Just Curious

[Darvin]

I'm curious as to some statistics that the devs are tracking, if they'd be willing to share:

1) Relative popularity of each of the Pactmaker's pacts

2) Most/Least played dungeons

3) Total number of successful dungeon runs / total number of attempted dungeon runs (ie, global player success rate)

4) Relative popularity of the seven races

5) The average number of dungeon runs before the user defeats Horatio in hard Gaan-Telet.

Don't know if you're tracking all this stuff, but I'd be interested to know if you are.

[dislekcia]

1. Nope, not tracking boons that solidly yet. Could probably expand on them but not sure about the benefit to the game right now - trying not to make large changes in design anymore.

2. Tracked, but requires lots of data manipulation to find out. Will probably do this eventually when I have a week to actually crunch the whole dataset after I've pulled it down.

3. Total attempted runs since we started tracking: 650634. Total victories recorded: 142132. Total runs fled from: 137316. (Total time played is trickier and again, requires me poking the complete dataset pretty hard)

4. Runs started with:
Human: 379212
Elf: 57855
Dwarf: 36803
Halfling: 61696
Gnome: 61768
Orc: 34048
Goblin: 15572

5. 3790 runs started on Tower of Gaan-Telet, 241 victories recorded. This may be only the vicious version though - will have to check up on proper dungeon names on our side. Not sure if story-win events are recorded right now. That is still out of a much smaller pool than the total runs - dungeon location tracking was only added about halfway through the logging process.

[Lujo]

4. Runs started with:
Human: 379212
Elf: 57855
Dwarf: 36803
Halfling: 61696
Gnome: 61768
Orc: 34048
Goblin: 15572

O.O

Never in a million years would've guesst the short buggers were so popular. And this is coming from a guy who never had a DD binge which didn't end up with a serious CP system abuse. If this is "all time" statistics, it would make sense, but small wonder the general public finds the game hard

5. 3790 runs started on Tower of Gaan-Telet, 241 victories recorded. This may be only the vicious version though - will have to check up on proper dungeon names on our side. Not sure if story-win events are recorded right now. That is still out of a much smaller pool than the total runs - dungeon location tracking was only added about halfway through the logging process.

Well, depending on when the tracking started, this is waaaaay off, since that place was crashing since forever. I would've bumped the win number up by a good 50 on my own if it wasn't for the crashes.

[Darvin]

1. Nope, not tracking boons that solidly yet. Could probably expand on them but not sure about the benefit to the game right now - trying not to make large changes in design anymore.

Boons are not so much a concern, because for most deities they are not mutually exclusive. With GG, whether I used humility or not has no bearing on my ability to use cleansing. With Pactmaker, picking a warrior pact precludes a body pact, and vice-versa. As a result, in-depth analysis of how his pacts are being used is far more meaningful.

My only serious balance concern is spirit pact, which I personally have not used ever since the orc nerf. Beyond that, I'm just curious.

2. Tracked, but requires lots of data manipulation to find out. Will probably do this eventually when I have a week to actually crunch the whole dataset after I've pulled it down.

Cool; I'm curious to see where most players are hanging out.

3. Total attempted runs since we started tracking: 650634. Total victories recorded: 142132. Total runs fled from: 137316. (Total time played is trickier and again, requires me poking the complete dataset pretty hard)

Does that also include disconnects where the user began the run and then navigated away from the page? There's enough guess-work here trying to filter out scum-runs.

4. Runs started with:
Human: 379212
Elf: 57855
Dwarf: 36803
Halfling: 61696
Gnome: 61768
Orc: 34048
Goblin: 15572

I'm actually surprised to see gnomes and halflings so low compared to humans. I would have thought beginners would be flocking to the potion-races in droves. Otherwise, not at all a surprising breakdown.

Never in a million years would've guesst the short buggers were so popular. And this is coming from a guy who never had a DD binge which didn't end up with a serious CP system abuse. If this is "all time" statistics, it would make sense, but small wonder the general public finds the game hard

The issue is that on the normal-difficulty 1-boss dungeons, a halfling or gnome gives you enough spiking power to absolutely steamroll just about any boss with just about any class. However, this damage-spiking becomes increasingly difficult to leverage as the dungeons get more intense and endurance-based. Ten healing potions will let you steamroll the Iron Man, but it won't necessarily get you past Anoobis.

[lucad]

Stats! That's cool as heck. Only thing I'd ask still is the popularity of the classes...

I couldn't help to bite the calculator and find a few numbers within yours. A little heads up, though, I don't know how the numbers are being generated but the fact that not all races are available from the beginning obviously give the early races a boostie. Nevertheless, if anyone's interested:

Humans hold more than half of the gameplay popularity. Since 58~% of the games thrown around are on their hands, it actually means they are played more than the other races combined. Leaving Humans out of the equation, 21~% of the games are on shorties hands (Halflings, Gnomes, and Goblins), while another 19~% is on the hands of the tall ones. Or medium sized. (Dwarves, Orcs and Elves)

So, in other words, in every other ten games, 6 humans, 1 halfling, 1 gnome and 1 elf are rolled. The last spot is usually a gruesome outcome between Orcs and Dwarves, with Goblins now and then taking the fight to themselves.

On the outcome side, only 22~% of the dungeon raids end up with the boss being dead, while a little less than half (43~%) of them end with the hero alive (Fleeing/Winning). The gruesome obviousity is that 57~% or so of the raids cost the life of our adventurer (Which, again, usually is a probable human).

So, in little numbers:
New to Desktop Dungeons? Prepare to lose 8 of every 10 games.

PS: Couldn't help but notice: The added sum of runs by race tallies up to 646954, while the number of raids disclosed is a little higher than that. Calculations where made with the first number for races while outcomes used the 65k-ish.

[Darvin]

On the outcome side, only 22~% of the dungeon raids end up with the boss being dead, while a little less than half (43~%) of them end with the hero alive (Fleeing/Winning). The gruesome obviousity is that 57~% or so of the raids cost the life of our adventurer (Which, again, usually is a probable human).

Until dislekcia clarifies on the issue of dropped connections, keep in mind that a lot of those "deaths" may actually be people who just quit by closing the browser window... or it's possible that disconnects don't count at all, in which case the total number of games played is actually much higher.

So, in little numbers:
New to Desktop Dungeons? Prepare to lose 8 of every 10 games.

It's worse than that; that's your community-wide average, including the experts who win 9/10 dungeon runs when playing outside of vicious. New players are going to be significantly below average.

That's partly why I want to see which dungeons get the most activity. If there's a lot of vicious dungeon runs, that would explain a lot of deaths, but if the activity is predominantly on the easier levels then these kinds of statistics paint a pretty dire picture.

PS: Couldn't help but notice: The added sum of runs by race tallies up to 646954, while the number of raids disclosed is a little higher than that. Calculations where made with the first number for races while outcomes used the 65k-ish.

Those would be the monster races. Since monster races are only unlocked by beating vicious dungeons, very few players have access to them so we'd expect them to make up a tiny number of the total dungeon runs.

[Lujo]

Oh, i didn't count the number of digits on humans, d'oh!

Wow.

Also interested in what dungeons get played the most, but if these numbers combined with old god usage stats give me a clue of anything it's that most people apparently give the game a try and give up once the difficulty starts gettign ridiculous, which is probably rather soon for regular people. I might be wrong, though.

[harfatum]

That's partly why I want to see which dungeons get the most activity. If there's a lot of vicious dungeon runs, that would explain a lot of deaths, but if the activity is predominantly on the easier levels then these kinds of statistics paint a pretty dire picture.

I'll bet there's quite few vicious dungeon runs, seeing as it takes quite a while to unlock them (and much longer if you are doing all the quests along the way / exploring all directions evenly).

[q 3]

Does each restart add to the number of runs? If so, I might be responsible for something like 10% of all vicious Tower runs.

[Darvin]

Oh, i didn't count the number of digits on humans, d'oh!

I did the same at first. Then I realized that it was in an entirely different order of magnitude.

I'll bet there's quite few vicious dungeon runs, seeing as it takes quite a while to unlock them (and much longer if you are doing all the quests along the way / exploring all directions evenly).

Oh, I concur. The question is how much fewer, and what kind of slope are we looking at along the transition?

Does each restart add to the number of runs? If so, I might be responsible for something like 10% of all vicious Tower runs

I would not be surprised if a dozen or so players are responsible for 90% of vicious Gaan-Telet runs. Most people would take one look at that level and run away screaming.