Analyzed: Tanki's Big Experiment

[Person_Random 4,991]

No, no, get out from under the table, we are not back in your nightmare science class, but indeed, Tanki did do an experiment. (No, the scientific method was not involved) We’ll be breaking down the main points and getting to specifics.

____________________________________________________________________________________________________________

 

Abstract

In case you aren’t aware, Tanki did some sort of experiment where some gameplay factors were affected. This change lasted for two weeks but felt like two years since I was playing the entire time. And of course, there were quite some significant effects on how Tanki gameplay changed. Let’s break everything down from the experiment one by one, starting with the purpose of the experiment.

 

Purpose

Many have been complaining about battles in maps too large for a team to play on, and it just isn’t fun. Some people wanted the maps to be removed from the MM maps list. In turn, these requests were granted, but some other changes had to be done to ensure that balance was maintained. But before they became final, Tanki wanted the players to test it out, and here we are.

 

Experimental Design

Concept: The goal is to see if the smaller 6v6 gameplay makes the game more team-oriented. By answering the analysis questions below, I’ll be able to give some insight on whether I believe this change allowed team gameplay to shine or not.

Summary: To track how the experiment affects me as a player, we’ll divide this experiment of the Experiment into two groups: the control group (pre-experiment game situation) and experimental group (Experiment game situation). The data numbers from each group will be recorded as well as a general overview of the game experience.

Procedure:

1. Log in to Tanki Online.

2. Play some MM battles before the experiment. Record full data in the following scenarios: complete win, draw/close match, complete loss.

3. Change some gameplay balance factors in the game:

• Hulls’ HP: +100%
• Turret Energy: +100% (Firebird, Isida, Freeze, Ricochet, and Twins)
• MM Map Pool: - Мagistral, Highways, Massacre, Stadium, Solikamsk, Brest, Wolfenstein, and Yorkshire
• Max Players on a team: -25% (8 → 6)
• Battle duration: -29% (7 → 5 minutes)
• Overdrives:
  • Active Time: +100% (Viking)
  • Damage: +100% (Wasp)

To check this out in action, I have linked a video that will help you grasp the meaning of this. I would have liked a side-by-side comparison, but none turned up and this is the best I could find. It’s still a decent comparison, so definitely watch if you need clarification. Check it out here.

4. Play some MM battles during the experiment. Record full data in the following scenarios: complete win, draw/close match, complete loss.

5. Repeat Step 4 multiple times to ensure a complete view of the experiment.

 

Results

Let's divide them into two categories and three subcategories: Pre-experiment and Experimental Conditions. I'll also add win, loss, and draw/close game situations.

Pre-experiment:

Spoiler

Loss:

 

Close Win:

 

Definitive Win:

Experimental Conditions:

Spoiler

Loss:

 

Draw:

 

Win:

 

 

I also analyzed the results and compiled them here on a table if you're too lazy to scroll through the slideshow. Enjoy the number crunching!

Pre-experiment:

Battle	Scenario	Score	Crystals	Kills	Assists	Deaths	K/D	Other (e.g. caps)
1	Loss	255	634	12	0	8	1.5	1 cap
2	Win (close)	310	970	14	2	6	2.33	1 cap, 1 cap assist
3	Win	385	694	23	3	7	3.29	1 cap assist
Avg	-	316.7	763	16.3	1.67	7	2.37	-

 

Experimental Conditions:

Battle	Scenario	Score	Crystals	Kills	Assists	Deaths	K/D	Other (e.g. caps)
1	Draw (TJR)	140	392	6	5	4	1.5	1 JGR kill assist
2	Loss	120	334	6	6	5	1.2	-
3	Win	270	712	12	3	3	4	1 cap
Avg	-	176.7	479.3	8	4.67	4	2	-

 

Side-by-side Comparison:

Stat	Pre-experiment	Experimental	%Change	Exp. (adj.)	%Change (adj.)
Avg. Score	316.67	176.67 (-140)	-44.21%	247.34 (-69.33)	-21.89%
Avg. Crystals	763	479.3 (-283.7)	-37.18%	671.02 (-91.98)	-12.06%
Avg. Kills	16.3	8 (-8.3)	-51.88%	11.2 (-5.1)	-31.29%
Avg. Assists	1.67	4.67 (+3)	+179.64%	6.54 (+4.87)	+291.62%
Avg. Deaths	7	4 (-3)	-42.86%	5.6 (-1.4)	-20.00%
Avg. K/D	2.37	2.0 (-0.37)	-15.61%	-	-

Obviously, I had fewer kills, score, and crystal rewards with shorter battles, even when adjusted, but that could easily be attributed to the shorter battle time. But one factor I’d like to note here is that I received much more experience through assists with the experimental settings, whereas I hardly received any experience through assists in the control settings. Due to the additional amount of “help” from my teammates, my average K/D in the experimental settings was also lowered, but not as much as I thought. 

Let's discuss some more additional general findings. Each gamemode became a bit more different in terms of scoring. For example, in ASL, it was usually a landslide in favor of the Red Team instead of a more balanced game, while in modes such as CTF, TJR, and RGB, we were met with an influx of scoring/boss kills due to the lack of defending team members. In CP, it stayed roughly the same, while in TDM, the scores were definitely a lot lower.

The largest difference was on mobile. I tested a variety of turrets and hulls on mobile (usually the hovering hulls), and the boosted health actually brought a boost in performance due to the increased health. Usually, I make pretty silly maneuvers and lose control, costing me some health, but with the double health, I was able to last a lot longer. For more findings, let's move on to analysis.

 

Analysis Questions

I’ll be using the analysis questions provided by Tanki to help determine if the update was beneficial.

 

Do you think that there were fewer occasions when you were destroyed where you spawned and were consequently completely useless to your team?

Short Answer: Yes.

Spoiler

Given that there were fewer people on the enemy team that were able to successfully reach my spawn point and that I was using Defender, I was spawn-killed less, especially on mobile. However, being useless to my team happened quite often, but that’s another story.

 

How did the benefit of well-timed overdrives and drones change? 

Short Answer: Not much for the drones I used, more so for overdrives.

Spoiler

I used Wasp for the majority of the experiment just because, and definitely, I’d have to say, the timing of a bomb had to work in your favor for you to either make a big play or die in a heap of shame. Several times, I placed a bomb down but ended up killing my opponent(s) before it detonated, and it was quite useless, while other times, I ran off with a few unexpected kills.

On to drones. For the entirety of the experiment (or the majority), I was using Defender, and I’d say it was active about 80% of the time when I had Double Armor equipped, so a well-timed drone wasn’t a factor that changed. Perhaps with other drones, like maybe Booster, but for the ones that I have (Defender, Driver, Trooper), no, it doesn’t seem to have a big change.

 

How did the importance of team gameplay change?

Short Answer: Team gameplay has definitely changed, but I’m not sure how the importance has changed.

Spoiler

Let’s break it down.

As seen in the first Red Alert CTF battle, where we were playing 4v6 against another team, it was definitely difficult when short-handed: losing just one player makes a larger difference so stop leaving when you die. Additionally, the smaller teams did make it harder to coordinate. 8v8 seems best in my opinion.

On the other hand, the team also “helped” a lot by spamming the spacebar towards the enemy or trying to capture the flag; everyone concentrated on the opponents or objectives blindly. This led to some interesting encounters. Once, I was trying to get the flag in my second Red Alert CTF battle, but this annoying Hunter decided he wanted to get yet another flag and tried to push me away. This sort of thing happens a lot in the regular MM battles, but with a shorter battle, there’s less time to net points and more pressure on the players, so kill stealing and flag fighting happened a lot more. (when in a competent team, of course; in an incompetent team, I’d be netting most all of the kills and driving around with an insanely high K/D, but that usually didn’t happen until after I processed my data)

However, I spent some time grouping, and it was definitely a lot better - it was a lot of one-two punches, continuous flag captures and scores, and in general, much smoother gameplay. We were usually on the winning team because of the better teamwork and coordination compared to being solo and fighting for the flag with other score-hungry players (I’m also pleased to report that I did brilliantly when facing much higher ranks totally not because I have an insane GS)

 

Could this experiment change your interest in the game?

Short Answer: Not really.

Spoiler

First off, I’d like to say that I am not very interested in playing the game; I used to get annoyed or upset if I wasn’t playing as competitively as I wanted. This was because I was only allowed to play a little bit during each day during the last great WAR, and needless to say, 10 minute battles and 8v8 gameplay did not work in my favor.

But I digress. Let’s move on to the real question. When I was playing in the control settings, I was quite relaxed - I was only bothered by the fact that these Thunders kept managing to hit me from halfway across the map and that an annoying invincible Mammoth just so happened to run into me while I was driving with the flag, but other than that, it was quite chill. However, I did remember exploding into mini-fits of rage when playing solo in the experimental settings - my overdrives proved mostly ineffective, my teammates kept hogging the flag, and every time I was about to get a kill, someone stole it and took it for themselves. In groups, it was mostly a positive experience given that Sabry and I were able to obliterate the other team with relative ease.

 

Do you have any suggestions on how to improve these changes?

Short Answer: There’s not much possible to do for the changes; however, it would be great if it were not to be implemented in PRO battles. 

Spoiler

My biggest problems with the experiment were that a team could be really short-handed if more than 1-2 players left the match and the pretty bland metas.

The first issue wouldn’t be very easy to fix. If you’re losing so badly and just want to leave, there’s no stopping you. After all, if you’re going to find a better battle and have an easier time there, then it’s not a bad idea as an individual. It just leaves your team struggling.

And as for metas, no matter the balance, there's always going to be a few standout metas that just completely blow the rest of the combos out of the water. It’ll need some more adjustments over time, such as maybe increasing fuel by a smaller amount for melee turrets and increasing the health of tanks by a smaller amount instead of a drastic doubling, but in the end, I feel that it is fine for me. Others may disagree, but they’re free to give their own opinion.

 

Conclusion

Overall, I’d say that the experiment was a more positive than negative experience in MM. Though I was not able to earn as many points (even when adjusted) and play as competitively as I wanted to at times, causing a bit of frustration, I was still able to have some strong moments in battle. Additionally, I feared having too short of a battle, but it turned out to be helpful at times - when stuck in a losing situation, I wouldn’t have to wait an eternity or quit. At first, it was harder to get accustomed to, but over time, I grew fond of the change. I’d just say to keep it in MM, make the changes a little less drastic, remove Archipelago (why is it still here), and adjust according to game balance.

 

Are you actually reading the caption?

____________________________________________________________________________________________________________

With that, it’s time to wrap up this lab report hopefully I got a 25/25 review. If you’d like to share your thoughts on my review, do so below; if you have some differing opinions, feel free to voice them here or on the Mad Tanks topic. Otherwise, that’s all for now; see you later!

[Person_Random 4,991]

Let's start off with a comprehensive review of the Mad Tanks event and experiment - Person_Random breaks down the pros and cons of the major changes right here. Enjoy!

P.S. Huge thanks to @Think for helping with the text of the image

[Table 0]

yes

[Tanker-Arthur 2,971]

Honestly, the double health was extremely fun for Siege Mode


Other modes, not so much. But in Siege, a Gauss couldn't wipe out a whole team in 1 shot, and I liked that.

[SONIC-BOOM 326]

12 hours ago, krish123_super said:

unfair random 

its mk3 or mk2 VS mk7 of urs 

thats what we are saying ... 

He has a point @Person_Random 1ST lieutenant with over 7k g/s.......I would say 90% of your battles you were way stronger than the enemy.

Conclusion......Their fault for staying in battle when you arrived on the scene.

I enjoyed the 16v16. I got a load of kills and usually came top, regardless which side won. Thats TO MM for you, never quite getting it right.