• Hey, guest user. Hope you're enjoying NeoGAF! Have you considered registering for an account? Come join us and add your take to the daily discourse.

Official May 2007 NPD Prediction Thread

sazzy

Member
[NDS] 525,000
[WII] 470,000
[PS2] 145,000
[PSP] 165,000
[360] 175,000
[PS3] 65,000
[GBA] 75,000
[GCN] 9,000
 

donny2112

Member
Deku said:
This way Pachter's NPD preview or other leaks won't sway the contestants.

It could only affect them for the worse. ;)

Deku said:
There's also a lot of copying going on where someone might randomly pick a few other guesses and add/subtract from there to arrive at their own.

As previously mentioned, a Price is Right scheme is doomed to failure. You'd never know whom to copy. Besides, they might actually learn something about sales in the process. "Why did they pick so much higher/lower for system X than I was thinking?"

Deku said:
I'm also unsure if Cheese's program can detect last minute edits by contestants after the numbers come out.

I have not seen any edits after the numbers came out this year, and I've never seen someone try to sneak an edit in, either. That said, a quick check for edit times shouldn't be hard to do.

bmf said:
There will be a cut off time for adjusting your numbers, and any further changes just won't register in the results.

Cheesemeister said:
I'd like to set a deadline, and the earlier the better IMHO.

4:30 p.m. ET on the day of release. That's when the numbers are released, and it's the traditional time that these prediction contests are closed. It's okay if the parser isn't run until later. A simple visual check for the "edited" line on posts will rule out any late predictors.

Thanks for setting up the parser, Cheesemeister! :)

bmf said:
I'm going to prep the next one well ahead of time and post it on the Saturday morning preceding.

I have been waiting until the Sunday before (I was out most of this Sunday). It could be released as early as the Thursday night before, since that's when the final Japanese numbers come in for the equivalent period. Are you saying you want to take over posting the official NPD prediction threads?

Edit:
Cheesemeister said:
Predictions are now closed.

Ummm... ;)
 
Cheesemeister said:
Predictions are now closed.


Why?

They aren't released for another 36 hours.

And I haven't made my predictions yet. Methinks the original poster was confused on the cutoff, because we've never cut it off before ~4:00PM EST on the day of release before.
 
sonycowboy said:
Why?

They aren't released for another 36 hours.

And I haven't made my predictions yet. Methinks the original poster was confused on the cutoff, because we've never cut it off before ~4:00PM EST on the day of release before.

Objectively, it seems awfully late, but we can keep it open until then if you insist.
 

donny2112

Member
Cheesemeister said:
Objectively, it seems awfully late, but we can keep it open until then if you insist.

In case you missed it in my previous (long) post.

donny2112 said:
4:30 p.m. ET on the day of release. That's when the numbers are released, and it's the traditional time that these prediction contests are closed. It's okay if the parser isn't run until later. A simple visual check for the "edited" line on posts will rule out any late predictors.
 
Wait, are you predictions allowed to be edited? Because I changed my mind a couple of days ago on a few figures. My numbers will still be accepted, right?
 
donny2112 said:
4:30 p.m. ET on the day of release. That's when the numbers are released, and it's the traditional time that these prediction contests are closed. It's okay if the parser isn't run until later. A simple visual check for the "edited" line on posts will rule out any late predictors.

Oh yeah, and I'll definitely be at work at that hour. I guess I'll have to set up a thread to execute the process to get predictions tomorrow afternoon.
 
Vagabundo said:
What about leaks? WILL SOMEONE PLEASE THINK OF THE LEAKS?

The data is not released in ANY fashion before 4:00PM EST. It's a voluntary compliance issue for the SEC so that insider trading doesn't run rampant.

I do agree that cutoffs are needed, I just didn't pay attention myself. Historically, we've had a good number of people making predictions in the last few hours (myself included). A cutoff of 3:30PM might be good just so that we're sure nothing funny happens. And to give time for the parser to run and have the results ready before the release.
 
Zerostatic said:
Wii - 392k
Nintendo DS - 389k
PlayStation Portable - 189k
Xbox 360 - 165k
PlayStation 2 - 160k
Game Boy Advance - 82k
PlayStation 3 - 84k
GameCube - 13k

Please use the parser format.

sonycowboy said:
A cutoff of 3:30PM might be good just so that we're sure nothing funny happens. And to give time for the parser to run and have the results ready before the release.

To that end, did you want to PM me with the actuals before posting the NPD thread so as to include the prediction results in the OP?
 
sonycowboy said:
Why?

They aren't released for another 36 hours.

And I haven't made my predictions yet. Methinks the original poster was confused on the cutoff, because we've never cut it off before ~4:00PM EST on the day of release before.


Yay! Quick!

[NDS] 500,000
[WII] 375,000
[PSP] 165,000
[PS2] 140,000
[360] 135,000
[GBA] 100,000
[PS3] 95,000
[GCN] 9,000
 

Parl

Member
MirageDwarf said:
Looks like NPD prediction on GAF is becoming serious business. I hope it doesn't get too serious.

As soon as some of the newly pushed standards are fully established, we can more simply do the whole NPD prediction, actual numbers and GAF analyst ranking process without too much red tape and people getting confused. It'll be simple.

Here's what I ask:

1) Each month, there's the traditional absolute numbers rating with lowest total difference being best. There's the official point system for the month, official point system for the last 3 months and official point system for the year (I don't particularly care for the last one).

With an emphasis on recent month's, not only is it more accurate at representing the current analytical skills of individual GAF members, it gives newcomers a better chance at actual rating well on more meaning charts that represent their armchair analyst skills/luck.

Also, that Michael Patcher be apart of our rankings.
 

donny2112

Member
Parl said:
As soon as some of the newly pushed standards are fully established, we can more simply do the whole NPD prediction, actual numbers and GAF analyst ranking process without too much red tape and people getting confused. It'll be simple.

Suggested Standards for NPD Prediction Threads:

1. Posted 1 week before the expected NPD release.
2. Predctions closed @ 4 p.m. Eastern U.S. Time on NPD release day.

First Post Required:
3. Post or link to last month's HW/SW results.
4. Coverage dates.
5. NPD release date and time.
6. Significant releases/HW price drops for the month.
7. Parser template posted.

First Post Optional:
8. Japanese sales in the equivalent time period.

Optional:
9. Last month's rankings (units and points).



How's that for a first go? If we do start including software prediction rankings, there'll have to be some changes, though.
 
Ok. I'm back. 2.5 hour commute to work today, car is in the shop. Updated the op to an undetermined cutoff until it's been nailed down again.

donny2112 said:
Are you saying you want to take over posting the official NPD prediction threads?
I guess I'm getting a little ahead of myself here. I'm definitely having fun with it.

I mostly ran it this time because I was sitting at work bored on a Sunday with little to do. Otherwise it's mostly been yours for the past few months, and I wouldn't have posted it up if it hadn't seemed to be rather late for it to not have been created yet.

In the future I'll be happy to leave it in your capable hands, but I'll also be happy to fill in for you in any case where you'll be otherwise occupied.
 

GhaleonEB

Member
m0dus said:
I'm ready for tomorrow . . . the question is, is GAF ready?

npd.gif
WTF :lol
 

Stop It

Perfectly able to grasp the inherent value of the fishing game.
Hopefully I am not too late, :)

[NDS] 480,000
[WII] 405,000
[PSP] 195,000
[PS2] 170,000
[360] 165,000
[GBA] 86,000
[PS3] 80,000
[GCN] 9,000
 
Cheesemeister said:
Please use the parser format.



To that end, did you want to PM me with the actuals before posting the NPD thread so as to include the prediction results in the OP?

Like last month, I might be in a situation where I'm not able to post before Gamedaily / Gamasutra / Next Gen. I'm trying to work around it, but I post as soon as possible.

I also think it's better to keep the prediction stuff in this thread and the analsysis of actual results in the true results thread. In fact, I'd like people to try and stick to that. Alot of folks don't care about the predictions and it follows a separate conversation flow than the data thread.

We don't need all those, "That killed my prediction, I'm so awesome, congrats, etc". It more properly belongs in this one. Plus it will allow for this thread to live on into the night with the top predictors giving their logic for their choices so we can continue to refine these threads and our predictions.

I do want to add SW next month, knowing it won't get nearly the action the HW predictions will. But, I think in a few months (and it's actually true now), that it will continue to become more and more obvious that HW is pretty linear for the most part, except at key junctures that aren't all that frequent and overall many will do quite well in the predictions.

SW is a whole different ball of wax. Very few would have predicted Pokemon to do 1.7M last month or God of War 2 to fall from 800k+ to 100k+ (or whatever the actuals were). Those predictions will seperate the men from the boys.
 

Deku

Banned
Early closures as initially planned are meant to not give late predictors any form of advantage in case numbers are leaked out early or news organizations, or the companies being tracked release statements that may materially affect that month's projections. It looks like no previews will be happening this time aside from Pachter's but I don't think it's such a bad idea.
 
sonycowboy said:
Like last month, I might be in a situation where I'm not able to post before Gamedaily / Gamasutra / Next Gen. I'm trying to work around it, but I post as soon as possible.

I also think it's better to keep the prediction stuff in this thread and the analsysis of actual results in the true results thread. In fact, I'd like people to try and stick to that. Alot of folks don't care about the predictions and it follows a separate conversation flow than the data thread.

We don't need all those, "That killed my prediction, I'm so awesome, congrats, etc". It more properly belongs in this one. Plus it will allow for this thread to live on into the night with the top predictors giving their logic for their choices so we can continue to refine these threads and our predictions.

I do want to add SW next month, knowing it won't get nearly the action the HW predictions will. But, I think in a few months (and it's actually true now), that it will continue to become more and more obvious that HW is pretty linear for the most part, except at key junctures that aren't all that frequent and overall many will do quite well in the predictions.

SW is a whole different ball of wax. Very few would have predicted Pokemon to do 1.7M last month or God of War 2 to fall from 800k+ to 100k+ (or whatever the actuals were). Those predictions will seperate the men from the boys.

Awesome...software is much more interesting. Are you thinking of doing a top ten including numbers? It will definitely be harder for Cheese and others to track, but it can be done as long as we keep the game name spelling consistent.
 
sonycowboy said:
I do want to add SW next month, knowing it won't get nearly the action the HW predictions will. But, I think in a few months (and it's actually true now), that it will continue to become more and more obvious that HW is pretty linear for the most part, except at key junctures that aren't all that frequent and overall many will do quite well in the predictions.

SW is a whole different ball of wax. Very few would have predicted Pokemon to do 1.7M last month or God of War 2 to fall from 800k+ to 100k+ (or whatever the actuals were). Those predictions will seperate the men from the boys.

Yeah, there's no way in hell I'd want to even try and shove SW predictions through a parser reading a thread here, as everyone would have to use only pre-defined strings from a lookup table. We've seen just how well people around here can follow simple instructions.

In order to process that sort of data without collecting it by hand, one would rather set up a website form populated with all of the pertinent titles, and then parse the results. I could set this up, but would probably ask for volunteers to prepare the title list. (I push enough titles for the MC threads. =P) Allowing form users to add new titles to the list would obviously be disastrous.
 
Cheesemeister said:
Yeah, there's no way in hell I'd want to even try and shove SW predictions through a parser reading a thread here, as everyone would have to use only pre-defined strings from a lookup table. We've seen just how well people around here can follow simple instructions.

In order to process that sort of data without collecting it by hand, one would rather set up a website form populated with all of the pertinent titles, and then parse the results. I could set this up, but would probably ask for volunteers to prepare the title list. (I push enough titles for the MC threads. =P) Allowing form users to add new titles to the list would obviously be disastrous.
A predetermined voting block could be posted in the OP, and then repeated at the top of each page, if a few of us could keep an eye on the thread and make a uniform first post every page. I'm sure that you, donny, and myself would be in for doing the work, and there are probably a couple of others we could recruit to help catch the page change.
 
[NDS] 450K
[Wii] 380K
[360] 200K
[PS2] 190K
[PSP] 180K
[GBA] 80K
[PS3] 75K

Unless the system code is case-sensitive? It feels wrong to put WII unless I need to.
 
bmf said:
A predetermined voting block could be posted in the OP, and then repeated at the top of each page, if a few of us could keep an eye on the thread and make a uniform first post every page. I'm sure that you, donny, and myself would be in for doing the work, and there are probably a couple of others we could recruit to help catch the page change.

Oh, hell no. If I'm going to go through the rigmarole of setting up a tracking system, I want it to at least have a chance of performing correctly without requiring babysitting. Programs are supposed to save work, not create it. This thread has certainly been a cautionary tale to that end.

JoshuaJSlone said:
Unless the system code is case-sensitive? It feels wrong to put WII unless I need to.

It's not case-sensitive.
 
Cheesemeister said:
Oh, hell no. If I'm going to go through the rigmarole of setting up a tracking system, I want it to at least have a chance of performing correctly without requiring babysitting. Programs are supposed to save work, not create it. This thread has certainly been a cautionary tale to that end.

It's not case-sensitive.

Do you think its still relevant to consider the GC in the prediction totals? Or is it excluded?
 
Earthstrike said:
Do you think its still relevant to consider the GC in the prediction totals? Or is it excluded?

It's excluded if no actual NPD number is provided. Of course, if we do get a sales figure for it and you haven't made a guess, you get that amount added to your total of absolute differences used in the ranking.

People who only made predictions for the new systems are screwed.

Empty_One said:
OK then, is it [GCN] or [NGC], or does it not matter? I only ask because it is both ways in the OP.

Either is accepted by the parser, even though GCN is the dumbest acronym known to man. This was discussed much earlier in the thread.
 

Empty_One

Member
OK then, is it [GCN] or [NGC], or does it not matter? I only ask because it is both ways in the OP.

Edit: Sorry, wasn't aware it had been brought up, I read only the first page, not the whole thread, before I made my post. I didn't want anyone else's opinion influencing me. :D
 

Lorul2

Member
Woot.....I'll give it a go!


[PS2] 207k
[PSP] 221k
[PS3] 76k
[360] 212k
[GBA] 62k
[NGC] 08k
[NDS] 447k
[WII] 297k


What do I win?
 

Fuzzy

I would bang a hot farmer!
I don't usually make predictions but I'm curious how close the numbers I posted in the Canadian thread might be.

[NDS] 485K
[WII] 319K
[PSP] 227K
[PS2] 196K
[360] 167K
[PS3] 74K
[GBA] 65K
 
Top Bottom