POTM / SPOTM voting discussion topic

[PorkScratching]

It sounds like a good idea, but I think you might be making too much work for yourself! Also, the extra effort required might stop a few people voting - perhaps. I think the current system is fine - there's enough of us voting to make it work well by the looks of things. Happy to give it a go though if you decide to go ahead.

[flour power]

Interesting idea Andy, however it sounds like a lot of work.

I guess that you would have discussed this already, but would it be easier just to have one vote?

In any case, if you're happy to run it as suggested we could give it a go.

Mike

[andygolfer]

thanks for the answers guys, anyone else got any thoughts on the suggestion please?

[andygolfer]

Interesting idea Andy, however it sounds like a lot of work.

I guess that you would have discussed this already, but would it be easier just to have one vote?

In any case, if you're happy to run it as suggested we could give it a go.

Mike

i'm happy to give it a one month trial on the current competition to see how it goes (and what difference it makes against 3 equal votes!)

[flour power]

Sounds like a plan, then.

[andygolfer]

As all the comments received were positive towards the 1st, 2nd and 3rd choice idea, I have set up the polls to trial the system this month now that the entries have closed.

The 3 polls can be found here:
the poll for your first place choice can be found here: viewtopic.php?f=110&t=19046

the poll for your second place choice can be found here: viewtopic.php?f=110&t=19045&p=60173#p60173

the poll for your third place choice can be found here: viewtopic.php?f=110&t=19044

How the voting and scoring will work:
The 3 polls are titled 1st choice, 2nd choice and 3rd choice, you will be allowed 1 vote in each. Every first choice vote will receive 3 points. The 2nd choice votes will receive 2 points and the 3rd choice votes 1 point. The winner will be the entry which gains the most points, NOT the most votes. In the event of a tie the winner will be the entry with the most 1st choice votes and if still equal the most 2nd choice votes and if still equal the result will be a tie.

If you have any questions please post them here rather than on the voting poll topics as that way they will be kept together on one topic (and the three topics will also remain together and in order on the forum).

[andygolfer]

So I trialled it for a month and it wasn't too much hassle for me, I set up an excel spreadsheet and updated the votes daily to make it easy at the end, this is a screen shot of the spreadsheet with the data on it. The only thing I had to enter manually was a ranking calculation (not shown), copying and pasting all the way down made it refer to the wrong top cell I will read up on it to see if I can correct it.



Rather embarrassingly - I WON! but the saving grace is that just when counting the total votes and not the corresponding points I was still ahead so not quite trialling it this month didn't affect the result.

The results weren't a lot different but if you look at the 3rd placed on both votes and points it shows that the greater number of 1st choices split the first and third by votes to 3rd and 5th with the 5th place moving up to 4th so the weighting did have an effect where the number of votes were close but the weighting did lead to a bit of a runaway winner (by 14 points) whereas on votes it was only 4 (13 v 9). Personally I think the weighting gives a more accurate reflection of preferences - your thoughts please?

There were 60 votes in total but not 20, 20,20 as you might expect! Someone only voted once, then another voted twice and then later another single 1st vote, followed yesterday by a second & third vote only so not everybody used all three votes. The total of votes is slighly up on previous months but not as higher difference as I expected to be honest, I was hoping for about 30% more and it was only roughly 10%

here's another interesting statistic (I love statistics and spreadsheets):

cumulative votes by end of each day, with POTM as well for comparison

SPOTM POTM
02/08/2014 32
03/08/2014 47 24
04/08/2014 47 24
05/08/2014 48 24
06/08/2014 48 24
07/08/2014 51 27
08/08/2014 54 27
09/08/2014 54 27
10/08/2014 55 27
11/08/2014 60 27
12/08/2014 60

the majority of votes appeared in the first 24 hours and then a few on the last day - either my reminder or tactical votes I reckon. POTM got off to the same quick start and then stalled so I wonder if it is necessary to run for 10 days, if it were only 7 would we see a rush of votes at the start a bit of a lull (shorter than at present) then a few last minute ones.

So should I continue with this format or not?

my personal opinion is that I think it gives a more accurate result and really wasn't much effort to set up especially as I collated votes as they were entered (or daily anyway) so I'm happy to do so (this summary has taken more time than the actual competition).

Was it a problem to have to use 3 different topics for your voting even with the link to the other 2 included each topic? I did realise one possible loophole almost straight away for anyone devious enough to want to use it but dont think anyone did - I won't publicise it in case nobody else has picked it up but hope that it wouldn't be taken advantage of if they knew.

So, what do you think please, would you like me to do it this way for future competitions (I'm happy to if that is the preference)?

And I really would value any feedback positive OR negative, thanks for reading to the end!

[PorkScratching]

Firstly thanks for your efforts on this Andy. My preference would be to revert to the single voting topic where up to three votes can be cast. This is because:

* Just one page to check to see how the images I've voted for are doing (because they are bold once voted for).
* If I've entered I can look at just the one page to see how my shot is doing.

[flour power]

Thanks Andy for a. running the trial and b. your comprehensive write up. Oh, and congratulations!

For me, the three vote system made it much more interesting as it wasn't obvious who would come 2nd and 3rd (the winner was pretty clear from the beginning) without working it out.

FWIW, my 1st vote would be; to continue with the trial. My 2nd vote would be; to continue with the trial and my 3rd vote; to continue with the trial.

EDIT: I wouldn't be too upset if the voting time was shortened to, say, 7 days. As you say, the majority of the voting is within that sort of timescale, anyway.

Cheers.


Mike

[andygolfer]

many thanks for the feedback so far - one for and and against and I've had a bit of feedback from other moderators as well but no strong feelings either way.

So has anyone got any more thoughts on this please? Currently I think my best offer is to give it a go again for the August comp and see how that goes and what more feedback I get - if the results are closer (the winner got well ahead very early on and stayed there) then we might get more of a feeling for which gives the better result and which is easier to use and manage (the latter is down to me).

I'm also interested in thoughts on shortening the poll time although anyone on holiday for a week could miss out on the chance to vote so I'm personally a bit anti that idea at present but open to discussions on the idea.

come on , lets have some more feedback please

[Mike]

Hi Andy, I know it's a lot of extra work for you but I'd personally like to see it continued for a month, or two at least.

Great work fella.

[andygolfer]

The voting for the August SPOTM has closed and the winners posted in the Hall of Fame, congratulations to Mike (Flourpower) who followed up last month's success in POTM with a win in SPOTM. Once again i trialled the ranked voting polls and this is the scoresheet showing the votes received and the resulting points awarded:

SPOTM august 2014 scores by sickbag_andy, on Flickr

Like last month the ranked votes made no difference to the result re the top three except last month we would have had joint 3rd places, this time the placing was exactly the same using either method. I have to say I am very surprised that the different methods didn't produce any variation in the top 3 placings I really thought it would.

My conclusion is that I really dont think it warrants the three different polls as it seems we will still get the same winner in the end and as Porkscratching said after last month's the single poll is easier and simpler. I would be interested to know how many people actually worked out the scores as the poll progressed, if you want to keep it private then PM me please.
Anyway, i've tried it twice and it has if nothing else, proved that the single poll does give an accurate result. I accept there might be an odd month where 3 polls make a difference but surprisingly the effect is minimal.

So, will anybody object if I return to the old system or would you like one more month of trying this just to prove the results? I'm happy to if feelings are strong enough but don't think anything radical will happen with the results if i do.

Feedback PLEASE!

[flour power]

Hi Andy.

Just to back up what I said in my PM, and to go on public record...

Thank you for your efforts and for making the outcome so transparent and clear to understand.

As a supporter of the ranked voting system, I'm also surprised that the result would have beem the same using either system. Given the evidence, I would agree that the extra effort isn't worth it and that we return to the previous, single poll, system.

Thank you for giving it a fair trial and thank you to everyone who voted for my image!

Regards,


Mike

[Mike]

I agree with everything that Mike has said above, I would also like to thank you for all of the effort you have put into this, Andy.

Have a great holiday, you lucky git.