Return of Reckoning

Wholdar wrote: Sun Jan 26, 2025 12:19 pm Update: Added Scenario-Wins graphs.
They're a bit boring since not many get concluded within each 5-minute period, but hey, the data's there!



Someone on Discord requested it.

Now, time to dig deeper into that API...

/j - https://maartenson.net/ror_graph.html

Great job! It would be nice to monitor activity and close discussion about <some fraction always wins>. Also highlights matchmaking problem, it impossible to write balancing algorithm when activity so low, without extra long wait time.

I think, for graphs "wins" above 24h it would be better to agregate data into chunks. Like for 7d graph agregate into 6h chunks, or summary in legend.

Also, interesting data is classes spread in each fraction. Can be helpful to analize how balance changes affects players behaviour. It just indirect marker and for deeper analysis need to have an access to raw data.

Don't know how balance team works, on one side you have to make a mathematical model of effectiveness and on another side you have to check changes analytically on live data. And, probably, you can be a hero for second part, if atm they don't have any analytics.

Activity is low bc the matchmaker is so poor used to be a super popular "mode" of WAR until it was redesigned to cater to 6-man easy mode.

The fact that a 6 man in q isn't queued for ranked at the same time says it all

Nice stuff with all the extra data added, and appreciate you hosting this!

Wholdar wrote: Sat Jan 11, 2025 1:47 pm

georgehabadasher wrote: Sat Jan 11, 2025 12:51 am Would it be possible to show population as a % of total players, similar to how the front page does it?

Hey!

I hope I understood you correctly - I changed the combined-total graph to reflect the % per realm, hope that works for you!



/j

This plot is going to be misinterpreted. I would suggest using a mirror histogram/density plot instead. Gl and hf with your project

Rubius wrote: Sun Jan 26, 2025 11:17 pm Nice stuff with all the extra data added, and appreciate you hosting this!

No worries, it's fun!

/j

NEW UPDATE!

I’ve added a new graph showing how “efficient” each realm is at killing in ORvR-T2+. We already track how many players each faction has in T2+ (though not all may be in RvR), and how many kills each side gets.

This graph calculates kills as a percentage of each realm’s T2+ population, then shows the difference.



It’s still not the whole truth—there’s no reliable way to know who’s flagged for RvR, or if they’re in an organized warband, their realm rank, gear level, if they’re blobbing, if one side is hiding in a keep, etc. But it should give a bit more insight into how effectively each realm is killing relative to its own population.

/j - https://maartenson.net/ror_graph.html?gt=kills_t2_pct

EDIT: Edited to make clear its a percentage, no point in going the "per 100 players" route..

Wholdar wrote: Mon Jan 27, 2025 11:25 am NEW UPDATE!

I’ve added a new graph showing how "efficient" each realm is at killing in ORvR-T2+. We already track how many players each faction has in T2+ (though not all may be in RvR), and how many kills each side gets.

This graph calculates kills per 100 players online in T2+ for each side, then shows the difference.



It’s still not the whole truth—there’s no reliable way to know who’s flagged for RvR, or if they’re in an organized warband, their realm rank, gear level, if they're blobbing, if one side is hiding in a keep, etc. - but it should give a bit more insight into how effectively each realm is killing relative to its own population.

/j - https://maartenson.net/ror_graph.html?gt=kills_t2_pop

I assume this is Kills per 100 player per 5 minutes, right? Think the time period the kills correspond to should be clarified somewhere in the graph.

leftayparxoun wrote: Mon Jan 27, 2025 11:49 am I assume this is Kills per 100 player per 5 minutes, right? Think the time period the kills correspond to should be clarified somewhere in the graph.

Thats correct! All data is gathered every 5 minutes. You're right in that that should be clear on the image itself. I'll fix!

/j

Wholdar wrote: Mon Jan 27, 2025 11:25 am I’ve added a new graph showing how “efficient” each realm is at killing in ORvR-T2+. We already track how many players each faction has in T2+ (though not all may be in RvR), and how many kills each side gets.

I think I confused even myself (happens alot at my age). Here's what I'm doing.

Population Data

I take the data from this graph, which shows the number of online players at rank 16+ for each faction over 5-minute intervals (fetched from RoR-homepage):



Kills Data

I also take the data from this graph, which shows the number of kills in T2+ for each faction over the same 5-minute intervals (fetched from RoR API):



Combining Them

For each 5-minute interval, we calculate kills-per-player for Order and Destro (kills ÷ population). We then subtract these two kill ratios to see which side is killing more efficiently (in terms of numbers of kills) relative to their own population. Positive values mean Order is performing better; negative values mean Destro is performing better. Finally, we multiply by 100 to convert that ratio difference into a percentage.



Now I hope it's crystal.

/j - https://maartenson.net/ror_graph.html?gt=kills_t2_pct

dizc0 wrote: Fri Jan 03, 2025 5:29 pm Are you only grabbing faction numbers or are you also getting class/race numbers? Would be fun to see a graph over archetype/class popularity.

I dug into this a bit today. Wasn’t sure at first how to approach it, but I realized that since there are “skirmishes” in the API, I could scan the ones that ended in the last five minutes, count unique class numbers over that time period, and maybe put that into a graph. Possibly per tier as well.

Haven’t gotten to the graphing part yet, but I pulled data from the last day or so. It took a while, but it’ll be much faster if I run this every five minutes like the other stats, not sure how enlightening it would be though, guess I'll just have to do it to see how it turns out.

--- Class Summary ---
Overall unique character counts per class:
SHAMAN: 691
CHOSEN: 631
WARRIOR_PRIEST: 620
ARCHMAGE: 524
DISCIPLE_OF_KHAINE: 505
ZEALOT: 496
CHOPPA: 495
WHITE_LION: 489
SQUIG_HERDER: 487
MARAUDER: 483
KNIGHT_OF_THE_BLAZING_SUN: 479
RUNE_PRIEST: 473
SHADOW_WARRIOR: 460
ENGINEER: 456
WITCH_HUNTER: 450
BRIGHT_WIZARD: 445
BLACK_ORC: 424
SWORD_MASTER: 423
IRON_BREAKER: 386
MAGUS: 381
SORCERER: 373
WITCH_ELF: 373
SLAYER: 332
BLACK_GUARD: 324

Unique character counts per class for levels 1 to 15
SHAMAN: 239
CHOSEN: 233
SQUIG_HERDER: 221
WARRIOR_PRIEST: 205
ENGINEER: 202
MAGUS: 183
WITCH_HUNTER: 180
ZEALOT: 179
CHOPPA: 178
RUNE_PRIEST: 170
BRIGHT_WIZARD: 170
DISCIPLE_OF_KHAINE: 169
SHADOW_WARRIOR: 165
ARCHMAGE: 160
KNIGHT_OF_THE_BLAZING_SUN: 157
BLACK_ORC: 156
SWORD_MASTER: 149
MARAUDER: 147
SORCERER: 145
IRON_BREAKER: 143
WHITE_LION: 132
WITCH_ELF: 130
SLAYER: 129
BLACK_GUARD: 109

Unique character counts per class for levels 16 to 39
SHAMAN: 169
CHOSEN: 160
WARRIOR_PRIEST: 150
SHADOW_WARRIOR: 143
MARAUDER: 141
ARCHMAGE: 131
DISCIPLE_OF_KHAINE: 129
CHOPPA: 122
KNIGHT_OF_THE_BLAZING_SUN: 122
WITCH_HUNTER: 121
RUNE_PRIEST: 121
ZEALOT: 120
ENGINEER: 119
BRIGHT_WIZARD: 113
SWORD_MASTER: 113
WHITE_LION: 109
SQUIG_HERDER: 107
BLACK_ORC: 101
MAGUS: 100
SORCERER: 88
IRON_BREAKER: 87
SLAYER: 83
BLACK_GUARD: 83
WITCH_ELF: 82

Unique character counts per class for level 40
SHAMAN: 333
WARRIOR_PRIEST: 287
CHOSEN: 276
WHITE_LION: 269
ARCHMAGE: 262
DISCIPLE_OF_KHAINE: 226
KNIGHT_OF_THE_BLAZING_SUN: 225
ZEALOT: 225
CHOPPA: 218
MARAUDER: 215
RUNE_PRIEST: 207
SQUIG_HERDER: 187
BLACK_ORC: 185
SWORD_MASTER: 185
BRIGHT_WIZARD: 182
WITCH_ELF: 182
SHADOW_WARRIOR: 178
IRON_BREAKER: 173
WITCH_HUNTER: 171
ENGINEER: 163
SORCERER: 159
BLACK_GUARD: 149
SLAYER: 139
MAGUS: 116

/j - https://maartenson.net/ror_graph.html