Well, the listing from 8/11/21 came through today 12:06 AM Purchased 119 for 750 each 12:08 AM Purchased 3000 for 750 each 12:08 AM Purchased 50 for 750 each 12:12 AM Purchased 1850 for 750 each 340 days.
Well, listing posted on 8/31/2021 is done: 7/30/2022 Purchased 2920 for 750 each 7/30/2022 Purchased 2080 for 750 each 333 days, and it's interesting that the whole listing got filled in only 2 parts...
Updated graph. (date format dd/mm/yy, I hate US format)
The effect of "mod23 AD sinks" can be seen.
I added the waited time at delivery date in case it can bring something to the table.
Prediction lines are calculated as if the waiting time was a distance from the order date, so the variation of the waiting time over time can be described with equations analog to kinematic equations in 1 dimension/time (rectilinear motion). "position", "velocity", "acceleration". The linear prediction is at constant "velocity" , the square prediction is at constant "acceleration".
It is interesting to see proof that adding AD sinks actually has an effect on the backlog.
I would not call it any proof. The change in waiting time is always dependent on everything that happened since posting, which is now a bit less than a year of progress. For us, it is impossible to accurately say if the core effect lies in the brand new AD sinks or the last autumns turmoil in player-base.
Anyway, I would expect the trend to follow the backlog, so my guess would call both the linear and quadratic predictions as highly optimistic. But we will see.
0
plasticbatMember, NW M9 PlaytestPosts: 12,405Arc User
It is interesting to see proof that adding AD sinks actually has an effect on the backlog.
I would not call it any proof. The change in waiting time is always dependent on everything that happened since posting, which is now a bit less than a year of progress. For us, it is impossible to accurately say if the core effect lies in the brand new AD sinks or the last autumns turmoil in player-base.
Anyway, I would expect the trend to follow the backlog, so my guess would call both the linear and quadratic predictions as highly optimistic. But we will see.
He did not say "adding AD sinks actually has an effect on the waiting time". He said "adding AD sinks actually has an effect on the backlog". It seems you agree with his statement that you quoted.
*** The game can read your mind. If you want it, you won't get it. If you don't expect to get it, you will. ***
It is interesting to see proof that adding AD sinks actually has an effect on the backlog.
I would not call it any proof. The change in waiting time is always dependent on everything that happened since posting, which is now a bit less than a year of progress. For us, it is impossible to accurately say if the core effect lies in the brand new AD sinks or the last autumns turmoil in player-base.
Anyway, I would expect the trend to follow the backlog, so my guess would call both the linear and quadratic predictions as highly optimistic. But we will see.
He did not say "adding AD sinks actually has an effect on the waiting time". He said "adding AD sinks actually has an effect on the backlog". It seems you agree with his statement that you quoted.
No, he said: "It is interesting to see proof that adding AD sinks actually has an effect on the backlog." And I say you do not see anything, you just guess. Not that it matters.
It is interesting to see proof that adding AD sinks actually has an effect on the backlog.
I would not call it any proof. The change in waiting time is always dependent on everything that happened since posting, which is now a bit less than a year of progress. For us, it is impossible to accurately say if the core effect lies in the brand new AD sinks or the last autumns turmoil in player-base.
Anyway, I would expect the trend to follow the backlog, so my guess would call both the linear and quadratic predictions as highly optimistic. But we will see.
For the most part, predictions can only be somewhat accurate inside a relatively constant situation. Given the "event" nature of Neverwinter, it is bound to be inacurate the further in the future we are from the last entry, and the closest the last entry is from an impactful event. A new module, an exceptionnal sale (blackfriday/jubilee as i'm yet to understand/model their recurrent effect), a big change (bonding stones), etc, all tend to create unpredictable singularities (big or more subtle). It's like trying to describe and predict the distance between 2 cars, one moving at constant speed on the road as a ghost with infinite fuel who goes through everything (it is the "order date"), the other one accelerating/decelerating/hitting the break or the accelerator pedals hard/changing gears/stoping to refuel inconsistantly because the driver reacts to things happening on the road or in the car itself (it is the delivery date).
But you are right, we can probably expect the waiting time to behave like the backlog very roughly, and with sort of an inertia, as the graph seems to point that out (there is only the strange may-june 2022 behavior that doesn't fit with this theory) Given the relative stability of the backlog from mid august 2021 to end of december 2021, I expect the predictions to start getting more and more accuracy again for orders made in october/november/december 2021, and the square prediction to be closer and closer to the linear one.
The current fall of the waiting time is probably mostly 2 or 3 things happening at the same time : - The AD sinks of mod23 are, I believe, at least temporarily efficient (and I agree it will fade away the older the mod is). It is also an incentive to spend Zen in the Zax for new unaware players, so maybe it also has improved the amount of Zen exchanged for AD. - Holidays are also a favorable period for a load of newplayers to arrive (potentially more unaware people to put their $ in Zen into AD with the Zax) - We are currently at the end of a "Bonding stone effect" cycle. End of 1st cycle was happening when the backlog line reverted on july-august last year (players who didn't get more than 18,75M AD, which were the most part). End of 2nd cycle was happening in june-july this year and so the 3rd cycle has started, but there are much less players able to refill from their bonding stones AD now : only those who had snatched over 37,5M with the bonding stones are concerned by the 3rd cycle (the last cycle for me will be 4th one ^^), and also less of them are still playing the game.
My data shows the vector acceleration pointing down since 14th june, while the vector velocity was pointing up until 1st july. After 1st july, the velocity started pointing down and since ~15th july 2022 the acceleration is less and less strong down. So we may have, at the end of summer a linear prediction going down but a square prediction reverting up, or square and linear merging in one line. It is not the case yet (quite far from it honestly ^^), but it is probable it can happen around mid september deliveries. If you want @rikitaki, I can add a cubic prediction, which would describe a constant variation on the "acceleration" :P (result would be a curve with 2 inflection points, square only has 1 inflection point)
@tchefi Statistically speaking you have no supporting data to choose the correct kind of curve your prediction should be - which means anything is fine as long as it is at least slightly believable. For the most part, the linear approximation is fine.
edit: Your linear prediction would mean the game is "stable", the quadratic - in this situation - would be "healing". If I remember correctly, neither can be said about Neverwinter in the past years.
As for 'proof': There is a clear break in the backlog line that corresponds to the release of mod 23 and the associated AD drains. While there of course could be other reasons, the most obvious reason is the significant AD drains of mod 23. As 'proofy' as it gets with these types of data and limited visibility to what happens behind the scenes.
And to add to @tchefi#6735 's list of reasons: The reduction of 50% players from last summer until this summer likely contributes significantly to the reduced size of the backlog due to less Zen slots being just replaced when they are filled.
As for 'proof': There is a clear break in the backlog line that corresponds to the release of mod 23 and the associated AD drains. While there of course could be other reasons, the most obvious reason is the significant AD drains of mod 23. As 'proofy' as it gets with these types of data and limited visibility to what happens behind the scenes.
...there is a clear break line at M23 release, OK. But how the hell can you "clearly" connect it with some meagre AD sinks? If you believe Steamcharts can present the rough trends in the player-base, there is a chance the game gained 20% bonus people (not just the time spent in the game). That would bring SOME impact to the ZAX trend as well.
And to add to @tchefi#6735 's list of reasons: The reduction of 50% players from last summer until this summer likely contributes significantly to the reduced size of the backlog due to less Zen slots being just replaced when they are filled.
Here, there are two 25% bumps down in the steam charts corresponding to the announcement of W7 support termination and actual W7 termination. There are two points to that - first: I think users of outdated system were more inclined to use ZAX than to put money into the game, so having them gone might heal the backlog quite a bit. Second: I am not sure, but W7 users might be more common/uncommon on Steam compared to the average, so the apparent 50% decline on Steamcharts might not represent the trend correctly.
I posted on 7.15.22 that I had received my Zen after 340 days; @qexotic posted yesterday that he received his Zen after 330 days. That is pretty encouraging. Over the span of the past month, it looks like the wait time for Zen has actually decreased by 10 days.
5K posted sep 27th 2021 and Received on Aug 19 2022. Seems like backlog/wait time reduction is accelerating.
I would say it is. You can discuss AD sinks, or whatever could possibly have pushed it quicker. But in the end, they've finally gotten to the point where the line isn't growing faster than what's spent. When they reset line and made it 750, from 500, it went to over 40 mil in a year? Then the bondings payoff, pushed it to 70+ mil. Hoping it keeps dropping 100k+ per day, and be around 40 mil at christmas.
Comments
Received: July 16
340 Days
12:06 AM Purchased 119 for 750 each
12:08 AM Purchased 3000 for 750 each
12:08 AM Purchased 50 for 750 each
12:12 AM Purchased 1850 for 750 each
340 days.
Received 07/23/2022
Wait time keeps going down.
7/30/2022 Purchased 2920 for 750 each
7/30/2022 Purchased 2080 for 750 each
333 days, and it's interesting that the whole listing got filled in only 2 parts...
(date format dd/mm/yy, I hate US format)
The effect of "mod23 AD sinks" can be seen.
I added the waited time at delivery date in case it can bring something to the table.
Prediction lines are calculated as if the waiting time was a distance from the order date, so the variation of the waiting time over time can be described with equations analog to kinematic equations in 1 dimension/time (rectilinear motion). "position", "velocity", "acceleration".
The linear prediction is at constant "velocity" , the square prediction is at constant "acceleration".
It is interesting to see proof that adding AD sinks actually has an effect on the backlog.
Unfortunately a lot of the immediate effects from dragons likely is burning out now as most players that are able to do dragons have their gear.
Anyway, I would expect the trend to follow the backlog, so my guess would call both the linear and quadratic predictions as highly optimistic. But we will see.
He said "adding AD sinks actually has an effect on the backlog". It seems you agree with his statement that you quoted.
Given the "event" nature of Neverwinter, it is bound to be inacurate the further in the future we are from the last entry, and the closest the last entry is from an impactful event. A new module, an exceptionnal sale (blackfriday/jubilee as i'm yet to understand/model their recurrent effect), a big change (bonding stones), etc, all tend to create unpredictable singularities (big or more subtle).
It's like trying to describe and predict the distance between 2 cars, one moving at constant speed on the road as a ghost with infinite fuel who goes through everything (it is the "order date"), the other one accelerating/decelerating/hitting the break or the accelerator pedals hard/changing gears/stoping to refuel inconsistantly because the driver reacts to things happening on the road or in the car itself (it is the delivery date).
But you are right, we can probably expect the waiting time to behave like the backlog very roughly, and with sort of an inertia, as the graph seems to point that out (there is only the strange may-june 2022 behavior that doesn't fit with this theory)
Given the relative stability of the backlog from mid august 2021 to end of december 2021, I expect the predictions to start getting more and more accuracy again for orders made in october/november/december 2021, and the square prediction to be closer and closer to the linear one.
The current fall of the waiting time is probably mostly 2 or 3 things happening at the same time :
- The AD sinks of mod23 are, I believe, at least temporarily efficient (and I agree it will fade away the older the mod is). It is also an incentive to spend Zen in the Zax for new unaware players, so maybe it also has improved the amount of Zen exchanged for AD.
- Holidays are also a favorable period for a load of newplayers to arrive (potentially more unaware people to put their $ in Zen into AD with the Zax)
- We are currently at the end of a "Bonding stone effect" cycle. End of 1st cycle was happening when the backlog line reverted on july-august last year (players who didn't get more than 18,75M AD, which were the most part). End of 2nd cycle was happening in june-july this year and so the 3rd cycle has started, but there are much less players able to refill from their bonding stones AD now : only those who had snatched over 37,5M with the bonding stones are concerned by the 3rd cycle (the last cycle for me will be 4th one ^^), and also less of them are still playing the game.
My data shows the vector acceleration pointing down since 14th june, while the vector velocity was pointing up until 1st july. After 1st july, the velocity started pointing down and since ~15th july 2022 the acceleration is less and less strong down. So we may have, at the end of summer a linear prediction going down but a square prediction reverting up, or square and linear merging in one line. It is not the case yet (quite far from it honestly ^^), but it is probable it can happen around mid september deliveries.
If you want @rikitaki, I can add a cubic prediction, which would describe a constant variation on the "acceleration" :P (result would be a curve with 2 inflection points, square only has 1 inflection point)
edit: Your linear prediction would mean the game is "stable", the quadratic - in this situation - would be "healing". If I remember correctly, neither can be said about Neverwinter in the past years.
And to add to @tchefi#6735 's list of reasons: The reduction of 50% players from last summer until this summer likely contributes significantly to the reduced size of the backlog due to less Zen slots being just replaced when they are filled.
Posted 16 Sept 2021. Received 12 Aug 2022
Has @hotfrostworm given up on maintaining the forecast spreadsheet? Cause it's now WAY off and useless.
Guild Leader
Neverwinter SOLO Alliance
Posted a month after the previous one, but received only 12 days after.
Guild Leader
Neverwinter SOLO Alliance