Data Analysis #3: The arrival of Statshark answers some old questions

Maybe you post 1.0-4.3 data, why you hiding it? Doesn’t suit your theories and conclusions?

image
image
image

In order: Naval, Ground, Air.

Air SB is not represented as we have a very different reward structure (see: Community Bug Reporting System). We have the same score/action as ARB, but instead we’re rewarded only for activity% over 15 minute cycles. Furthermore, activity% caps out at 92% regardless of how much you exceed 1050 score, giving a fixed maximum potential reward that’s around 2000 RP without premium in a 136% RP economy plane for 1050, 1950, 2650 score alike (that’s ~2 kills + 1 crit, 4 crits + 3 crits, 6 kills + 1 crit, or whatever other rough combination of captures, kills and assists and crits), with a bonus ~+500 RP if you survive the full 15 minutes and land safely PROVIDED you got 2000 RP (ergo, you receive 80% of the 92% reward, and must land to receive the remaining 20% of 92%.)

You get 0.03 more RP per second in Naval AB flying planes compared to Air AB, and 0.07 more RP compared to GAB.

In contrast, a kill is worth 48 RP in Air AB and there’s guaranteed by definition to be more planes for you to shoot and kill, whereas in GAB one kill is worth 45 RP and in NAB only 40 RP.

Your claim is only very technically and marginally correct (there’s a 0.03 RP/second advantage ignoring however the activity% function is parametrized), at cost of -8 RP per kill compared to AAB.

Nowhere is it backed up regarding “you get triple the amount as in AAB” however -

Nor is “NRB” gives less RP for using planes at all substantiated.

Nota bene:

For air kills, reward is represented as 2 values unless you had scored a critical + crash, pilot kill or tail separation:

You receive 80% of your reward under “Severe Damage” heading and 20% for whatever they call “Kill confirmed” to be. If kill is confirmed by an ally you, don’t get the remaining 20% and instead they receive 40% of the reward for kill credit (this occurs if they shoot a plane with a deal engine, more than 50% of the wing cut off, all control surfaces disabled/removed or… black horizontal stabilizer.)

This might be the cause for confusion as a blackened horizontal stabilizer barely impacts flight performance with full-real controls in SB, much less with arcade boost and instructor in arcade so it’s very likely you scored 40% value kills in air arcade due to density of air combat and had assumed it was full reward. This is the only plausible hypothesis I can muster to justify your claim that NAB gives 3x the value of kills compared to AAB.

Example of kill credit being really weird in air battles. P-51C maintains near-full combat power (its right horizontal stabilizer IS shot off, but he's still turning and evading and diving.) has wing tip shot off, leading to me being rewarded an ASSIST when they die while a guy who damaged the horizontal stab long-since died being rewarded the kill. Preusmably, if I had had shot the pilot instead I'd have received 0.4x450=180 score for killing the P51C (4.0) rather than the expected 450 for a full kill (3.3 plane shooting 4.0 gets full score)

https://www.youtube.com/watch?v=05UvhsvAylE

Another possible explanation would be that you are shooting downtier aircraft more often in AAB and are suffering the BR correction’s consequences (which is probable due to no SP cost to spawning and lineup respawns meaning people may keep respawning to keep fighting in planes whose BR is a significant difference from yours):

image

Of course, if you collate sufficient data to provide tables on par of the linked article with proper methodology, I am willing to concede that the cited work is outdated in its reward quantities and requires updating. However, as of presently I trust its veracity given my own experiences with the score I earn for my actions lining up as expected.

Example of perfect line-up:
image

Fw190A4 (4.0) kills F6F-5 (3.7) through direct pilot snipe, resulting in no critical hit score being awarded. This is evidenced by the battle log reading "Shot down", rather than "Severely damaged" or "Critically damaged" in additional to the kill being credited while the visual model is still intact, rather than requiring a crash or bailout beforehand.

BR step difference is 0.3, therefore 450 x 0.986 = 443.7.
In game, I was awarded 444 score which is 443.7 rounded up, thus confirming that the formulae and scoring system remain applicable within ASB at least.

Yeah, this misunderstood fact underlying the original claim here was crew skill gain in AB is tripled compared to RB, because it’s an AB mode. Naval is the same as ground in that respect.

In every other sense, SL, RP, score, naval RB is the higher payout mode. Thankfully, all anyone has to do these days to know this is look up their favourite ship (or any other vehicle) on Statshark and toggle between AB and RB stats to see the differences in SL per game and RP per game, to see what they’re missing out on.

(Note however those totals are not inclusive of all end game rewards, your service records only totals the in-game stuff, so they’re an undercount.)

1 Like

Will be interesting to see if Naval remains dead even after the huge update it received.

1 Like

Still trying to pep talk myself into grinding the bismark and I even have a talisman on an aced sharnhorst.

Ships are really cool, but the gameplay is boring, and for a realistic game its the same handful of ships in multiple every match, no place for cruisers or destroyers, just fast coastal boats as a means to cap on 3cp maps.

1 Like

If someone is interested in seeing how effective the new Rank VII battleships were in June 2025, I have made Excel tables based on StatShark data.

Naval Arcade:

Naval Realistic:

4 Likes

Roma 8.3 btw.

1 Like

It’s fixed now. Thanks for pointing it out.

Full June statistic released. I wonder now Bruce will compare February to June as always? Or will it find some excuse? Cmon Bruce we counting on ya.

Graphs updated with June 2025 data:

Arcade and Realistic stats combined:

Even the major update couldn’t stop Ground and Air modes from losing players in June compared to May.

Naval gained a lot of players in June - clearly a result of the latest major update, which added iconic battleships.

3 Likes

Yes however Air and Ground lost way more than Navy’s gains… Navy’s gains however , might not really be …gains.
There are people that play everything (like me) and that we will play Navy doesn’t mean we won’t play Air or Ground.
The game lost people if anything… But it’s logical , it’s summer .

1 Like

This graph shows the number of players in battles. StatShark calls this data “Games Played”, but that name isn’t entirely accurate. If there is a battle with let’s say 12 playes and 20 AI bots, the “Games Played” stat will increase by 12. If one player plays 100 battles in a month, this stat will increase by 100 from this player alone. In short, it’s the total number of battles played by all human players (AI bots are not counted).

This is the best stat to show how popular the game mode is at a given time.

Yes, it’s visible in the stats. But if the latest major update didn’t matter, you would expect all 3 game modes to drop in the number of players in battles. But you see a big increase in Naval, so players clearly play Naval more than before.

It’s also reasonable to assume that if someone plays more Naval than usual, they can’t be playing Air or Ground at the same time, so those stats will drop. For example, if you normally play an average of 100 Air battles, 100 Ground battles, and 10 Naval battles each month, but in June you decide to play 100 Naval battles, it’s very unlikely you’ll still manage to play 100 Air and 100 Ground battles as well.

That’s likely part of the reason for the drop in Air and Ground stats. But there are clearly other factors as well. The latest major update also may have felt disappointing from the perspective of Air and Ground players. We also can’t forget about many bugs that were introduced, which had a major impact, especially on Air battles (e.g. malfunctioning missiles and guns).

Considering naval growth I think June is still not fully representative cause update landed on 25th. July will be way more interesting, given Naval event coming soon

1 Like

I’m more excited to look into combat stats of certain ships (hello Belfast) and overall, after recent BR changes

Considering coming naval event yep it is.

The game will decline if they won’t fix the crap they do.
Adding crap doesn’t work anymore. It works for a month (maybe) and then everything goes down again…
Now you see people playing naval. They will play for a month, they will get bored because it added nothing in gameplay. They won’t stay. Meanwhile without events (practically ^paying^ people) Air/Ground decline…

That’s why pervious event system with all types of vehicles was better it disperse players more equally.

Going from May to June there’s like 2m more sessions in Naval, meanwhile Ground and Air combined lost like 22m sessions. With this in mind, it’s safe to assume latter two modes lost most of that to other games or real life, and not Naval.

Since this update dropped late last month, seeing stats for July will be really interesting, as this is basically the best case scenario for Naval’s growth.
That mode received a massive update with decompression alongside it, while most Ground additions were lackluster to say the least and it was focused around the least played vehicle type.
Air didn’t have much more luck either.

So, at least in my opinion, if Naval’s sessions in July don’t spike like 2-3-4x in comparison to June it might mean that mode will never have a healthy playerbase.
Gaijin dealt perfect cards to Naval, we’ll see will it fold or not.

2 Likes

The correct month to compare with is March, the last one with a major update (and also a naval event).

It should probably be bad news in Gaijin HQ that after ALL that effort, they didn’t manage to get even 10% above the number of sessions they got in March for naval arcade. Statistically the only real growth they got above baseline for their naval efforts has all been in NRB, their most irrelevant non-sim mode (but still gaining now on a faltering AB).

Naval as a whole is still just a rounding error though.It’ll be interesting to see if it finally pulls itself above 2% of game sessions (AB/RB) in July.

The Naval story is a blip in the much more significant WT statistical trend we’re seeing, the growing dominance of ground RB as the “core mode” in the game’s later life now comfortably above 50% of all user sessions, and still growing, even against the seasonal summer drop-off trend. Maybe they should have tried to knock out WoT instead of WoWS.

You see it (and how consistent the player modes actually are relative to each other, single-vehicle events have only a minor effect here) when you look at Ground RB relative to the other modes over the last five months:

Spoiler

You can factor in how much of this is seasonal by looking at Steam users per day for the same period vs player sessions per day. While not a perfect comparison (Steam users could be falling for other reasons besides seasonal, and players and sessions are obviously different) it still somewhat shows that while the rest of the game is following the standard summer decline pretty closely, Ground RB is growing AGAINST the seasonal trend, 30% net growth June over Feb.

Spoiler

2 Likes

Just to manage expectations down a little… there were 6 days of new update in June out of 30. Even if ALL the growth shown in Poul’s graph happened in those 5 days, and that continues for the full July period, so people kept playing at the same intensity as they did in the first week of launch basically, that would still only be a 3x multiplier July over June at most. For 4x there would have to be massive positive word of mouth as well, a “rolling snowball” effect. I think if they get up to a 3x, equating to 5% of all random-match non-sim user sessions, or 13 million user sessions in absolute numbers, in July Gaijin should be absolutely ecstatic at that result, as statistically that would seem absolutely the most they should hope for.

The other interesting question with naval for me is whether AB and RB are inverting in popularity. AB fans should be a little concerned, because, while the data was partial, RB, despite still being the less popular mode (still at 40% of matches, although it’s been climbing lately), had significantly more (52%) of the new top tier BR (8.0-8.7) game sessions in those first six days, 75,512 vs 69,304. So the players that are trying the new top tier ships were tending to use more RB to do it. Given that for the last few months the only place where AB was gaining players WAS top tier, that could indicate any inversion is now getting accelerated by the update and the new ships, but it’s too early to be very confident in that… we’ll see in a month.

2 Likes