Calculating handicap index increases/decreases

[Trodski]

Hi,

I understand how index is calculated. Best 8 of last 20 score differentials averaged.

Logically it follows that when a counting score is lost/replaced to calculate the change you do; previous counting score - new counting score / 8.

This season I have calculated the expected change and been out by 0.1 a few times. Am I missing something from the calculation? To give you an example;

Previous differential 13.8 - new differential 7.1 = 6.7. 6.7 / 8 = 0.8375. HI reduction should be 0.8?

Actual reduction was 0.9. Anyone know why?

Not that big an issue but 0.1 can change the number of shots you get in knockouts etc.

 

[Alan Clifford]

I think the problem is the decimal points in the average of the other 7. Try adding up all eight before rounding.

 

[Colin L]

Hi,

I understand how index is calculated. Best 8 of last 20 score differentials averaged.

Logically it follows that when a counting score is lost/replaced to calculate the change you do; previous counting score - new counting score / 8.

As Alan says. In applying your logic, you have overlooked that score differentials are not whole numbers but are rounded to one decimal place. You have to aggregate all 8 counting differentials, divide by 8 and then round up or down to the nearest tenth.

 

[Imurg]

You also have to take into account that PCC might, just might, kick in a spoil your Maths.
You also need to watch weather you've reached your Soft Cap...increases are halved....
It's Witchcraft.....

 

[Voyager EMH]

Score Differentials should retain two decimal places in order for the sum of eight of them, divided by eight is accurate to one decimal place.
Each Score differential has a positive or negative error or tolerance due to rounding to one decimal place.

There must always be, at any given time, a small percentage of players whose HI is "wrong" by one decimal place.
A small percentage of these will be tipped over the borderline to a "wrong" Playing Handicap.

All this could have been greatly avoided if Score Differentials had been kept to two decimal places.

 

[wjemather]

Also, the soft/hard cap may be in play.

 

[Alan Clifford]

Score Differentials should retain two decimal places in order for the sum of eight of them, divided by eight is accurate to one decimal place.
Each Score differential has a positive or negative error or tolerance due to rounding to one decimal place.

There must always be, at any given time, a small percentage of players whose HI is "wrong" by one decimal place.
A small percentage of these will be tipped over the borderline to a "wrong" Playing Handicap.

All this could have been greatly avoided if Score Differentials had been kept to two decimal places.

Interesting. HNA keep the differentials in their database to 4 decimal places. But rule 5.1a states "An 18-hole Score Differential is calculated as follows and rounded to the nearest tenth, with .5 rounded upwards". And, thinking about it, that doesn't even make sense; how can .5 be rounded upwards if you are rounding to the nearest 10th?

 

[rulefan]

Interesting. HNA keep the differentials in their database to 4 decimal places. But rule 5.1a states "An 18-hole Score Differential is calculated as follows and rounded to the nearest tenth, with .5 rounded upwards". And, thinking about it, that doesn't even make sense; how can .5 be rounded upwards if you are rounding to the nearest 10th?

HNA - What they?

 

[D-S]

HNA - What they?

Handicap Networks Africa

 

[PJ87]

I normally just times my index by 8

Then when I know what score diff is dropping off say 25.4 being replaced by a 22.4 I'll take 3 off that number and divide by 8 to give a rough figure of what my index will be the next morning

 

[Voyager EMH]

Interesting. HNA keep the differentials in their database to 4 decimal places. But rule 5.1a states "An 18-hole Score Differential is calculated as follows and rounded to the nearest tenth, with .5 rounded upwards". And, thinking about it, that doesn't even make sense; how can .5 be rounded upwards if you are rounding to the nearest 10th?

It has happened to me twice since the start of WHS in UK.
Here is one of the times. I was given 3.0 but I was 2.9 if two decimal places had been used.

 

[Alan Clifford]

It has happened to me twice since the start of WHS in UK.
Here is one of the times. I was given 3.0 but I was 2.9 if two decimal places had been used.

View attachment 48662

I suppose the world famous statisticians didn't get involved in the arithmetic

 

[Trodski]

I think the problem is the decimal points in the average of the other 7. Try adding up all eight before rounding.

Thanks, make sense

 

[Trodski]

You also have to take into account that PCC might, just might, kick in a spoil your Maths.
You also need to watch weather you've reached your Soft Cap...increases are halved....
It's Witchcraft.....

No PCC or soft cap but logic flawed as Alan states above

 

[rulie]

Hi,

I understand how index is calculated. Best 8 of last 20 score differentials averaged.

Logically it follows that when a counting score is lost/replaced to calculate the change you do; previous counting score - new counting score / 8.

This season I have calculated the expected change and been out by 0.1 a few times. Am I missing something from the calculation? To give you an example;

Previous differential 13.8 - new differential 7.1 = 6.7. 6.7 / 8 = 0.8375. HI reduction should be 0.8?

Actual reduction was 0.9. Anyone know why?

Not that big an issue but 0.1 can change the number of shots you get in knockouts etc.

Our system recalculates the HI as soon as a new score is entered (and is confirmed after midnight).

 

[backwoodsman]

It has happened to me twice since the start of WHS in UK.
Here is one of the times. I was given 3.0 but I was 2.9 if two decimal places had been used.

View attachment 48662

But isn't that an 'if my uncle was a woman he'd be my aunt' argument? Things could be done differently. But they aren't.

 

[Voyager EMH]

But isn't that an 'if my uncle was a woman he'd be my aunt' argument? Things could be done differently. But they aren't.

It wasn't an argument.
It was an illustration of accumulating rounding errors in HI calculation, which is what I believed the thread was mostly about.
I accept the reality that a small proportion of HIs, at any given time, are incorrect by one decimal place, because this is how it is being done.

 

[rulefan]

It wasn't an argument.
It was an illustration of accumulating rounding errors in HI calculation, which is what I believed the thread was mostly about.
I accept the reality that a small proportion of HIs, at any given time, are incorrect by one decimal place, because this is how it is being done.

It it is being done as specified by the designers then it won't be incorrect.

 

[Voyager EMH]

It it is being done as specified by the designers then it won't be incorrect.

And I am not incorrect in pointing out that there is an accumulation of rounding errors leading to to an incorrect result of a calculation sometimes.

I accept the reality of the system that is being used. All our HIs are what the authority decides - the reality.

I do not accept that the arithmetic is flawless.
It could be improved.

 

[backwoodsman]

You are indeed correct that the accumulation of rounding leads to different results.
But the rounding is not an error. So the results are not incorrect.
And yes, the methodology could be different - which would lead to a different arithmetic result.
But it isn't.
(Ps. me not going to bother any further ...)