miscellaneous

There’s probably a ton of variables to take into account when asking this question, but with the clan race this month having lots of competition, it made me wonder - what have been the most active months, whether based on bot activity (average bots online?) or like most energy gained in a month (buffs change this obviously)

Maybe @Ender has a way to kind of track this or can provide some cool numbers, especially with the new players coming in from some of the marketing and stuff Ed has been doing.

Also probably could go in a suggestion, but it would be cool if you could tag folks in posts with like @Ender or ^Fishwick or something, granted with there being like 8 active posters maybe it’s not the most exciting addition, but if/when global logins are enabled, it would be cool to be like @Nos and regardless of which of the 92000 bots he owns he would know he was mentioned.

I don't know if it's exactly what you're looking for, but you can see some graphs showing which time periods were the most active at https://bots4.net/online/history

Would be all for @Nos style interactions but imagine it would need some sort of top level account and bot management system which has been wanted for years but lots of work to actually do

Uh oh, you two are talking about topics near and dear to my heart and giving me an opportunity to ramble! I've been thinking quite a bit about a lot of this, happy to share my current thinking...

it made me wonder - what have been the most active months, whether based on bot activity (average bots online?)

A couple days ago, I noticed thirty-something bots online at one point and made me wonder the same thing: Has there been an upward trend lately?

Ideally the online history page that Fishwick linked could answer this. Unfortunately, the "online count" chart is basically useless. Although data is collected every minute, the chart shows a daily sample (i.e. 1 data point of 1440) from a drifting time of day. If you blur your eyes and look at multiple years you can kind of pick out trends, but even month-to-month analysis is impossible because of time-of-day bias. The "N day actives" charts are fine because they don't have time-of-day bias, but the chart interface could probably still benefit from being refreshed.

Because of the limitations of that page, I have explored rewriting it, but that's fairly involved and low priority. To get a quick answer, I actually happened to dump this data into a spreadsheet yesterday. Online counts only began being collected every minute in June 2016, so this data starts then, but it shows (a) average bots online each day and (b) max bots online each day. For both of these values, it also has a 7-day average (smooths out day-of-week bias) and 28-day average (even more smoothing).

Anyway, I'll cut to the chase:

• Max bots online: 2 days ago (March 23), we hit 35 bots online (!). The last time we had that many or more online was November 2024. Pretty cool.
• Trends: The 28-day average for both metrics are trending upward since the ad campaign began at the end of January. Max bots online 28d-average had been 15-16 bots or so and is now 23-24. Average bots online 28d-average had been 6-7 bots or so and is now 11-12.

I've also been watching players that found the game from ads:

• 592 bots created.
• 170 reached level 2+, 46 reached level 10+, 15 reached level 30+.
• 117 played 5+ mins, 52 played 30+ mins, 32 played 1+ hour, 8 played 5+ hours.
• 40 signup-to-lastactive 1+ hour, 23 1+ day, 13 3+ days, 7 1+ week.

All of these stats leave me "cautiously optimistic". They're certainly trending in the right direction (yay!), but there's still a ton that can be improved and optimized in terms of both the game (new player experience and retention) and the ad campaign (more/better conversion tracking, better keywords, better ad copy, etc.). Expect to see more in this direction over time.

or like most energy gained in a month (buffs change this obviously)

Not sure how well this captures this, but this is the sum of the monthly energy of the top 10 clans each month (ignoring tarnished results). But yeah, it's subject to bias from various system changes over time, Monthly Crescendo, etc.:

mysql> SELECT
    ->   year,
    ->   month,
    ->   SUM(energy_month) AS top_10_energy
    -> FROM history_clans2
    -> WHERE
    ->   `rank` <= 10 AND
    ->   is_tarnished = FALSE
    -> GROUP BY 1, 2
    -> ORDER BY 3 DESC LIMIT 10;
+------+-------+---------------+
| year | month | top_10_energy |
+------+-------+---------------+
| 2012 |     9 |      46270598 |
| 2012 |    10 |      46266705 |
| 2025 |     5 |      45185518 |
| 2025 |    10 |      41282868 |
| 2012 |     8 |      39741267 |
| 2024 |     5 |      38528434 |
| 2012 |     7 |      37817353 |
| 2012 |     6 |      36702352 |
| 2026 |     2 |      35817492 |
| 2024 |    11 |      35413776 |
+------+-------+---------------+

it would need some sort of top level account and bot management system which has been wanted for years but lots of work to actually do

I may end up eating these words, but the V1 of this system may be closer than you suspect...! It's something I've been actively working on lately (well, again, to temper expectations - the very first step). I've been working on the game's Discord integration, and to have that, I'd like to have some standard definition of a player's "identity" across bots, which requires some sort of top-level account object. More to come.

Does anyone know the background for the big scoring months in 2012? Was there a different energy system or just a super active race?

Back then the EPH wasn't capped at 60, so there was a large amount of passive energy generated by bots with decent wins and whatnot (including dumpers). So while the very active scorers are getting a lot more currently from crescendo, it would appear the lower half of the top 10 clans back then were generating energy comparable to the difference. Not to mention the roster being cut in half.

The month on top of Ed's list there was very active for the time, but make note of the energy offsets. Notably the amount of energy some of the top 10 clans have while still having a neutral 1.0. Some bots getting over 200k idle energy a month if you take a look at the individual list.

Matter of fact, looking again I see all those 2012 months in the top 10 are consecutive, June through October. March that year was the month Ed first tested the uncapped EPH and I guess it had a more linear increase considering the energy bots like Myriad, Rene and Neps had. Don't recall specifics in April, but the energy offsets look more normal. Perhaps it got tweaked mid-month since May is when all the bots start having relatively low offsets and were likely still changing strategy to utilize the system, which led to the following 5 months being in that top 10 list. After that in November, based on the energy total of clans that month, it must have been when Ed changed it to the hard cap of 60 EPH that we're familiar with today.

I believe the initial testing in March is when there was an extremely high transfer rate too, basically like a crescendo buff on crack. That got tamed down within the same month, don't recall it lasting more than a week or two.

On the subject of stats, are there any numbers on browsers/device? I’m pretty confident that I’m in the minority by mostly playing on mobile, especially since most players generally multi bot, but maybe there’s shifts over time, or with the new players

With a disclaimer that these stats are possibly inaccurate (lots of spam in GA4, hard for me to filter to actual players), here are some stats from the past 28 days, I think filtered to users that have logged in:

• Device: 81% desktop, 16% mobile, 3% tablet.
• Browser: 63% Chrome, 12% Safari, 11% Edge, 9% Opera, 5% Firefox.

So yeah, mobile isn't super common, but also not totally uncommon either.