Last week I asked people on WOWs forum what they think about decline of DD population in random games. I had only limited data at that time (my own games for a single day), and that indicated significant decline in DD numbers.
Somebody posted link on MapleSyrup site (
http://maplesyrup.sweet.coocan.jp/wows/ranking/index.html) which has really interesting collection of historical statistic for WoWs. I don't know where this data comes from and how good it is, so take everything below with container of salt. As every data scientist will tell you – garbage in garbage out. Means everything below could be total bologna. However, we have no other data sources, so let's stick to MapleSyrup weekly statistics.
First of all, let's figure out some topics not directly related to CV rework. For example – people keep saying they are leaving game because something just happened. I'm playing since October and have seen few waves of rage quits like that. Is it actually happening? What's going on with players population in general?
MapleSyrup tracks number of accounts per region.
Picture 1: Number of Accounts https://imgur.com/LnMKSHl
Data shows steady growth in number of accounts across all markets for the last year. EU grows faster than any other region, RU – slower, but it's already largest market, so it could be closer to capacity.
There is no indication about account activity however. So let's check other stats, number of battles played.
NB: From now on I will only look into NA numbers (preliminary estimates show it reflects the whole picture quite well). If I'll have time later, I'll add all other markets.
Picture 2: Number of battles per tier
MS differentiate low tier ships (1-6) and high tier ships (7-10) into two different categories.
Number of battles played with high tier vessels is growing (17.2% increase per year), but number of low battles decreased by 46%! These trend numbers based on linear regression, so it's not that accurate, but still. If population is indeed moving to T7-T10 ships, that could create potential matchmaking issues for some tiers (hello, T8s!)
Picture 3: Total number of battles
Overall it looks like despite growing number of accounts on NA, number of battles goes down (-15% for the year). I'm looking forward to see same stat for other regions – if things are the same there, it's sign of trouble for the game.
Now let's slice the original question I asked myself – How CV rework changed players population? First of all, let's see how many players actually care about CV.
Picture 4: Players per Class, High Tier https://imgur.com/erxV2b0
That's interesting. Before rework high tier surface classes, especially BB and CA were significantly more popular than CV was. For every CV player we had on average 17.3 BB players! And after rework… it's the same, but for lesser degree. Even DD – least popular and most affected by recent changes in spotting class is 4.2 times more popular than CV among players.
For low tier ships it's similar, however, more favorable to CV:
Picture 5: Players per Class, Low Tier https://imgur.com/FP0CSYk
It's interesting to see how numbers of players for specific classes were affected by seasonal events. RN DD line launch made DD most popular class in low tiers and end of PEF marathon made people play low tier BBs to earn steel.
Let's also check how number of battles per class changed.
Picture 6: Battles per Class, High Tier
Here's where we can start asking questions about CV rework. First of all, DDs are indeed suffer from it. First week after 0.8 showed year-low number of battles played by DD, it's getting better after two hotfixes, but still not looking well. BB and CA population appears to be fully recovered after two weeks.
Picture 7: Battles per Class, Low Tier
Again, it's similar, but not as dramatic on lower tiers. One surprising data point though – weekly number of games played by CV on second week since 0.8 is only 1.6 time higher than it was during New Year holidays. In absolute numbers CV population didn't grow up as good as one could think. And absolute numbers for other classes suffer. In other words – some people play CV more than before, most people play game less!
Absolute numbers however are not really good if one wants to know what is likelihood of meeting certain class in random games. Let's recalculate number of battles per class into standard 12 people composition for Randoms.
Picture 8: Random game composition, High Tier https://imgur.com/8lgDFwk
Before rework it was one CV for every 5 games, give or take. Least favorite surface ship in high tiers was DD – 3 vessels per side in random on average. After rework it's 2 games of 3 with CV in expense of… DD again. By the way – see that spike in number of CV for first week of 0.8? It doesn't look good for future popularity of CV.
Picture 9: Random game composition, Low Tier https://imgur.com/pCxNcv1
T1-T7 bracket has 1.2 CV per side on average. If we'll remember that there are no CVs in T1-T3, that would mean about 2 CV per side on average for T4-T7. Which class suffered most from recent demographic changes? Surprisingly this time it was BB! Could be explained by pathetic AA for BB in T3-T5, making them helpless against T4-T6 CVs.
So, answering my original question about DD population in randoms: yes, in higher tiers average number of DDs per game substantially decreased. And overall player base is moving to high tier games. So one single class paid the price for dramatic balance change, caused by CV rework.
But it's not World of Destroyers! some people would say. Okay, sure. Let's see was it ever.
Picture 10: Average Damage per Battle per Class, High Tier https://imgur.com/8SCASVf
Picture 11: Average Damage per Battle per Class, Low Tier https://imgur.com/lmCkOjM
DD was and is class with lowest average damage output per game. In low tiers it's not that bad, CAs are only marginally better. In high tiers DDs are only outsiders. High tier CVs were not affected by rework at all – they are still compete for first place in damage meters with BBs. In low tiers CVs looks more balanced, but, again, it's not as relevant, as high tier numbers.
Last data point to look at – Average Kill / Death ratio per class.
Picture 12: K/D per Class, High Tier https://imgur.com/9nepskM
Picture 13: K/D per Class, Low Tier
Again, if anything, it's not World of Destroyers and it never was. Quite opposite – in terms of balance between risk and rewards CVs were and still are (at least on T7-T10 bracket) best ships to play.
And despite that they are still least popular class. And likely will be.
Feel free to make your own conclusions and ask your own questions.
Source: Original link
© Post "Statistical analysis of per-class balance in WoWs for 2018 and CV rework (mostly NA)" for game World of Warships.