One hundred trials later...


Adeon Hawkwood

 

Posted

Since the drop rate for Incarnate trial components was adjusted in the April 26th patch, I've been collecting data on every trial I've run. A little while ago I finished trial #100, for an average of 1.28 trials per day, and I want to share the information I collected. The raw data is here. In total, I ran 48 BAFs, 28 Lambdas, and 24 Keyes.

Component Drops

We know there is a threshold of activity below which a player only gets 10 threads, and above that they get a Component. Assuming that there's only one threshold (you either get threads, or a component of random quality), we can gather drop rates from it. And my numbers are...



For me, over 100 runs, the drops heavily hinted at 10% Very Rares, 20% Rares, 30% Commons, and 40% Uncommons. The drops were somewhat streaky, though: I got three Very Rares in a row, and two Rares in a row twice.

Note that I was using two characters to run trials, and one of them got 8 Very Rares, and the other got 2, both running a very close number of trials. This is a prime example of the RNG being random; all comes down to luck. I went 30 runs without a Very Rare, and a the moment it's been 19 runs since my last Very Rare.

Completion Time

I measure the time from the first "radio transmission" to the moment the final AV is defeated. I don't use zoning into the map as a starting point because then the opening cutscene time would appear in some runs but not in others, screwing up the data; also, not everybody zones in at the same time. The first speech bubble gives me a good consistent point in which to begin measuring.



BAF: The fastest BAF run took 13:34; the worst, 35:33. Don't ask about that one. It was a horrible, horrible situation and I swore to never speak of it again. The average time to complete was 17:45; the median time, 17:09.

Lambda: The fastest run took 19:33; the slowest, 34:06. The average completion time was 25:08, and the median, 24:26. While it is longer than BAF, it's not that much longer; and usually, at least on Freedom, people run both back to back.

Keyes: Oh boy. Keyes. The fastest run took 31:46, and the slowest, 54:20; the average time to complete was 41:19, and the median, 41:15. If run with people who know exactly what to do, sub-40 minute times are the norm. If you have to explain things to people, better schedule an hour for this trial.

I sorted the data in order to show the "time trends" better:



The worst BAF runs will still be faster than the best Lambda runs (I'm still not talking about that extremely bad run that soils the graph) and the worst Lambda runs will be faster than the best Keyes runs. It's no surprise that broadcast in Pocket D sees a lot of "looking for anything but Keyes".


www.SaveCOH.com: Calls to Action and Events Calendar
This is what 3700 heroes in a single zone looks like.
Thanks to @EnsonsDeath for the GVE code that made me VIP again!

 

Posted

Quote:
Originally Posted by Golden Girl View Post
I just knew that someone was going to do a chart like this eventually
Would have been a lot earlier if I didn't take a complete break from COH in May. I'm about due for another break now; got some games in the last Steam sale that are waiting for my attention.


www.SaveCOH.com: Calls to Action and Events Calendar
This is what 3700 heroes in a single zone looks like.
Thanks to @EnsonsDeath for the GVE code that made me VIP again!

 

Posted

The run-time dichotomy is a big thing imo.

Such that I think that perhaps it needs to have it's reward readdressed, such as a better reward rate [ie, more likely to get Rare/V.rare whatever] or an increased number of awards [it takes twice as long, give it double awards].


Let's Dance!

 

Posted

Very interesting...

Kudos on all of the work putting this information together


Andy Belford
Community Manager
Paragon Studios

 

Posted

Quote:
Originally Posted by reiella View Post
The run-time dichotomy is a big thing imo.

Such that I think that perhaps it needs to have it's reward readdressed, such as a better reward rate [ie, more likely to get Rare/V.rare whatever] or an increased number of awards [it takes twice as long, give it double awards].
The average inf and iXP per minute would also be interesting; especially since i suspect that Keyes will also show a much lower rate of iXP and inf gain.

(Unfortunately iXP is very badly borked right now, so that will have to wait. i have Brute who has run at least nine trials in the last two weeks and has yet to unlock a single power. i already had slotted the Alpha before i started running the trials.)
Edit: Judgement is at 38% and Interface is at 40%. Without burning a lot of threads and Astral merits it will take another couple of weeks to unlock just those two. OTOH i will have components piled to my neck for making the powers by the time i do unlock them, so that's a plus i guess.


Dr. Todt's theme.
i make stuff...

 

Posted

Quote:
Originally Posted by Schismatrix View Post
Unfortunately iXP is very badly borked right now, so that will have to wait. i have Brute who has run at least nine trials in the last two weeks and has yet to unlock a single power. i already had slotted the Alpha before i started running the trials.
I can't gather iXP stats because my characters already have everything unlocked. I don't have the patience to wait for iXP, I just burn Threads into iXP. I could gather Influence gains, though; but I never cared much about influence gains from defeating enemies. The market is where all the influence is at.

Edit: logs for the last 3 days parsed (sadly I already deleted everything else). The numbers are:

1,566,948 BAF*
1,565,836 Lambda*
1,887,800 BAF
2,638,230 Lambda
1,411,972 BAF*
1,624,462 Lambda*
1,287,620 Keyes
2,395,050 Lambda
1,190,609 BAF

(*) Supergroup mode was enabled in this character; I doubled the inf gains to compensate.

The influence gains jump around too much to be of any use. Remember that influence comes from defeats, and it's up to the team if they want to clear the Lambda road or just head inside, or the amount of ambushes in BAF if the fight drags long, or the number of respawns in Keyes. The amount is not very significant in any case.


www.SaveCOH.com: Calls to Action and Events Calendar
This is what 3700 heroes in a single zone looks like.
Thanks to @EnsonsDeath for the GVE code that made me VIP again!

 

Posted

Quote:
Originally Posted by Leandro View Post
I can't gather iXP stats because my characters already have everything unlocked. I don't have the patience to wait for iXP, I just burn Threads into iXP. I could gather Influence gains, though; but I never cared much about influence gains from defeating enemies. The market is where all the influence is at.
i always burn threads to xp as well whenever i get the chance, even before ixp was borked the gain rate still sucked, i think it might improve when it gets fixed though


 

Posted

Quote:
Originally Posted by Schismatrix View Post
The average inf and iXP per minute would also be interesting; especially since i suspect that Keyes will also show a much lower rate of iXP and inf gain.

(Unfortunately iXP is very badly borked right now, so that will have to wait. i have Brute who has run at least nine trials in the last two weeks and has yet to unlock a single power. i already had slotted the Alpha before i started running the trials.)
It could be interesting, especially since there is more spatial arrangement going on with Keyes than slaughterfesting.

Quote:
Originally Posted by Leandro View Post
I can't gather iXP stats because my characters already have everything unlocked. I don't have the patience to wait for iXP, I just burn Threads into iXP. I could gather Influence gains, though; but I never cared much about influence gains from defeating enemies. The market is where all the influence is at.
Perhaps thread count instead?


Let's Dance!

 

Posted

Quote:
Originally Posted by reiella View Post
Perhaps thread count instead?
Again, that depends on the number of defeats. A Keyes trial where people kill everything in the bunkers to get to the glowies will have less drop chances for threads than a Keyes where an illusion troller (that is, me) sneaks through the back door and gets all the temps without killing anything. Also, threads are divided among the entire league; maybe I got 2 threads but someone else on the league got 7, in the same trial. Random drops + Variable number of enemy defeats = No good data.


www.SaveCOH.com: Calls to Action and Events Calendar
This is what 3700 heroes in a single zone looks like.
Thanks to @EnsonsDeath for the GVE code that made me VIP again!

 

Posted

Very cool, Leandro. Thanks for taking the time to put this all together!

-Buxley



"Don't cry because it's over, smile because it happened." -- Dr. Seuss

 

Posted

Nice work.

Math work follows that probably has some errors: if the numbers are really 10% for a Very Rare, it means that in 10 runs you would only have around a (9.^10) 35% chance to not get a V Rare. If those odds are real, then in terms of a single Very Rare, it is actually better strategy to run the BAF over and over than chase Empyreans. If you took an average amount of time to run each of the 3 trials every day based on these times, a Very Rare in Empyraen would be worth about 84 minutes of mission time across 10 days, or the equivalent of 14 hours of playing trials (not including set up time). If you ran the BAF 40 times you'd have a less than 1% chance of not having your V Rare within 11 hours, and some chance of getting more than one. Of course, you can do both.

However, this also means that after 40 runs around 1% of players will not have found a V Rare. Multiply that by the number of people playing and you've got your conspiracy theories right there, and some (IMO somewhat justifiably) annoyed players.

I'm still curious whether team performance changes the odds and in what way.


 

Posted

Quote:
Originally Posted by Leandro View Post
Again, that depends on the number of defeats. A Keyes trial where people kill everything in the bunkers to get to the glowies will have less drop chances for threads than a Keyes where an illusion troller (that is, me) sneaks through the back door and gets all the temps without killing anything. Also, threads are divided among the entire league; maybe I got 2 threads but someone else on the league got 7, in the same trial. Random drops + Variable number of enemy defeats = No good data.
Hmmm, very true. Without putting together recurring teams for the purpose of running each trial the same way and then recording stats for all teammates... yeah, no.

Thank you very much for collecting and analyzing the drop data. That's quite interesting in itself.


Dr. Todt's theme.
i make stuff...

 

Posted

Quote:
Originally Posted by Oedipus_Tex View Post
However, this also means that after 40 runs around 1% of players will not have found a V Rare. Multiply that by the number of people playing and you've got your conspiracy theories right there, and some (IMO somewhat justifiably) annoyed players.
I used two characters to run the trials; one of them got 2 Very Rares, and the other got 8. The character that got 8 already had everything at T4 before April 26th. As a result, I now have a character that still needs 2 Very Rares, and another that has 6 Very Rares and nothing to do with them. This mildly annoys me.

If the two characters had belonged to different people, one would have seen about 4% VR drop rate, and the other 16%.


www.SaveCOH.com: Calls to Action and Events Calendar
This is what 3700 heroes in a single zone looks like.
Thanks to @EnsonsDeath for the GVE code that made me VIP again!

 

Posted

Quote:
Originally Posted by Leandro View Post


For me, over 100 runs, the drops heavily hinted at 10% Very Rares, 20% Rares, 30% Commons, and 40% Uncommons.
The sad part is that the ratios should be ...
10% Very Rares
20% Rares
30% Uncommons
40% Commons

This would explain why every character I've run trials with always winds up FLOODED with piles and piles (and piles!) of Uncommons that I'm never going to need or use ... while I must go begging for Commons to finish out my Tier 1/2/3 slots with (and am pretty much ALWAYS forced into converting Threads for).


It's the end. But the moment has been prepared for ...

 

Posted

Quote:
Originally Posted by Redlynne View Post
The sad part is that the ratios should be ...
10% Very Rares
20% Rares
30% Uncommons
40% Commons

This would explain why every character I've run trials with always winds up FLOODED with piles and piles (and piles!) of Uncommons that I'm never going to need or use ... while I must go begging for Commons to finish out my Tier 1/2/3 slots with (and am pretty much ALWAYS forced into converting Threads for).
i downgrade most of my uncommons into common components. Much more useful.


Dr. Todt's theme.
i make stuff...

 

Posted

Seems like, from a casual glance at the charts, that the information you've gathered is exactly what people should have expected to see. Over enough runs, the drops find their norms, just like a good, weighted table drop should do.


Loose --> not tight.
Lose --> Did not win, misplace, cannot find, subtract.
One extra 'o' makes a big difference.

 

Posted

Quote:
Originally Posted by Redlynne View Post
T
This would explain why every character I've run trials with always winds up FLOODED with piles and piles (and piles!) of Uncommons that I'm never going to need or use ... while I must go begging for Commons to finish out my Tier 1/2/3 slots with (and am pretty much ALWAYS forced into converting Threads for).
Begging? They aren't tradable

Plus, why not just downgrade the uncommons?


@Golden Girl

City of Heroes comics and artwork

 

Posted

Why are uncommons more common than commons?

That has been my experience too.


[U][URL="http://boards.cityofheroes.com/showthread.php?t=251594"][/URL][/U]

 

Posted

Awesome, awesome information, Leo. Thank you for all the hard work and analysis.




And for you, you know who you are, who just have to come in here and dance around making worthless vacuous waste posts--for God's sake, can't you just stop your mouth from moving for ten seconds?


There are no words for what this community, and the friends I have made here mean to me. Please know that I care for all of you, yes, even you. If you Twitter, I'm MrThan. If you're Unleashed, I'm dumps. I'll try and get registered on the Titan Forums as well. Peace, and thanks for the best nine years anyone could ever ask for.

 

Posted

Quote:
Originally Posted by Oedipus_Tex View Post
I'm still curious whether team performance changes the odds and in what way.
Take this for whatever its worth, but I believe that the "intrinsic" rate of commons is actually higher than uncommons, but the problem is that the average team/league participation bonus pushes uncommon over common in actual practice. Since the vast majority of players do more than enough to qualify, and leagues composed of such players are likely to thus score at least nominal bonuses, that's why "uncommon" is so common.

I believe this is part of the source for the devs' original miscalculation on how "common" the uncommon drops were at release. They didn't factor in the fact that the average league was going to have a consistently non-zero random roll bonus.

I will also say something else I really cannot elaborate upon, but those numbers are consistent with what I believe are the drop percentages that good leagues will tend to get assuming very good (but not necessarily difficult to get) random roll bonuses. This would suggest that its not hard to basically saturate the best possible league bonus with normal (but reasonably good) league play in general.


[Guide to Defense] [Scrapper Secondaries Comparison] [Archetype Popularity Analysis]

In one little corner of the universe, there's nothing more irritating than a misfile...
(Please support the best webcomic about a cosmic universal realignment by impaired angelic interference resulting in identity crisis angst. Or I release the pigmy water thieves.)

 

Posted

Quote:
Originally Posted by Leandro View Post
Assuming that there's only one threshold (you either get threads, or a component of random quality), we can gather drop rates from it.
I don't think we can make that assumption. Remember what they said a while back:

Quote:
Originally Posted by Baryonyx View Post
After evaluating the data since the launch of Issue 20 we’ve found that players are being rewarded with the Uncommon table more frequently than we anticipated.
Just Uncommon? And it's more frequently than the flat-rate, weighted distribution they were anticipating?

I don't buy it. Participation plays a factor in what Components are rolled.


 

Posted

Quote:
Originally Posted by GuyPerfect View Post
I don't buy it. Participation plays a factor in what Components are rolled.
The thing is, I've been in leagues where I was on autopilot and barely doing anything, and I got a Very Rare out of it; while in other leagues I was going crazy trying to do thirty things at once, and I got commons. Whatever way participation is being weighted, it doesn't seem to do all that much.

However... there's something that really bugs me in the data, and maybe Arcanaville can say me if it's just random being random, or there's actually something here. Sort the drops by trial, and we get these numbers:

BAF: 15 common, 18 uncommon, 13 rare, 2 very rare.
Lambda: 8 common, 15 uncommon, 3 rare, 1 very rare.
Keyes: 8 common, 7 uncommon, 2 rare, 7 very rare.

This could be just random being random, but... yeah. Keyes dropped a Very Rare on me 1/3 of the time. It really bugs me.


www.SaveCOH.com: Calls to Action and Events Calendar
This is what 3700 heroes in a single zone looks like.
Thanks to @EnsonsDeath for the GVE code that made me VIP again!

 

Posted

Quote:
Originally Posted by Leandro View Post
The thing is, I've been in leagues where I was on autopilot and barely doing anything, and I got a Very Rare out of it; while in other leagues I was going crazy trying to do thirty things at once, and I got commons. Whatever way participation is being weighted, it doesn't seem to do all that much.

However... there's something that really bugs me in the data, and maybe Arcanaville can say me if it's just random being random, or there's actually something here. Sort the drops by trial, and we get these numbers:

BAF: 15 common, 18 uncommon, 13 rare, 2 very rare.
Lambda: 8 common, 15 uncommon, 3 rare, 1 very rare.
Keyes: 8 common, 7 uncommon, 2 rare, 7 very rare.

This could be just random being random, but... yeah. Keyes dropped a Very Rare on me 1/3 of the time. It really bugs me.
Probably random. At this point I've gotten six very rare drops total. Four have been on Thursdays. Statistically speaking, that's pretty weird because most of my trials are run on weekends. In fact, I've only run about fourteen trials on Thursdays, for a 4 out of 14 very rare rate. Of course, there's no possible way the participation system could be weighted for Thursdays, so that's just what happens when you stare at raw data too long: random chance says you will find patterns if you look long enough.


[Guide to Defense] [Scrapper Secondaries Comparison] [Archetype Popularity Analysis]

In one little corner of the universe, there's nothing more irritating than a misfile...
(Please support the best webcomic about a cosmic universal realignment by impaired angelic interference resulting in identity crisis angst. Or I release the pigmy water thieves.)