5v5 and TTK: An Analysis

Since Season 1, the term “stomps” have been littered in threads and posts throughout the forums. Many could write it off as the usual low-post-count people complaining they are not having fun, but by the sheer amount of them which always seems to be related to talks about Matchmaking seems abnormal. This is especially true for any of you who have read Blizzard’s latest dev post here which share similar sentiments to the few previous ones as well which I will quote:

Now, I would like to do some comparisons of TTK real quick before we get into further talks. This short table will just be a basic, vague address of TTK averages of listed games and we’ll detail some information after.

Game TTK
Call of Duty: Black Ops 300 milliseconds
Valorant 120-450 ms (based on weapon)
League of Legends 10-20 seconds (Full Team Fight)
Paladins 1-3 seconds
DOTA 2 ? (Higher than LoL)
Counter-Strike 0.4-1 second
Smite ? (Higher than Dota 2)

Now, there is a small issue when trying to accurately gather data and show it in a condensed and small table as there are MASSIVE variables which decide TTK in games. As well, the genre massively changes the numbers obviously.

Black Ops, Valorant, and Counter-Strike

These are variously similar times but listed slightly differently when looking at available data. Now, mind you, there are a few big important considerations when averaging TTK for the game as a whole. The main factor is a weapon’s damage. If you notice the spectrum of Valorant and Counter-strike, the data looked at was analyzed by a chart of all given weapons and their various abilities to kill. Pistols tending to take the longest, with many weapons having instant kill potential in both. However, they all on average stay below one second.
The other variable, and much harder to analyze, is map designs and layout. I imagine this effects weapon choice and the average distance players must fight one another so it is very difficult to get an actually accurate time with this. However, the fact maps are broadly mentioned often and that “Distance” is usually included in these tables meant it was worth mentioning.

Paladins

Paladins was a difficult one to look at. For one, every character has a different health pool, damage resistance, and more which changes the numbers. (Average HP range of HP can be anywhere between 2k and 4k)
Not only does the target change these, but paladins has an unusual means of combat compared to the above topic that is beyond just guns and their damage. Damage mitigation comes from more aspects than, say, Counter-Strike as not just from “Body Armor”. There are many abilities and builds meant to completely avoid or negate damage numbers. Paladins deserves it’s own section in this discussion as it is the most comparable type of gameplay to Overwatch with a TKK spectrum fairly comparable.

League, DOTA, High TTK Explained

With what was mentioned earlier in Paladins, the same can be said here and more. In a game where concepts like Aim and Headshots do not apply combined with various damages, health pools, and other concepts meant to negate damage numbers drastically affects TTK in various ways to the point where the only method you could try recording the numbers in a hyper specifically observed situation. You would need the characters involved, their levels, items, abilities, and number of people involved in a fight. Getting the average time for all ten people fighting in a 5v5 was the only easy average to gain.

This is because both sides will equal out relatively to a similar pool of health split between five people in an engagement that starts and only ends when one side is essentially eliminated. Getting a rough estimate of a group “All-In” is essentially the only means of giving a short average time without it deserving it’s own graphs and break-downs entirely. The same sentiment can also be given for Smite or most other MOBAs or generally just games built around High TTK.

What was the point of these comparisons?

Simple: Genres have a purpose and somewhat good idea of why their gameplay is represented by TTK. Things like Call of Duty have millisecond TTK due to it being certained around lightning-fast engagements. A fight can be almost instantaneous and this similar sentiment can be given to Counter-Strike and Valorant. Though, with those games, their TKK changes considerably as matches progress and higher damage-scaled weapons are available.

MOBAs tend to have WAY higher levels of TTK due to the genre difference. The sheer difference of something like 300 milliseconds looks dramatically different to observe if you have ever tried to compare a CoD montage to someone’s MOBA comedy video of, say, two Tank characters dueling for roughly 50 seconds.

How this relates to Overwatch and Overwatch 2

For numerical data, this is a chart from 2016 by a reddit user named Don’t_Rush:

(h ttps://docs.google.com/spreadsheets/d/1wY_yJgMaFZqgRwo7GM5vvpL7rHIPWTkDk162AazxVPM/edit#gid=0)

Understanding different genre games and their types of gameplay, we can understand what goes on in Overwatch which arguably has a unique style of combat unlike other shooters where combat can’t be measured easily in milliseconds alone. This chart is also just measured by different weapons and draining someone’s health pool without considering things like mitigation, sustain, and so on. Even without those, it still ranged from around 1 second and even 10 seconds.

If we are going to discussing something about Overwatch 2 such as the state of the game and these proposed “Stompings”, taking a look into TTK of matches is important. Not just roughly by weapon damages alone, but everything discussed above which can effect the time of engagement:
Map design, health variances, mitigations, healing, team formation design, META, possibly more variables I forgot to mention.

These are all concepts that must be deeply looked at if we are to view Overwatch 2’s TTK being objectively lower per match compared to what it used to be and what changes have been made to reach this number. As well, what changes are good and what are bad. If a game does not have every character, team, and formation to be of equal value, it is quite possible there is some variables which create an unfair potential for a “Time of Engagement” for one team and another.

tl:dr - What’s the point of this thread?

  1. TTK and how they drastically differ from genres

  2. TTK and how it has become different from OW 1 and 2 (Possibly?)

  3. What makes Overwatch different from a standard FPS like CoD

  4. Presentable topics and data when referring to other topics as of late such as what Overwatch has changed or kept, “One-Shots” in general, map design in OW effecting state-of-play, and whatever else you can conceive from looking at this

  5. These variables make simple concepts like “Nerf/Buff Genji” way more of a complex topic than damage number changes

6 Likes

I think a lot of the stomp problem is the move to 5v5, and increase amount of impact per player.

Now, one particularly good or bad player (or a player having a particularly good or bad day) has a larger effect on the outcome.

One of the selling points was “an increase in impact per player” and the stomp problem is just that being felt.

19 Likes

This one sentence is kind of ONE of the many topics this thread can kind of cover.
In another FPS game, a pistol to a machine gun can change from 500 milliseconds to150 milliseconds. Fair. That range will never change.

A game where you can completely tell damage “No”, has healing, certain aspects means data on if Genji’s shurikens can get an elim in 2 seconds or 11 seconds, by itself, meaningless by itself.

A “Stomp” happens because a team has a ridiculous difference in Time of Engagement and the factors of what changes that can be a massively detailed conversation. If we want to talk about how Widow can kill someone instantly and Genji needs two seconds, that’s fair as a SINGLE variable of dozens.

Especially since before when conversations about things like one shots were discussed, the answer was that shields meant they didn’t matter because they could be completely negated and often.

1 Like

I don’t get this the whole selling point of 5v5 was hay dps mains now you will have some solo carry potential.

Mind it even now a single gm sniper can’t put do 5 bronze even with buffed stats.

But we are able to make plays, but immediately it’s oh my god one player ruins the whole game.

Nerf dive, nerf range, ne f dove ne f range

Buff heall buff heal

I’m sick of this.

Tanks have been allowed to be raid boss, supports have been buffed to hell, i play support to so don’t categorise me as dps main.

Dps can’t even 1v1 tanks which they did in ow1.

Fine

Now they shouldn’t even kill squishies?

At what point will this stop.

1 Like

One of the way one shots (or any burst) do add to stompyness, by reducing the number of heroes which can have impact in a fight.

If one hero has burst damage, and the other doesn’t, then burst will be resolved “first”, and non burst resolved second.

Healing for instance only can take effect if the hero isn’t burst down first.

And by removing impact from some players via burst, you are effectively reducing down the “spread” of impact. You are running a game with “less players”. Which increases stompyness.

You point of time till engagement ALSO reduces the number of people in a fight.

4 Likes

Matchmaker is still bad in a lot of cases and there is skill difference in teams.

So do you bring mei up to tracer level or bring down tracer to mei level ?

I’m not sure that’s really it…

The problem with 5v5 isn’t 5v5 per se, yet still enforcing Role Queue with now only one tank which is a key role that can make other roles powerless, especially support players.

In 6v6, way before Role Queue, each player got impact, no matter the hero they played so I’m not sure 5v5 offers more impact per player, it’s just more random and brawly and less constructed in terms of strategies (that were more complexe and made the most of each hero’s kit in 6v6 Open Queue - and multiple synergies available).

That’s clearly the case because of the solo tank issue as tank reworks weren’t enough to make them all tank material. This is again an issue created because of forced Role Queue (with now 1-2-2).

Plus, I believe that now that the matchmaking issues were exposed, it’s safe to say that they’ve probably been there since Day One of the F2P update last October (probably because of new players at first and now because of a lack of players to find a balanced match).

That was the official explanation from the OW team but I don’t believe it’s true. All players that are familiar with the franchise since 2016 probably have the same impact as they did back then when it was 6v6.

Most players on these forums were gloating about how 5v5 will help them climb easier in ranks once the game goes 5v5… Never heard of those players since. They probably got the same rank as on OW1, or one or two ranks below since the matchmaking is random and the tank issue still not addressed.

2 Likes

I also believe this, but I also believe another variable that adds to it makes it feel worse.
For example, say you play a game against someone who has slightly higher skill than you. (Whatever that means.)
That doesn’t mean you don’t have a fighting chance. You can still beat them. However, how do those chances change in a game where not everyone is equal in mechanical stats, abilities, and damage?

1 Like

i would add the team fortress 2 ttks to your list aswell, to give a broader view of hero shooters

Or you accept the game will be more stompy.

Here is the thing, decisive fights are not always bad in a game. The trick is how do you make them feel good?

Yep, he more impact is unbalanced between players, the more stompy the matches will be. Making tanks a raid boss makes the games more stompy.

It’s just maths. in 6v6, you accounted for 1/6 of the impact from your team on average.

in 5v5, you account for 1/5, which is a larger amount.

3 Likes

If the difference is between dps or support it’s closeable.

Because you have more options and two player each there.

If it’s tank diff it’s gg most of the time.

There doesn’t even need to be skill diff.

If you get the dumber tank that’s enough.

All tank mains are dumb, it’s only question of who gets the biggest idiot.

You mean this depressed bunch probably spray something from helicopters.

I wanted to, but I searched everywhere and there didn’t seem to be an easy table to see this. You can find character weapon stats, damages versus health pools, and more. I think it’s also harder because TF 2 has a concept that other faster FPS don’t have and that’s number of players in a game. Which dramatically can effect things.

It’s… kind of hard to talk about a game that can have 30 players in one place and an engagement that doesn’t take less than a second to win.

1 Like

And here I was thinking, if you get mad at your games feeling “outside your control” the brightest people would take up the heroes which stop them feeling that way.

Here is the proof the collective intelligence of the tank mains in that thread won’t go past single digits

These are the people having the highest impact in your games.

Just read the copium replies there.

The other part which is interesting is “what is a stomp, and what feels like a stomp” are not the same thing.

It could be there is a solution which the games are JUST as stompy, but they don’t feel anything like as much.

Hell, I’ve seen people complain how stompy the match was, and blaming the matchmaker when it was a comp game where they stomped in one direction on the first map, and got stomped on the other two.

It isn’t a matchmaker problem when that happens, it is a gameplay issue.

I think longer fights ultimately feel less stompy, and shorter team fights feel more stompy.

I also think a lot of how overwatch is set up for fights ALSO causes issues. It is a drinking game, a team which wins the first match on KOTH maps, now had ult advantage, AND position advantage.

Meaning they are more likely to win the next fight as well.

4 Likes

This is true with support and dps as well though. People don’t always notice it, but there is a huge difference between a support duo that outputs 20k healing and has imo and rez and a support duo that outputs <8k healing without imo and rez. It doesn’t really matter if the tank/dps are significantly better at that point.

Likewise, if one dps duo runs an efffective double sniper or has an effective echo or pharah and the other is running Reaper/Torb and getting sniped across map or having damage rain down on them from above, it’s going to be very difficult for the rest of the team to compensate.

I do think that outside of Masters/GM dps diffs tend to be the easiest for the team to absorb (but the higher the rank the more the dps diff tends to become the determining factor in who wins the game), but support diffs are felt every bit as much as tank diffs, and I think we’d all be better off if the community as a whole realized that. Supports can have way more impact than they recognize and it’s really hard to win a match when the supports think they have little/no impact on the outcome.

2 Likes

Currently you cannot make up for tank diff in ow1 doom mei brig could offset one bad tank a bit.

That’s no longer the case.

A mei or brig can help a good tank but not make up for bad tanks.

1 Like

As a Mei main I am well aware of this :slight_smile:

2 Likes