer, though, and am open to words of wisdom greater than mine.

 

You wouldn't necessarily need to have team building set to 0.

For example, you could have limitations build around deviation from the mean.

 

I think it might be quite interesting, if nothing else.

[dontforgetthefueltapsbruv 10,706]

 

 

OK, so you are saying there would need to be the same number of teams, and the same number of riders, for it to be meaningful.

Actually, I think you are also saying it must be the same pool of riders too.

 

Would it actually need to be the case though? I'm sure you know there are variations in ways of calculating the standard deviation for a sample.

I'm not sure about the answer, though, and am open to words of wisdom greater than mine.

 

You wouldn't necessarily need to have team building set to 0.

For example, you could have limitations build around deviation from the mean.

 

I think it might be quite interesting, if nothing else.

I agree it is an interesting concept but is probably too complicted to be considered for implementation.

[bigeddiechek 1,094]

I agree it is an interesting concept but is probably too complicted to be considered for implementation.

 

There are so many uncontrollable variables that would totally knacker it, but that is what happens with the current system, I suppose.

It might make it a lot more difficult to manipulate rider ratings for team building.

 

In terms of the public, then it could be presented as below average, average, above average, and maybe extreme deviant?

[dontforgetthefueltapsbruv 10,706]

Extreme deviant?...are you still talking statistics or have you been speaking to some of the women I used to know!!!