Comparing Accuracies between Input Methods

Hey everyone,

I’ve always been interested in this discussion and Halo Infinite is the perfect game to discuss this topic over because we finally have a competitive game with seperate queues so we can stop throwing anectodal evidence at each other and look at raw data.

I wrote a small program that grabs the Top N players from the halotracker leaderboard and then calculates the average, minimum and maximum accuracy from all players in the dataset. Nothing fancy here, just adding up all the accuracies and dividing by N to get the average.

Small disclaimer: Since the accuracy stat is tracked over the entire profile and not per queue it was a bit tricky to decide wether a player had achieved that accuracy with a controller or a mouse (or both), so I excluded players who also had a ranked rating in the other respective queue. This means I’m only looking at players that have exclusively queued in either Controller or MKB queues.When I’m looking at the Top1000 controller players and one of them is also rated in the MKB queue, that player is ignored. This was the only way to come close to a valid result since the api doesn’t give you anything that helps you in that regard.
IN THEORY someone could play 200 ranked games in the Controller queue but also play another 100 games in Quickplay only using MKB. I would expect though that these types of players are exceptionally rare because most players (especially the ones pushing for high ratings) have settled on one input method. The results pretty much verify this:

Top N players KBM Avg. KBM Min. KBM Max. Controller Avg. Controller Min. Controller Max.
25 47.83 42.99 53.44 54.78 50.51 59.13
50 47.75 39.57 56.12 56.20 50.51 61.75
100 47.26 39.57 56.12 55.10 45.46 61.75
200 46.67 35.78 56.12 54.50 45.44 66.59
1000 44.58 32.11 56.12 53.26 38.93 66.66

What you can see here is what pretty much everyone had already expected. Overall controller players have much higher accuracy regardless of how you look at it. 8-10% doesn’t sound like much but this is the difference between hitting the headshot on your 4th burst and killing the opponent or having to wait for the 5th burst because you wiffed one or two shots from the bursts before. The differences here are massive ingame. There isn’t even a single stat here that makes MKB look at least a little competitive.

There’s really only a couple of options here:

  1. Mice are a handicap and Controllers are actually the way to go when it comes to precise aiming and tracking in FPS games.

  2. The MKB players in this game are just bad. The better players whose tracking could match Controllers are simply not playing Halo Infinite.

  3. Aim assist values are slightly overtuned (spoiler - it’s this one right here).

What I’m taking away from this:
Never even once following the Controller-MKB discussion have I cared about which is better/ easier/whatever. What I care about is this: If you throw all players into one pool (like quickplay or the open queue) it’s not fair for MKB players full stop.

Quickplay is the first point of contact for new players and straight up getting recked for playing on “the wrong input method” won’t feel good for most players.I also don’t like the point beeing made alot on this sub of “Halo has always been a console game, therefore Controllers should always feel better.” Here’s my rebuttal:
FPS can be played on both MKB and Controllers. There’s no reason to artificially set the aim assist to an amount where MKB players get the lesser end of the stick. The devs literally control how hard Controller players will hit and within the snap of a finger Halo could become a “MKB game”.

I think we can agree that without aim assist the data would look the exact opposite so what 343 should go for are aim assist values that actually give both inputs a fair chance against one another. The current ones are clearly not ideal. Aim assist exists because Controllers need help when aiming - but not this much.

343 has to make sure that in MKB queue Controller input is completely ignored because you could just queue for MKB and then actually use your Controller to gain an advantage. Also if you are on MKB I would stay as far away from the open queue as possible unless you absolutely have to play in a party.

Thanks for reading

16 Likes

Out of curiosity where are you getting these stats from? I can’t even look up my own stats for this game. Let alone manage to scrape data like this for “Top N”.

EDIT:
nevermind… I think halo tracker was down for a bit. Well it was for me anyways. There wasn’t any infinite stats loading for me. Now they are.
I guess that there’s pretty solid evidence there. The only question I could ask further would be what are the median accuracies across the inputs… as most of the people complaining I imagine are not sweaty leaderboard types.

Personally I found that the beta was bad… but I haven’t been having issues on full release.

For me however, in order to make the inputs actually fair 343 needs to investigate this at the gun level, and just show the community the facts… they probably need to tone down the BR and AR a bit and buff the sniper and commando. Or they need to make sniper harder to use for PC players. To me when I hear mnk players complaining I keep having flash backs to MCC snipers and getting no scoped consistently across the map and I am like yeah… mnk seems like a real disadvantage.

But again the same statement will be made… mnk players are asking for 343 to make controller players compromise on the gameplay they have enjoyed for 20 years.
Splitting input types (or making it easier to choose like in MCC) is really the only fair solution.

5 Likes

Excellent analysis. However, there are other variables to compare. For example, a gamepad might be more “accurate”, but the mouse can turn on a dime; mkb can run circles around gamepad at cq.

My point being is that there is much more to compare than just accuracy. Even if you had a similar study analyzing other variables, how do you compare the significance of an accuracy advantage to a turning one. And one wonders at what point do all advantages and disadvantages wash out?

Finally, in theory, shouldn’t SBMM level everyone out regardless of inputs? Whose to say whether or not a gamepad player is matched with a higher or lower skilled mkb, but does that matter so long as matches are photo finish?

6 Likes

You mean the best Halo players who are mostly on Controller are the best? what a surprise.

If you could compare a median group of average players, you’d get a more accurate view.

4 Likes

Imagine 343 removed aim assist and you started getting trashed by MnK players. Then someone says to you “but SBMM will level the matchmaking out regardless of input method, so everything is gravy”. See how dumb that sounds?

The point is that MnK players don’t want to feel like their input is inferior. If a MnK player loses to controller, they want to know it was because the other player is better, not because the other player is using a superior input.

2 Likes

Nah, it still kinda makes sense to me. If SBMM matches mid-tier controllers with bottom-tier mnk, what does it matter? So long as their apparent ‘skill’ is comparable and the match is a close finish I likely wouldn’t notice the difference anyway.

My point being that there are too many micro-variables to control for, so a tiered macro-level matchmaking system (such as SBMM) should in theory wash out the differences if it’s doing it’s job. No?

1 Like

If you only play socially, then that logic is fine.

However, if someone is playing ranked, then that player presumably wants their rank to be reflective of their skill, which is reflective of their performance. Currently, there is a disproportionate degree of influence that input has on performance, and therefore rank.

Players who are competitive by nature want to beat the other team because they are better. They don’t want to win or lose due to any other reason, including a difference of input.

Nobody is saying the two have to be perfectly balanced, because they can’t be, but the difference as of right now is too obvious to ignore.

3 Likes

I don’t buy it. I’ve played Onyx Controller players on small maps then matched them again on Behemoth. Evenly matched tough fights to they might as well have been Bronze players. The illusion is just gone. Controller players are all being hard carried by software.

1 Like

You so badly want to believe that you are owning mkb players on your own skills alone, don’t you lol. Unfortunate for you and other seething controller players, the stats are there in the OP. Controllers are OP in this game, which shouldn’t come as a suprise for anyone who isn’t a gasoline huffing clown. Cool that you took the time to wonder if there’s other factors that should be taken into consideration, when weighing the pros and cons of different inputs. However, aiming is the only thing that requires any mechanical skill in Halo. There’s no bunnyhopping or strafejumping or crouchsliding or anything that would require the level of precision a mkb can offer. And when the only thing that requires any mechanical skill is done for you by an aimbot, it becomes very clear why controllers are so OP in Halo. In the current build there will never EVER be a pro team where every player plays with mkb.

2 Likes

As cool as this data is - I dont think it is by any means a complete picture. I also think there might be a problem with the data

  1. accuracy isn’t a great representation of skill - especially when you consider techniques like jiggle peaking where it’s not some much about hitting all your shots as it is about providing as much pressure/damage while minimizing the amount you receive. For jiggle peaking for example, your accuracy is going to take a dump.

  2. you need to include the standard deviations here. Min and max tells us nothing.

  3. why does the average accuracy fall off a cliff for kbm after the top 100 players. I refuse to believe that in a game with (probably) over a million players, the top 100-900 have such drastically lower accuracy than the top 100. And I refuse to believe my accuracy is regularly better than that on Kbm. Edit: I just realized this is probably because of the collection method where you exclude people that did mixed input queues so maybe you are just grabbing from VERY small player pools.

1 Like

You are aware that the aim assist in this game is EXTREMELY low compared to all the other halo games right? Even compared to other fps games the aim assist is really low. Which means it’s not an “aimbot” by any means. So no the game is not doing all the work for them it’s not even close to that.

2 Likes

My only response to that is that my cousin, who is SUPER casual (and self-admittedly bad) finds the AA to be very strong and sticky in comparison to other games like CoD. Although personally I do feel like the AA on pc with controller felt different than on console and he tried on console.

1 Like

There was apparently a bug on pc that drastically increased aim assist when both M&K and Controller were connected along with the dead zones being set to 0. I’m not sure if it’s fixed yet or not.

1 Like

Would have been interesting to see these figures before they adjusted the magnetism cone pre-pre-launch.

Thanks for your efforts…

Idk what to tell you dude. Look at the numbers in the OP…and there you go. That’s reality. Also, if there’s software that helps you aim – that’s an aimbot there’s no two ways about it. The game literally does most of the work when it comes to aiming. Controller players don’t know this, but I’ll let you in on a secret…aiming on mkb is hard. Especially tracking. You’ll never understand how easy aiming is on controller with aim assist until you get good at tracking with mkb. You have 0 perspective. I have lots of perspective. I get 99 percentile scores in Kovaaks tracking scenarios. And all this is fine. This game was made for controllers and it will stay that way. Lucky you haha!

2 Likes

This right here. Exactly my thoughts.

2 Likes

Because of the difference in how you play the 2 games where in 1 you essentially become a stationary target in a fight the moment you scope in and in the other it’s a literal duel and strafe fight where 2 players are actively moving around dodging shots while trying to land there’s. There has to be some aim assist on a gamepad. The issue with khm vs gamepad is just that. It isn’t fair. There’s circumstances where 1 has the advantage over the other but players probably find themselves in 1 circumstance way more often.

The solution is to allow input filters. It’s the only way. Not sure why it isn’t there in a game that’s free that should have a huge player base. If anything…leaving it as it is can hurt the player base and cause what would be thousands of mkb players to just say screw this game…I don’t want to use a game pad…onto the next.

I play on an Xbox series S with a game pad. I’ve not used a mkb in so long like the last game was guns the duel…lol but reading all this feedback really demonstrates what is going on here. Here’s how I myself first felt in the game

  • bro its SO unfair that we play against mkb. The advantage is SO significant for them dude.

Until I went into theater to watch a match I had played. I played good in the match, lost, but wanted to see what went wrong. I immediately noticed a mkb player was on my team by how his reticle moved and his turn speed and I found it interesting to watch. That’s when I realized that what i felt was a huge advantage really wasn’t. He didn’t do the best on our team, and I understand there could be extremely talented mkb players…but the point I found was that…there’s this like automatic belief some game pad users have where it’s like wow mkb really? Unfair. It isn’t true though…but the level of aim assist is Lower than in the other titles. I dont believe that’s what should be messed with because it’s just impossible to balance it all out where 1 person is aiming with 1 digit and 1 with their entire hand. 1 will always suffer more than the other whatever you attempt to tune in terms of having them play one another. So solution is the simplest. Input filtering across the entire board imo.

But yea halo and cod are horrible to compare to one another in terms of how it feels to aim and shoot and just the general sensation of combat. They’re 2 extremely different games played differently.

(Editing this in) input filter optional. I have a good friend on mkb I play with. He’s significantly better using a mouse, (not at all saying it’s a magic fix for everyone, but it became apparent he had already used a mouse for other games and only struggled in halo 5 with us when we played and he was on his xbox. )and we have fun playing together. An opt in should still exist to not segregate players. That have mixed input fire teams …so this whole thing is difficult to tackle. What would it do to MM searches? An actual player population counter per playlist and filter would be ideal imo

I mean personally I use MKB and feel like I do decent at least for my skill level. I mainly play casual modes like btb and can often end up in the higher end of the leaderboard. This is good info though because the back and forth of which side has an advantage is crazy. I personally feel like I move more fluidly on controller but shoot better on mouse.

I wanna throw a quick counter-argument in here. Note this is only based on my own speculation.

Most controller players complain that there isn’t enough aim assist, but this could be because it was really aggressive in MCC, like REALLY aggressive, and they are used to that.

One thing I think it comes down to is Halo has (for the most part) been a controller/console only shooter for the majority of its history. Most people interested in Halo are gonna be playing on controller because that’s what they’ve always done. They’ve had 2 decades to “git gud” at Halo while M&K is a relatively new thing to the game excluding Halo PC and Halo 2 Vista.

3 Likes

Although I appreciate this “analysis” - it doesn’t prove much of anything.

“Accuracy” could be skewed a million different ways (controller players tend to take less risky shots [i.e., more up close fighting] because it’s harder to aim, etc., while MKB players take more risk [360 noscope kids, etc.] dragging down their average accuracy) - yet MKB players perform better overall (on average) than controller players. But, who knows? 343 should release stats on this so we can see what Crossplay has done to Halo.

Here’s what I’m interested in knowing: what are the input/console/PC breakdown in each ranked bracket? What does that look like in the Crossplay playlist? The Input Method playlist? If we had two teams of equal skill - one team on console, the other on PC - who would win? Why?