DOW competition up/down/same

[ale]

There seems to be some discrepancy with the three levels - at least 2 people have chosen a value that is in the 'lower' range but the system has put them as predicting 'same'.
Or have I misread it?

 

[FetteredChinos]

also, now that volatility isnt what it was, shouldnt the "same" threshhold be slightly tighter??

 

[User]

A clarification of point scoring should be given on the Dow competition page.

 

[counter_violent]

"same" is based on 1% + / - index close value each friday .....may seem a wide band , but then again there is no limit on upside or downside

 

[ale]

My point is that, when entering the competition, you choose a value and that value fits into one of the 3 bands. These are labelled UP, DOWN or SAME.
The values chosen by some people have, in the table, been placed into a category different from where their chosen value should have put them.
Thus, the system that puts the label UP, DOWN or SAME onto the chosen value in the table seems to have different limit values from the defined limit values on the entry page.
This may have the effect of skewing the stats.
It would be a good idea to have the limits visible - at present they can only be seen on the entry page.
As for the definition of SAME, let the debate continue.
(IMHO 1% is too wide)

 

[Sharky]

I agree with your suggestion about making it so that a prediction and a direction must match. I'll add this feature. With regards the 1% +/- I'd be happy to change this if it's felt it's too wide - of course, we need to also be careful not to make it too narrow - suggestions would be appreciated, anyone have any stats as to the average weekly move of the DOW? Likewise, ill check over the past competition to see how many have been lower, same and higher - and post the results here.

 

[ale]

mmm. I said that I thought 1% seemed too wide.
Perhaps I was too hasty.
Having just looked (by eye) at the previous 12 months I'd say that the noise is +/- 40 points on a value of 10000. Thus, 10000 with noise of 0.8%.
In other words, I agree that 1% seems a reasonable value, although if the calculation is based on +/- then it may be better at +/- 0.5% (until the DOW halves or doubles in value!)
What do others think?

 

[Racer]

dc2000 you are being a little optimistic? And how come that is entered as the same!

Members Prediction Points
1 Pat494 10302 SAME 5+5
2 Salty Gibbon 10396 SAME 5
3 BRunner 10398 SAME 5
4 archimg 10285 SAME 5
5 KiltedWarrior 10285 SAME 5
6 Bluewave 10265 SAME 5
7 skeltonr9 10432 SAME 5
8 ajaskey 10243 SAME 5
9 dc2000 21415 SAME 5
10 bullboy 10240 DOWN 5
11 counter_violent 10164 DOWN 5
12 Joules MM1 10117 DOWN 5

 

[ale]

The definition of SAME is automatic and there is something wrong with the assignment (the reason this thread started).
Any joy is sorting this one Sharky?
As for dc2000's optimism, looks like his trigger finger slipped.
I hope he's more careful when trading for real!

 

[ale]

No - think I'm wrong there.
I thought the assignments were automatic based on the values entered - assuming that people put the values in the correct box.
I've just entered my value which is LOWER but deliberately put it into the SAME box which is where it has been assigned.
People clearly cannot be trusted to select the correct box in which to enter the value!
Sharky, how about if people enter only the value and the assignment of HIGHER / LOWER / SAME is done automatically? Is that possible? Also have the definition values on the list page.

 

[counter_violent]

While we're on the subject of changes , would it be possible to have a cut off point on day three of the Dow comp.

I realise this would spoil it for the spoilers but the Comp is supposed to be all about where the Dow will be in a week's time !!!!!.......I'm getting a little hacked off with the One day to go type entrants who's only motive can be to wreck the Comp for others.

Maybe there are others who feel the same.

 

[Sharky]

Yes, we could change it to -+ 0.5% with a 1% spread. Also I agree by entering a value we can determine the direction - but members have the option of choosing a direction without entering a value - which they then couldn't do. Is this a problem I wonder? Finally we could cut off entries sooner. Best to decide all the changes then we'll update it for the next competition - but with the next one starting next week, that doesn't leave us much time.

 

[ale]

My vote goes to a +/-0.5% band for "same". That's still 1% peak to peak.
I see no value in just being able to pick "up" or "down" without a value - how could that person win?
Thus, select a value and let the system specify the direction.
Anyone else got a view?

 

[counter_violent]

My vote goes to a +/-0.5% band for "same". That's still 1% peak to peak.
I see no value in just being able to pick "up" or "down" without a value - how could that person win?
Thus, select a value and let the system specify the direction.
Anyone else got a view?

Sounds OK to me.

Thanks to Sharky for noting the problem with "spoilers" and cut off point.

 

[Sharky]

I've now made the following updates:

i) -/+0.5% with a total 1% spread for 'same' direction
ii) you must now make a points prediction in order to enter
iii) the direction of your prediction is determined by the points you predict, not the box you enter it in.
iv) Cut off date for entries is recuded by a day from 4 days to 3 days. So lowest point score available is now 2.

 

[Bluewave]

Sharky, can you please check this weeks' Dow Comp scoring (3rd June). The Dow closed at 10461, yet "I Will Win This" didn't get 2+2 points for their nearest guess of 10465! But "shrenickshah" was awarded 3+3 points for their guess of 10495.

Why should someone win with a guess 34 points away just because they entered their guess in the SAME box?! Surely the final value is more important than the LOW / SAME / HIGH guess?

Pat494's guess of 10594 is 134 points away from today's close, yet they got 5 points for using the SAME box!

 

[Sharky]

Don't worry Bluewave - seems to be a bug with the software, following the changes I made a couple odf weeks ago. I'll make sure the scores are all updated with the correct values. Thanks for bringing this to my attention.

 

[Bluewave]

I've now made the following updates:

i) -/+0.5% with a total 1% spread for 'same' direction
ii) you must now make a points prediction in order to enter
iii) the direction of your prediction is determined by the points you predict, not the box you enter it in.
iv) Cut off date for entries is recuded by a day from 4 days to 3 days. So lowest point score available is now 2.

Sharky, is the +/- 0.5% for the SAME direction working properly? 0.5% + 10461 (Friday's close) = 10513. Therefore "shrenickshah" (10495) and "tradingintime" (10501) should have got their points as they both predicted SAME.

Also last week, "Joules_MM1" guessed DOWN with a value of 10174. That's 2.82% from the Dow close, yet points were still awarded because he guessed the right direction. This seems unfair on those that were closer to the close but got the direction wrong. I think there's too much weight put on the direction rather than the actual value guessed.

 

[Pat494]

Scoring Dow comp

I've now made the following updates:

iii) the direction of your prediction is determined by the points you predict, not the box you enter it in.
.

Glad to see you are improving the Dow competition.

Does this mean that we only get points for the direction, regardless of how close ??

May I suggest the number of points away from the Friday close is measured and then subtracted from say 200.
e,g. 50 points away from friday's close equals 200-50=150 points.
Therefore it is the same difference if one is 50 points over the value or short of it.

 

[EK1]

Bluewave/Pat,

I'll look into your queries about the competition.
Leave it with me for now.

Cheers

EK1