Dev Request if Easy and Fast... Data Mine Task Force and Trial Times

[BlackSpectre]

Hello my favorite devs,

 

How easy would it be to data mine the completion times for each of the game's trials and task forces? The best format would be in a spreadsheet form (but any format would be fine). I could do the data analysis and plotting with the data to see any important patterns and to get the average, mode, and range.

 

This information would be HIGHLY useful for players when trying to decide on whether or not to take on a task force or trial.  

 

Anyway, I'm only asking if something like this would be easy and fast. I don't want to take up much of your valuable time. If it would take a lot of effort, just disregard and file it under pie in the sky. 🙂

 

Thanks a bunch!

[Greycat]

On 6/24/2024 at 7:46 PM, BlackSpectre said:

How easy would it be to data mine the completion times for each of the game's trials and task forces? The best format would be in a spreadsheet form (but any format would be fine). I could do the data analysis and plotting with the data to see any important patterns and to get the average, mode, and range.

 

This information would be HIGHLY useful for players when trying to decide on whether or not to take on a task force or trial. 

 

 

... not a dev, obviously, but how would this really be useful or accurate? The times are going to depend on the group, how they *want* to run it (granted, most are going to do it in one shot, but some do run them over multiple days/sessions,) clear all vs speed, how well the group handles it, who's exemped down vs at level and in what combination, etc.

 

I mean, if someone sees Posi 1 at - say - 15 minutes, because there are a bunch of groups exemping down and speed running it during the WST, and they run it with their friends or a pug at level 8-10, it's *not* going to take 15 minutes.

[BlackSpectre]

You can do all sorts of things with numbers and statistics, and actual data can tell you all kinds of things. A range from the fastest to the longest is useful, knowing how long the vast majority of trials and task forces take to complete is very useful, and knowing an average is also very useful. The idea is to get an ESTIMATE or an idea of how long it might take, not to nail down an exact time. You are right that there are too many variables to settle on an exact time that is true for everyone. That's not the goal. The goal is to get an idea of about how long it might take, a starting point, that's all. Then it's up to you to factor in the other variables, team size, player skill, speed run or kill all, etc.

 

Patterns in data have the possibility to tell you almost exactly how long a trial or task force might take to complete, or potentially tell you nothing worthwhile. It all depends on the data. And without seeing the data, it's impossible to draw any conclusions about it. 

[SeraphimKensai]

As much as I love data.... Cipher hasn't given us new stats on AT/Powersets since 2020, so good luck getting TF times.

[UltraAlt]

On 6/24/2024 at 8:46 PM, BlackSpectre said:

How easy would it be to data mine the completion times for each of the game's trials and task forces?

 

So we are talking about a wide diversity of times even within the same task force or trial.

 

Is it a speed run?

Is it a leveling run?

Is it with a full team?

Is one person running it set on +4/x8?

How experienced are the players in the team running the content?

What level are the characters that are running the content?

Are all the members on the team exemplar'ed down to run the content? (aka have powers above what characters would have playing at "level" with the content)

 

There are a whole lot of factors that are going to skew the output that I believe you are looking for.

[BlackSpectre]

Those are all good things the player needs to consider when estimating the time to complete a task force or trial. However, those things will not be visible in the data… which is why you don’t want to use them when anayzling the data. 
 

For example, if the data shows that 90% of all task forces are completed between 30-40 minutes, then the estimate I would go with is that range… 30-40 min. I might even state the percentage… 90% completed within 30-40 min. 
 

The data might show that 50% are completed in 20 minutes, while 40% are completed in 45 minutes. And then 10% scattered all over the place. This might indicate the times for speed runs (20 min) and kill all runs (45 min). Not conclusive unless additional testing was done to confirm the times, but it definitely shows a pattern. 
 

The best way to communicate the patterns in the data is through a graph.
 

i’d probably get completion times for 1 year’s time. That should probably be good enough. 

[TheZag]

Merit rewards are already based on the fastest clear times of a task force or trial.  Just take a look at cavern of transcendence and its paltry reward when most teams took over 30 minutes and speed runners were finishing in less than 6 minutes.  I dont know how much clear time data the team collects but it isnt none.  One of the GMs even mentioned how many attempts and successes there were of the original Master of LRSF on torchbearer (in game) the day it came out (its not clear times but still data collection).

 

Getting data released is probably the most difficult hurdle.  Im pretty confident that there would be spikes at speedrun,  kill all and 'regular' teams even with the differences in team composition,  gear and skill spread throughout.

 

[Bionic_Flea]

Be careful what you wish for.  I'm all for more info and data.  But do not be surprised that the merit rewards for things like Tin Mage and Apex get adjusted downward about the time that data is provided.

 

Players are like water -- they both choose the path of least resistance -- and gravitate towards the fastest rewards/time.

 

While I don't particularly want to see rewards nerfed anywhere, I wouldn't mind some downward adjustments if that also included some upward adjustments to keep merits/time relatively balanced while also acknowledging that harder content deserves better rewards.  And harder content doesn't only mean 4*.  It can be acknowledging that you don't have as many tools at level 10 as you do at level 50+3.

[Arc-Mage]

Example why these numbers will be almost useless: Several years back they had a contest to see what team could run +4 x8 ITF the fastest. It was a pretty big event with lots of prizes. Every team started at the same time. Every team was balanced. 
 

One thing everyone forgot about was map spawns. The Cysts map in the ITF has several different map layouts that can spawn. The fastest map is the straight (shotgun) map. The shotgun map only spawned for one team. That team blew everyone else’s time out of the water. It wasn’t even close. 
 

The newer TF/SF have multiple different maps that can spawn. This can greatly alter the time for completion. 
 

You can pretty much judge how long an TF/SF should take to complete by the number of Merits rewarded. 
 

 

Edited July 1 by Arc-Mage

[Jacke]

19 hours ago, Bionic_Flea said:

Be careful what you wish for.  I'm all for more info and data.  But do not be surprised that the merit rewards for things like Tin Mage and Apex get adjusted downward about the time that data is provided.

 

Players are like water -- they both choose the path of least resistance -- and gravitate towards the fastest rewards/time.

 

While I don't particularly want to see rewards nerfed anywhere, I wouldn't mind some downward adjustments if that also included some upward adjustments to keep merits/time relatively balanced while also acknowledging that harder content deserves better rewards.  And harder content doesn't only mean 4*.  It can be acknowledging that you don't have as many tools at level 10 as you do at level 50+3.

 

@Bionic_Flea again getting a good bite into the situation.  😺 

 

The devs do gather data to help them see where the game is going and some of the details on how changes are affecting the game.  Some of that analysis affects which things should be considered for changes in the future.

 

When looking up data that's been shared on these forums, we have to be careful not to read too much into the data.  Especially if that data is even a couple years old, because the game is changing in significant ways.  (Example: I've yet to make up new builds for all my Toons for the big changes in Issue 27 Page 5, because those changes affected all Toons.)

 

BTW, the Apex and Tin Mage II Task Forces are also locked at Notoriety +4.  I asked about this and was told this was to allow them to keep their current rewards.  I highly suggest having at least a Tier 3 Alpha Boost slotted so it's only effectively +3 to a Toon (but virtually all Pets don't inherent the Castor's Global Buffs, so it'll still be +4 or worse to them).

[BlackSpectre]

On 7/1/2024 at 6:15 AM, Arc-Mage said:

You can pretty much judge how long an TF/SF should take to complete by the number of Merits rewarded.

Sadly that’s not the case. Originally, that was the live devs’ intention, to link the amount of reward merits to the amount of time it took to complete the task force, trial, story arc, etc.  My first attempt to determine times for task forces was to use reward merits to calculate the average completion times, but the resulting times were all over the place and did not reflect how long the completion times actually were.  I did this about 3 years ago and since discarded the info I gathered as useless.  
 

in general, one thing I concluded was that the merit/time ratio was 30 reward merits per hour. This ratio fits many things in the game, but at some point the devs stopped using it to determine the amount of reward merits. More than likely they just kept the reward merits for prior tasks the same, and then diverged with new tasks.  So not even the 30 reward merits per hour is not consistent. 
 

Fundamentally, linking reward merits to time is unfair and in general not a good idea. Some players complete tasks faster or slower than other players. Does that mean the players that take the most time to complete should earn the most reward merits and players that do it faster earn less? Or should it be flipped around so that players who are able to complete tasks faster earn more rewards? Don’t we want to reward players for accomplishing amazing feats such as completing tasks forces fast? I mean, it takes some skill, knowledge, and thought to do speed runs. Meanwhile players who decide to clear the map are putting in much more work and time to complete. At some point how many reward merits a task gives becomes arbitrary. 
 

In the game Red Dead Redemption Online rewards are in large part tied to how much time a player spends completing the task. The more time spent, the more rewards. This builds in an exploit and players are happy to take advantage of it. They do everything required for the task, and then right before they’re about to complete the task they sit on it. Go take a shower, get a coffee, eat lunch, go to the grocery store, pick up their kids from school, etc. After 30 minutes or so of being AFK they come back and finally end the task and reap the bigger rewards. It actually takes a lot of fun out of the game. 
 

I think the live devs realized all of this and is why the amount of reward merits earned is not consistent. This begs the question, then how should reward merits be allocated? 

it seems to me that the best way to determine reward merits would be to create a formula that considers the completion of the task, the type of task,  the difficulty of the task, the size of the team/league, and the amount of teamwork and coordination required for the task.  Time should not be considered (except possibly as part of determining difficulty). 
 

What would this prompt players to do? Team instead of solo, have bigger teams, do the important or signature tasks of the game, tackle the most difficult tasks in the game, become skilled in leadership and cooperation. 
 

At some point how many reward merits a task rewards becomes a judgement call, even if it does factor in some objective items. 

Edited July 3 by BlackSpectre

[Arc-Mage]

On 7/3/2024 at 12:53 AM, BlackSpectre said:

Sadly that’s not the case. Originally, that was the live devs’ intention, to link the amount of reward merits to the amount of time it took to complete the task force, trial, story arc, etc.  My first attempt to determine times for task forces was to use reward merits to calculate the average completion times, but the resulting times were all over the place and did not reflect how long the completion times actually were.  I did this about 3 years ago and since discarded the info I gathered as useless.  

With an average team that knows what they are doing and all working for the same end goal you can absolutely get get averages out of every Task Force. Those averages equal the time spent vs Reward.

 

I've been playing this game for almost 20 years straight. Literally run Thousands of Task Forces from lvl 02 to 50+. Every last Task Force takes about the same time to complete every time. Just one example, the first TF, Posi 1 and 2, both should always take about 30 minutes  each to run with an average team with Teleport. You may have a outlier from time to time, newb or someone who has a bad build or is just a bad player or no TP on the team. All these things will stretch out the time of course but more often then not you will not have this problem. The vast majority of those playing have been playing for years and know what they are doing. 

 

Look, I was on the Secret Server. I know for a fact when they couldn't create new content they spent time on game balance. For years it felt like that's all they did was game balance. The game as it is now is probably more balanced then it has ever been. The last few updates have continued this trend. 

On 7/3/2024 at 12:53 AM, BlackSpectre said:

Fundamentally, linking reward merits to time is unfair and in general not a good idea. Some players complete tasks faster or slower than other players. Does that mean the players that take the most time to complete should earn the most reward merits and players that do it faster earn less? Or should it be flipped around so that players who are able to complete tasks faster earn more rewards?

So this is what it's about? Time vs Reward is absolutely the best and fairest way to distribute reward in an online video game.

 

I have a Def Pure Emp that I have to slog my way through Mishes with, I also have a Tank Willpower/Energy that crushes in every mish. My point is if you have a toon that doesn't crush in every mish/TF, build one. Maybe it's your toon that's as fun as a Happy Day at the Park but it just doesn't give the results you are looking for. Change it up until you find that toon that absolutely crushes the TFs so you get those averages everyone else is getting.  

 

Happy Hunting.

[Without_Pause]

On 6/28/2024 at 7:57 PM, SeraphimKensai said:

As much as I love data.... Cipher hasn't given us new stats on AT/Powersets since 2020, so good luck getting TF times.

This. I don't doubt the devs have lots of data they use for various things, but giving players said data feels as it isn't something they are going to do. The thread in question gets repeated calls for updated data and yet four years later there has been none. 

[BlackSpectre]

On 7/4/2024 at 11:17 AM, Arc-Mage said:

reward merits…
 

So this is what it's about?

Not at all.  Just trying to help fellow players have a better idea of what they’re getting into time-wise when contemplating doing a TF.  Everything else in this thread is a side conversation, more or less.  
 

Our volunteer devs are doing a fantastic job. Any criticism of a game mechanic is not meant to criticize… well, anyone. My apologies if anyone felt otherwise. 

 

 

P.S. I went ahead and re-created the reward merits to completion times for task forces and posted it over on this thread so you can see the data for yourself. 

 

 

Edited July 10 by BlackSpectre
added the P.S.

[BlackSpectre]

After looking over the reward merits for incarnate trials, and comparing them to typical completion times given by players, it looks like the reward merits for incarnate trials are awarded in a 1 to 1 ratio. Meaning, 1 reward merit for every 1 minute of the average completion time, or 60 reward merits per hour. If interested, you can see the data over at this post.