The Box Office Buzz and Tracking Thread: Electric Boogaloo

[Biggestgeekever]

Endgame seems to have broken Movietickets too. No update from the top 5 this hour.

[Porthos]

9 minutes ago, Biggestgeekever said:

Endgame seems to have broken Movietickets too. No update from the top 5 this hour.

Trying to get data now is like trying to construct a mnemonic memory circuit using stone knives and bearskins.  

 

Spoiler

Google the quote, you heathens. 

Spoiler

Also, get off my lawn. 

 

 

[cax16]

Here is EG as of now for Friday May 3rd at my theatre,

 

10:00am- 4/306

10:30am-18/392

11:00am-21/336

12:30pm-23/355

2:00pm-15/306

2:30pm-38/392

3:00pm-12/336

4:30pm-21/336

6:00pm-4/306

6:30pm-48/392

7:00pm-146/336

8:30pm-23/355

10:00pm-13/306

11:00pm-39/336

[dakus]

Is the pulse tracker being down an issue with the tracker or Fandango Pulse itself?

 

I’m guessing it should probably correct itself once the weekend is over and the numbers drop to less astronomical levels?

[Biggestgeekever]

9 minutes ago, dakus said:

Is the pulse tracker being down an issue with the tracker or Fandango Pulse itself?

 

I’m guessing it should probably correct itself once the weekend is over and the numbers drop to less astronomical levels?

Yeah it's a problem on Fandango's end: https://www.fandango.com/dataviz/#

 

The ticket sales popping up are all for showtimes that happened on Friday.

[dakus]

14 minutes ago, Biggestgeekever said:

Yeah it's a problem on Fandango's end: https://www.fandango.com/dataviz/#

 

The ticket sales popping up are all for showtimes that happened on Friday.

Thanks! I checked out Pulse itself, but couldn’t tell if it was up to date or not. I didn’t think about looking at the showtimes.

Edited April 28, 2019 by dakus

[Cooper Legion]

On 4/27/2019 at 9:21 PM, Thanos Legion said:

17:00 Pacific 4/27/19 (End of opening Sat)  

1	90.7%	Avengers: Endgame
2	2.1%	Captain Marvel
3	1.5%	The Curse of La Llorona
4	1.5%	Breakthrough
5	1.2%	Shazam!

CM is cooling a tad on MT, though being #1 of holdovers in week 8 is pretty wild still. Not sure it will be above Llorona in dailies. Endgame still repping nonsense numbers, of course.

17:00 Pacific 4/28/19 (End of opening Sun)  

1	89.9%	Avengers: Endgame
2	1.9%	Captain Marvel
3	1.9%	The Curse of La Llorona
4	1.4%	Shazam!
5	1.3%	Breakthrough  

 

Llorona seems like it’s returning to being above CM for the dailies? And Endgame now with a downright embarrassing sub 90 😛    

 

Should be a quiet week, all upcoming openers are so small they might not  even show up til Thurs.

[Porthos]

Detective Pikachu Greater Sacramento Area Premiere Night Seat Report: T-11 Days and counting

 

	Sellouts	Showings	Seats left	Total Seats	Perct Sold
TOTALS	0	75	8030	8648	7.15%

 

Total Seats Sold Since Thr:                103

 

I don't have any good comps like Incredibles 2, so these will have to do.  Use with caution:

 

.3257x as many tickets sold as Fallen Kingdom 11 days before release.           [JW2 had 22 days of pre-sales while Pika Pika has 29]

.3121x as many tickets sold as Crimes of Grindelwald 11 days before release. [FB2 had 30 days of pre-sales while Pika Pika has 29]

NOTE: If a theater locally that is now selling tickets for reserved seating was selling them for the above movies, but as non-reserved seating for that movie, it is not counted toward the ratio of 'as many tickets sold' for the movie in question.  If a theater is brand new to the region and wasn't yet open for one of the above movies, however, it will be counted toward the ratio.  It will also be counted toward the ratio if it was playing in a different number of theaters locally.

 

JW2 (T-11)             198  tickets sold in that time period [0 sellouts/97 showings     |   8507/10113 seats left  | 15.58% sold]

Pika (JW adj)* (T-11)     91  tickets sold in that time period [0 sellouts/75 showings     |    7169/7601 seats left   |  5.68% sold]

FB2 (T-11)              184 tickets sold in that time period [0 sellouts/94 showings     |  11605/13377 seats left | 13.25% sold] 

Pika (FBadj)** (T-11)     95 tickets sold in that time period [0 sellouts/75 showings      |    7561/8114 seats left  |  6.82% sold]    

*  Pika (JW adj) is the number of tickets sold at the same theaters I had tracking info for Fallen Kingdom

**Pika (FB adj) is the number of tickets sold at the same theaters I had tracking info for Crimes of Grindelwald

 

====

 

Nightly updates starting tomorrow.

[Porthos]

PULSE IS BACK!!!!!! 

[Eric Prime]

PULSE IS FIXED...kinda.

 

Akvalley's charts finally updated, but the data from the past two days aren't up. But hey, any data is better than no data.

[Cooper Legion]

Endgame claimed a total of about 5 Fandango days then, yeesh. Could go years without another movie claim a single one!

Edited April 29, 2019 by Thanos Legion

[Porthos]

6 minutes ago, Thanos Legion said:

Endgame claimed a total of about 5 Fandango days then, yeesh. Could go years without another movie claim a single one!

With @stfletch's comments about capping (which I haven't fully looked into/thought about) I actually wonder if it's not so much traffic crashing Pulse as some sort of internal integer/variable hitting an overflow and causing the app to hard freeeze.

 

That is, and bear with me while I be stupid about computer programs for a second, there might be some sort of variable that tracks total tickets sold/monitored in a time period.  Or maybe something else.  If they didn't set the max value of that high enough, it could be an overflow error of some sort and isn't fixed until it is cleared out by some process.

 

Like, lets say they were using a 16 bit integer for something.  Signed gives a value from −32,768 to 32,767.  Unsigned caps out at 65,535. If that number overflowed, it could lead to all sorts of problems.

 

I'm not saying this IS an overflow situation. I left my computer programming days behind me over two decades ago.

 

But as I think about it, damn does it sound like it might be one.

 

If it just crashed under the onslaught of sales when tickets first went up, that'd be one thing.  But to go down Fri morning as well when, while busy, things should have been more spread out?

 

I dunno.  I might be wrong, but it makes sense to me as a possible culprit.

 

We'll probably never know, but that's a suspicion I now have.

[Nova]

4 hours ago, Porthos said:

With @stfletch's comments about capping (which I haven't fully looked into/thought about) I actually wonder if it's not so much traffic crashing Pulse as some sort of internal integer/variable hitting an overflow and causing the app to hard freeeze.

 

That is, and bear with me while I be stupid about computer programs for a second, there might be some sort of variable that tracks total tickets sold/monitored in a time period.  Or maybe something else.  If they didn't set the max value of that high enough, it could be an overflow error of some sort and isn't fixed until it is cleared out by some process.

 

Like, lets say they were using a 16 bit integer for something.  Signed gives a value from −32,768 to 32,767.  Unsigned caps out at 65,535. If that number overflowed, it could lead to all sorts of problems.

 

I'm not saying this IS an overflow situation. I left my computer programming days behind me over two decades ago.

 

But as I think about it, damn does it sound like it might be one.

 

If it just crashed under the onslaught of sales when tickets first went up, that'd be one thing.  But to go down Fri morning as well when, while busy, things should have been more spread out?

 

I dunno.  I might be wrong, but it makes sense to me as a possible culprit.

 

We'll probably never know, but that's a suspicion I now have.

 @BudStarr has also talked about Fandango having a Cap  (sorry I had to) per hour. They could probably explain how they came to that conclusion and stuff. 

 

*But yea if there is a Cap  I'm gonna be cautious with how I track Aladdin's first week sales then. But I also wanna know if there is proof of there being a Cap  because if Aladdin's first week sales are through the roof (or if they're not) then I would wanna know what role a Cap is playing in its sales. 

[BudStarr]

2 hours ago, Nova said:

 @BudStarr has also talked about Fandango having a Cap  (sorry I had to) per hour. They could probably explain how they came to that conclusion and stuff. 

 

*But yea if there is a Cap  I'm gonna be cautious with how I track Aladdin's first week sales then. But I also wanna know if there is proof of there being a Cap  because if Aladdin's first week sales are through the roof (or if they're not) then I would wanna know what role a Cap is playing in its sales. 

 

None of the overnight hours have hit the cap yet but, later today, add up the total number of tickets sold by all movies in a single hour and it will always come to 4000.

[Nova]

44 minutes ago, BudStarr said:

 

None of the overnight hours have hit the cap yet but, later today, add up the total number of tickets sold by all movies in a single hour and it will always come to 4000.

So in a single hour, the most amount of sets of tickets that can be shown on the Fandango app is 4,000? Is that what you are saying? 

 

If so, how does Fandango determine which movies get accounted for and which films get depressed? Or do all films see a depression? So like even End Game (which is what caused the cap) would see its numbers depressed as well? I actually believe that because last week before the server crashed, the second movie was selling like 40-50 tickets or so which could be understandable given how huge End Game became but I what was most odd to me was that the number of movies listed that had sold tickets had shrunk. So instead of seeing a string of movies that had sold 1 ticket, that list was pretty small compared to normal 

Edited April 29, 2019 by Nova

[BudStarr]

6 minutes ago, Nova said:

So in a single hour, the most amount of sets of tickets that can be shown on the Fandango app is 4,000? Is that what you are saying? 

 

If so, how does Fandango determine which movies get accounted for and which films get depressed? Or do all films see a depression? So like even End Game (which is what caused the cap) would see its numbers depressed as well? I actually believe that because last week before the server crashed, the second movie was selling like 40-50 tickets or so which could be understandable given how huge End Game became but I what was most odd to me was that the number of movies listed that had sold tickets had shrunk. So instead of seeing a string of movies that had sold 1 ticket, that list was pretty small compared to normal 

Yes, that’s exactly how it works. And sure enough, already we’re hitting the cap - add up the tickets for the 9am hour - 4000. 

 

I witnessed this throughout the whole of last week and I think it works like this. Pulse shows us the last 1000 transactions and updates this every 15 minutes. If it hits 1000 tickets after 10 minutes, then we don’t see any of the sales in the remaining 5 minutes. If it hits 1000 after 5 minutes then we miss out on 10 minutes of data. So if 8000 transactions take place in a single hour then we will only see half of them. 12,000 and we will only see a third. Assuming the numbers that we do see represent a roughly proportional sample then all numbers should be suppressed reasonably equally - including Endgame’s.

[Nova]

8 minutes ago, BudStarr said:

Yes, that’s exactly how it works. And sure enough, already we’re hitting the cap - add up the tickets for the 9am hour - 4000. 

 

I witnessed this throughout the whole of last week and I think it works like this. Pulse shows us the last 1000 transactions and updates this every 15 minutes. If it hits 1000 tickets after 10 minutes, then we don’t see any of the sales in the remaining 5 minutes. If it hits 1000 after 5 minutes then we miss out on 10 minutes of data. So if 8000 transactions take place in a single hour then we will only see half of them. 12,000 and we will only see a third. Assuming the numbers that we do see represent a roughly proportional sample then all numbers should be suppressed reasonably equally - including Endgame’s.

This is actually very valuable info. I wonder if this has happened in the past and it’s skewed our data. Just as an example: Last Kingdom opened after Incredibles 2 which had insane fandango sales the week it opened and I wonder if had an effect on Last Kingdom’s reported Fandango sales. I know. I know. Last Kingdom was more walk up friendly than usual but even still its Fandango sales were pretty mute for a movie that opened close to $150M but I’m wondering if a similar scenario happened then that may caused its sales to be more depressed 

Edited April 29, 2019 by Nova

[Cooper Legion]

13 minutes ago, BudStarr said:

Yes, that’s exactly how it works. And sure enough, already we’re hitting the cap - add up the tickets for the 9am hour - 4000. 

 

I witnessed this throughout the whole of last week and I think it works like this. Pulse shows us the last 1000 transactions and updates this every 15 minutes. If it hits 1000 tickets after 10 minutes, then we don’t see any of the sales in the remaining 5 minutes. If it hits 1000 after 5 minutes then we miss out on 10 minutes of data. So if 8000 transactions take place in a single hour then we will only see half of them. 12,000 and we will only see a third. Assuming the numbers that we do see represent a roughly proportional sample then all numbers should be suppressed reasonably equally - including Endgame’s.

Great breakdown, I’d long suspected it was something like this but never got around to digging for details. So Pulse is actually somewhat similar to MT when a huge movie is around, in that the %s should be right but absolute numbers can't necessarily be compared apples to apples with other time periods. All Pulse data for Pika, Wick, Aladdin, etc over the next 7 days at least will probably display some of this effect.

[BudStarr]

2 minutes ago, Nova said:

This is actually very valuable info. I wonder if this has happened in the past and it’s skewed our data. Just as an example: Last Kingdom opened after Incredibles 2 which had insane fandango sales the week it opened and I wonder if had an effect on Last Kingdom’s reported Fandango sales. I know. I know. Last Kingdom was more walk up friendly than usual but even still its Fandango sales were pretty mute for a movie that opened close to $150M but I’m wondering if a similar scenario happened then that may caused its sales to be more depressed 

Yes, I only noticed this for the first time last week but, once I realized it, some previous anomalies suddenly started to make more sense. It has definitely skewed comparisons in the past. I think this Pulse data is generally very good for comparing sales on an average week-by-week basis but if there are 1 or 2 in there that are going through the roof then it skews the data for everything else.

 

Pikachu and Wick got off to a reasonably good start every day last week, consistently showing numbers of 10-12 per hour through the night and early morning, only to collapse to low single digits later in the day. Meanwhile, the niche films that were selling 1-3 tickets per hour disappeared completely. And Endgame itself carried on chugging along doing the same 3500 every hour when it must have actually been selling much more than that. 

[BudStarr]

This is the other thing I’ve witnessed. The 10am hour has just updated on akvalley and it currently adds up to exactly 3000. Check it again in an hour after the next update and it will be 4000. The last batch of 1000 hasn’t been added yet.