[Vehicle] Stats Interpretations Question: The PPA & The Prowler.

HE and HEAT have been eschewed for AP in MBTs in as far as most factions are concerned. I would have to run numbers against the Oracle but it feels like the VS held out on MBT HE the longest. The most common HE you'll find are probably on Lightings. Check the second part of my reply for other commentary on this point.

And it's not fair, in my opinion, to compare a weapon that can only damage Infantry with a weapon that can damage both Infantry and Armor (I suppose the distinction is moot - everything can damage Infantry). Presuming there is some correlation between what a weapon can damage and what the weapon will be used against, that means that, whether or not you can compare the dynamics of a Prowler in Vehicle kills versus occupant kills, you are, in the end, going to end up killing a half-full Sunderer every now and then and a two-crew tank once in a while. It's not like farming pure Infantry where you kill one with one shot; multiple shots are needed to get the effective kill between Vehicles, the last shot may not always come from the tank that performed all the previous shots (darn kill stealers), and all occupants of the target Vehicle will die at once if they are in the Vehicle at that time. You could argue there is some statistic that coorelates number of seats versus number of shots required thus that the ratio approaches 1:1, but it's still an unfair assessment.

While I do not have all the time right now to play around with the Oracle, if it's KPU, then here's all of the MBT AP turrets. Here's the same turrets, but with their Vehicle KPU. They look very close. What you are describing is not a problem with the treatment of the P2-120 compared the PPA, or the treatment of the P2-120 AP compared the P2-120 HEAT, but a failure of the dervelopers to make HEAT meaningful in the long-run term in the eyes of players. Though you lose splash in transition from HEAT to AP, and after the last adjustment of the tank cannons you needed to land near-perfect shots, AP gives you one extra shell on Armor over HEAT in most cases. HE is right out: it's two shells more than AP in most cases and the splash is still not a comparative advantage against HEAT. That leaves most players using AP against everything, because nothing else is truly stand-out-ish in the ways that it matters.

How are you quantifying the ability of the Maggie to traverse difficult terrain and get to places that other mbts can't? The extra ability of the prowler anchor can be taken from damage stats. While the vanguard could be seen from damage taken with their shield. But where do you see the effects of afterburner and being able to ignore certain terrain. Essentially making it a floating harasser with bigger guns.

Seeing as anchor mode is the tr special ability I can totally understand why it is a top performer.

Also tactically how is the tank being used? Prowlers are more artillery. Vanguards are more line busters and the Maggie is a anti infantry farming machine as well as very survivable with the right load out.

You will also need to take be level into account vs and nc have lots of new people. So the performance metrics will take a hit.

Probably want to use only high level be stats as those people probably know how to use their mbt to a better degree.

To reach a conclusion? No, not really - this is all hypothesis based on the available data. I would go far as to say that it gives us an extremely good indication of what the conclusion will look like, which allows us to pass along a "Hey, head up" to Daybreak's folks that address balance, but it doesn't paint the entire picture.

Not especially, it's gathering more kills / KPU then anything the other MBT primaries can do - this alone is enough to tell us that it may be better at AP to a significant degree. If it really is getting that many more kills via crew deaths, something may be amiss. If it's not getting that many more kills via crew deaths, but instead farming them directly ... same thing - something may be amiss.

Yep, that it is - I never claimed anything more then my opinion. I am, however, basing my opinion on logic and performance stats, how we interpret them is open to argument. I changed a few of my conceptions / opinions based on doing this analysis - I thought that Pounders were being used for AV more then they are, and I was under the impression that Comets were doing better against vehicles then they are, as a few examples.

I can imagine such a weapon, that's the Lancer / Vortex, but as we're comparing Nanite-using weapon platforms, let's go with the Vortex (The reason being that the unit it's equipped on cannot be pulled infinitely for free.) - The Vortex doesn't meet your criteria of insta kill against vehicles, but it is quite useless against infantry in general, so KPU will be deflated, as many vehicle crews will escape. Still, the massive range allows insta kills in unexpected scenarios, so i'd say it's probably close.

For pure kills, the Left Vortex has the lowest of any MAX AV in the game, although that's less of a surprise given it's 124 unique users (2nd lowest, behind Right Vortex) - The Left Vortex has 2.65 KPU, and a V-KPU of 1.58. The KPU being higher then the V-KPU, I would say that it is indeed getting quite a few multi-crew kills with the vehicles it downs, so your point is fair. That said, if the AP Prowler truly is getting a significant portion of it's KPU from vehicle crew kills, i'd say it's AV potential needs to be addressed as well, as it performs well, well over any of the competition.

Perhaps, but if the Prowler was (and still is) getting that many infantry kills compared to the prenerf PPA with crew kills as a major component...? That throws more alarm bells for me then the original data.

As we cannot see any data for attacks / kills from the rear, this is difficult. Given the very clearly inferior primary cannon performance, we can only assume that whatever advantage the Magrider has is insufficient to shore up it's lack of overall performance. The Magrider is the slowest ground vehicle in the game, so any positional advantage must be obtained and exploited rapidly in order to compensate for changing battle conditions. The Magrider has the highest skill cap for this reason alone. Due to it's inability to go head to head vs other tanks, the Magrider must flank / gain a positional advantage to be fully effective - other MBTs require less positioning to be fully effective, and the sheer output of the Prowler's primary cannon is a massive force multiplier when flanking. How much damage can you output in those critical moments of ambush, while the enemy is reacting to your tank's position? Would you rather have a Lockdown AP prowler, or a lackluster Magrider FPC?

Me too - what is already the shortest reload primary becomes even shorter, and gains a massive velocity boost. The only thing I can think of with higher DPS is C4 or tank mines.

With the guns to back it up? I see the Magrider as a contender, but it has neither the Primary weapon performance, nor secondary weapon performance (due to constant nerfs) to back it up. The Magrider has become rather underpowered, as the platform itself once served as a force multiplier. Getting into odd places is hardly useful if you don't have the output to take full advantage of ambushes, especially when that capability comes with the inability to compete head to head with armor.

The Magrider no longer has the advantages that justify it's lesser speed / greater inertia / turning it into an overall weak armor piece compared to the Prowler / Vanguard. Skill can make up for a sizable amount of disadvantage, and the fact that the VS are using overally higher Battlerank players, and outputting less performance stats is quite telling for this.

The Magrider? I personally believe that it's best use is flanking and shot-trading from range. The Gatekeeper has extremely superior performance compared to the Saron, so the Secondary weapon advantage the Magrider was intended to have no longer exists. The Prowler was intended to have better primaries, and lesser secondaries. The Magrider was intended to have better secondaries, and lesser primaries. Now that this is no longer the case, there only real, effective engagement tactic is ambush from the rear, and long-range area denial via pinging away at things. This is a tactic that ceases to be effective once enough air has been pulled to remove Magriders from whatever perch they have, turning the Magrider into either a delay tactic until superior forces can be summoned, or a solo ambush / flanking hunter.

It used to be. The complete obliteration of the PPA's effectiveness took care of that - the Magrider is now an unloaded gun when it comes to AI capability. The platform may still be there, but the ammo is gone.

Very true - the NC have lesser battleranks on average, and are still competing on par or better in many cases. I did a complete overview of every class of weapons the other day for someone - as it turns out, the NC are the only faction with at least 1 weapon in the top 3 of every single category. There's no faction with a better array of choices than the NC, for any given situation. Their gear is overall, the most balanced. (I blame this on Higby's favoritism, but that's another topic entirely.)

I'm still not clear why you opted for KPU instead of KPH. Traditionally KPU has been suspect because it is affected by the amount of time each unique player uses a weapon.

Say we have two weapons, A and B, with the following characteristics. We assume that the number of uniques is the same (for math, here set at 1).
Weapon A is used for 5 hours a day and gets 8 kills per day.
Weapon B is used for 2 hours a day and gets 4 kills per day.

By KPU, Weapon A is the superior one, with a KPU of 8 versus Weapon B's KPU of 4.
But by KPH, Weapon B is better, with a KPH of 2 instead of Weapon A's 1.6 KPH.

The question is, which is a better metric for the "best" weapon? I believe that time used is a necessary consideration. If I have to spend much more time getting each kill, then that is not an effective weapon. This is why most people will examine KPH instead of KPU.

Examined under the lense of KPH instead of KPU, data from around September 14, 2014 looks like this:



Which is significantly different from the story that KPU told:



So in terms of KPH, the difference was extremely significant between the performance of PPA and Prowler main cannon. You did not have to spend as much time or effort to get a kill with PPA compared to getting a kill with Prowler cannon. This is why PPA was seen as OP.

This trend continues even to today, despite the PPA nerfs:



And again, compared to your graph of KPU:



Why is this the case? Well it has to do with how long each weapon is used for. Taking a look at September 12, 2015:

Hours Per Unique
PPA- 0.059 (7.89/134), 3.54 minutes
P2 AP- 0.409 (510.73/1,250), 24.54 minutes
P2 HEAT- 0.300 (400.68/1,337), 18 minutes

And there's your problem with using KPU. On average, each unique user only uses the PPA-H for 14.4% of the time that Prowler AP is used, and 19.7% of the time that Prowler HEAT is used!

In other words, what is actually going on around September 12, 2015:
Your average PPA-H user plays for 3.54 minutes to get 4.21 kills, and needs to spend 50.5 seconds to get one kill.
Your average Prowler AP user plays for 24.54 minutes to get 10.35 kills, and needs to spend 142.3 seconds to get one kill.
Your average Prowler HEAT user plays for 18 minutes to get 6.59 kills, and needs to spend 163.9 seconds to get one kill.

I'm also not sure why we're comparing a Harasser weapon to MBT weapons. Or why we would compare MBT secondaries to MBT primaries.

Shouldn't we just be comparing one Harasser weapon (like the PPA-H) to another Harasser weapon (like the Marauder)?

I'm legitimately curious why you're comparing two weapons that are on two different platforms - not to mention that you're also comparing an AV and an AI weapon.

You need to be comparing the PPA-H - Marauder-H - Canister-H for this analysis to work.

Reading through this it seems to me that

The KPH stat seems to say that in a smaller fight squad vs squad the PPA will do far better than an AP prowler which makes sense because in a small fight if you can find the spawn you can farm that single point very easily which tends to suppress the fight because the spawn point is unescapable. Alternately an AP prowler will do far better by KPU than a PPA in a large scale engagement 40 vs 40 ish or higher this too makes sense because an AP prowler can engage more targets.

But if I'm understanding the two stats correctly there are more potential KPU in a zerg and more KPH in a smaller engagement with less uniques.

All told though I think the Mag needs a buff not necessarily the weapons but the platform itself there really isn't that much terrain that can be exploited that quickly that would protect more than one or two mags in a particular assault.

They shouldn't have to...even as a point defense weapon it's downright terrible. I tried...I tried very hard with the nerfed PPA. I did everything I could. Me and my gunner experimented with tactics and certain types of terrain...nothing. The fact it takes near a full clip to kill infantry beyond the range of your bumper(with clip size increase) renders it near worthless. You can zoom in...get a kill...thenr eload and have four guys with C4 and rocketlaunchers shoot you and no way to respond.

Honestly...I don't know what goes through the Dev's heads. They never seem to put any actual thought about balancing thing that are OP they just....seem to TRY and ruin it. It's not like they even try to fix it.

Fractures need some bite back. Velocity...really all they need. But if they left them their current speed they need a little more damage.

ZoE simply made MAX's to fast. Dial back the speed increase so they were faster but not as fast as normal infantry. Likely would have solved the issue.

PPA could have been 3 round burst fire swith a smaller CoF so sniping at long ranges wasn't as effective (which was the true problem with the PPA ) Well the medium/close range was still very threatening to infantry.

But instead they just like to use a sludge hammer to tap things into place. Then seem surprised that they break it all.

why are we comparing a main gun with a secondary gun? perhaps compairing main guns would lead to more accurate stats....

I'll field these questions now, as i have time to do so - I'll check back later for the others.

The original goal of the thread was not to directly compare two weapons.

The short answer to this, is that I picked two weapons with similar AI performance, and a similar user base, on platforms that both cost nanites. Both are used extensively for AI - Engagement range is similar, and both have long effective range. The PPA-H and the Prowler are both at the absolute top of their respective empire's AI performance, and were used in similar numbers - I am comparing the two as an exercise in determining acceptable KPU, as performance is similar.

I was going to use the Magrider PPA, but as the question is "Why is this KPU on unit A acceptable, and not unit B, when the range of both and overall performance of both is roughly equal." - It's not a direct comparison of tank primaries or secondaries, I'm comparing two long range weapons used for Anti Infantry in similar combat scenarios.

I would have expected the Prowler AP to have gotten less infantry kills and more vehicle kills (crew kills are another story, but I expect that the majority bail out, in general, or at least pale in comparison to farmer-obtained infantry kills), and the PPA to have gotten many more infantry kills. As it turns out, the PPA-H was only just beginning to outperform the P2-120 AP in terms of infantry kills, so I'm curious as to why the Prowler's nearly identical infantry KPU is acceptable, when the PPA's performance was not.

Given that the original asymmetrical design intent of the Prowler was to have superior primary cannon performance, and inferior secondary weapon performance... and the Magrider was designed to have inferior primary cannon performance, and superior secondary cannon performance, I see the second case you mentioned as a valid comparison. I was originally going to use the Magrider's PPA, but as the PPA-H was yielding superior stats, and the question was in relation to acceptable KPU in a ground combat, vehicle oriented scenario, I chose to use the PPA-H's superior performance. The Magrider's PPA had ~20% worse KPU then the P2-120 AP, at least the PPA-H surpassed it prior to the nerf.

As for why we would compare Harasser's weapons to MBT weapons, given that the Harasser secondary weapons are worse, but otherwise identical versions of the MBT versions, this lets us gain at least some insight into any number of things.

If a -H Version of a weapon is outputting superior performance, this can mean anything from: it's Alpha / ambush potential is quite high (as is the case for weapons you can spam at point blank (Saron), or weapons that can OHK (Halberd) - Or it can mean that the weapon is simply extremely easy to use, and that the ability to maneuver around the battlefield at high speed is of greater value then normal, as the weapon can still be used at high speed (Generally, higher velocity weapons, or weapons with high splash.

The Gatekeeper is a good example of the first, the Marauder is a good example of the second, the Marauder currently leads Harasser weapon KPU

Shortly after the Gatekeeper came out, I posted a nice pile of stats for all the MBT / Harasser weapons - after a great deal of stats discussion and analysis, I came to the conclusion that KPH is too skewed of a stat to really use.
(Context: https://www.reddit.com/r/Planetside/comments/3lyvpl/piles_o_data_for_new_secondaries_mbts_and/ )
Here's the AI -H weapons in the time immediately before and after the PPA nerf.

As you can see from the above, the Marauder-H was competing nearly on par with the PPA-H, and surpassing it at times. The Canister was doing nearly as well, too - These stats do not reflect actual combat performance. For KPH, the way this is calculated in PS2 Oracle is not clear. Kills/Playtime. When is time measured? Is time measured from the moment you pull the harasser, and last until you loose the harasser? Does the clock keep counting down for a few minutes after every kill? Does it count whether or not the user of a weapon got a kill at all with that vehicle? There are too many unknowns, but the above screenshot I posted makes me highly suspect it's validity.
Population density also plays a large role in this - I know I can drive around for a half hour without seeing any enemy units at certain times of the day, and this may vastly drag down the stat, if that time spent doing nothing is included in the stat. For that matter, if we assume that time is only counted when a gun is manned, this will drag down MBT Primary / Lightning KPH in relation to secondary guns due to the fact that the pilot is manning the gun for as long as the tank exists.

If this is how KPH is measured, then empty secondary guns have even more skewed KPH due to solo driving / gunning harassers wandering the battlefield. If pilot only hops in the gunner seat when there's something to kill (in the case of the KPH graph set you posted) That PPA-H gunner will simply hop in the turret, starting the clock... get 5-6 kills, then hop back in the driver's seat, stopping the clock. There has been no less time spent, but the way the stat was measured implies that it's less useful then the same prowler primary that simply has more time spent due to the driver also being the primary gunner.

This would also explain the very low KPH stat of the Prowler P2-120 AP - Prowers are best used shelling from afar. If a Prowler gets 6 rapid infantry kills at 150 meters, then backs off and re-engages a few minutes later ... is the clock still ticking? By the same logic, did the PPA Harasser blaze through enemy lines and mow down those same 6 people, did the clock stop when they died seconds later? The Prowler may have obtained those kills in the same period of time as the Harasser, but the Prowler's KPH will be vastly lowered by the fact that it survived, keeping the clock running.

There's also the issue of the relatively short average base-capture time when population is at it's highest. Dropping a platoon on a base will clear out the place in minutes - there are only so many kills to be claimed before the enemy is gone, and the more effective the weapon is, the faster this goes. This results in many short-duration spikes in performance, which are then flattened by long periods of inactivity. With KPU, it's clear, cut, and leaves much less room for unknowns. How many kills did this user get in a given day? Ok, count it towards the stat for that one user, then average all users... That's all. It's a good way to measure average weapon performance, and it's the best way I know of without access to Daybreak's internal metrics.

With Racer, both the vanguard and magrider go 65, 67 for the mag if you go at an angle.
That said stock/rival mags are the slowest and need to be bought up to 55 base speed like Vanguards.

Yep, this is correct - and I agree with the solution you suggest.

Mag goes at 64*. The angled speed of 67 also comes with the drawback of not being able to fire in the direction you're going, a unique trait of the Magrider.

Are you sure? You can get it to go 65 on flat surfaces like the spawn area in VR (being the flatest are in there) and on esamir rivers too. The 64 to 65 is pretty slow and takes a bit, so its probably just cusping on whatever the game measures as 65.

I'm certain, although at the end of the day the difference is minimal.

All this stats negate the skillcap of the weapon.

For example you can say pounder is op but take a guy who have learn on raven and put it on a pounder.
The first attempt is always a big fail, you can't be the lob king with the first attempt.
This require some train.
The problem of ppa it's just the weapon have large mag big splash , pinpoit no drop.

PPA not need an improve but a total rework to make a close range ai weapon.

I'm still not sure what exactly we can draw from the data you presented. Definitely the idea that you seem to suggest, that the PPA was unfairly and overly nerfed, cannot be supported by such a comparison.

When you compare the PPA-H to other ESAI -H variants, this is what they look like today:

-H ESAI by KPH
Marauder- 114.1
Canister- 97.13
PPA- 67.96

MBT ESAI by KPH
Canister- 60.5
Marauder- 56.69
PPA- 36.5

So you can still make an argument that PPA has been overnefed using direct comparisons to factional equivalents. There's really no need to compare it to unlike things, because that comparison doesn't really tell us anything useful (if those things are over or underperforming).

KPH is measured by dividing the total number of kills made by a weapon by the total amount of time that someone is behind or holding that weapon within a set period. If the vehicle seat is empty, it does not count that time.

Sure KPH isn't perfect. Two good examples of that are C4 and sidearms, where the weapon is only "drawn" for short periods of time and then put away. So they get high KPHs because of low usage time.

On the other hand, that's why comparing a Harasser weapon to a tank primary weapon is inherently flawed. Comparing one Harasser weapon to another Harasser weapon makes more sense, because in the even that KPH is skewed, it will be skewed more or less equally. But if you compare one Harasser weapon to an MBT primary, you are going to get very different usage patterns. That's not limited to KPH, it would also flaw any comparisons made with KPU as well.

This image, right here, is the one that shows the PPA nerf and it's accompanying drop in performance.



This is meant to depict just that - the drop in relative performance to another faction's premier ground-vehicle based ES weapon performing extremely well against infantry. I went as far as to state that I am not defending the PPA or the P2-120 here, and I was not attempting to elicit a specific response, or make a suggestion in my original post. I was keeping my own opinions out of the data, and attempting to present facts.

This is the question I asked is the following.

Now, to address my own opinion, I made a post on that a while back, it's much too long winded to go into here, but yes, I think the PPA was overnerfed.
https://forums.daybreakgames.com/ps...roton-ppa-ii-thoughts-and-suggestions.234054/

This is quite correct, and when comparing factional equivalent weapons, this is a good approach. I have done this at extreme length, to the tune of 50-100 pages total over the last few months, this thread was an attempt at something different. It's still a valid topic, however. Looking at ES-AI alone, however, is not good enough - one must take into account the asymmetrical balance intent behind the weapons and units of various factions. The Prowler's primary and secondary vs the Magrider's primary and secondary are the best example of this I can think of. Prowler's are intended to have excellent main cannons and inferior secondaries, Magriders have poor primaries, and excellent secondaries.

This has been the general design until now - Currently, the PPA is underperforming compared to the Marauder, The Saron is underperforming compared to the Gatekeeper, and the P2-120 is still vastly superior to the Magrider's primary lineup. When you examine the KPU / V-KPU of other systems such as Fractures vs Vortex, Lancer vs Striker, etc - you see that the TR does indeed underperform in other areas, thus, it does justify having superior performance in others. The Gatekeeper, however, stacks the strength of an excellent Primary with the strength of an excellent Secondary, something no other faction has - this serves as a massive performance multiplier, when one considers that the Marauder is also performing extremely well compared to it's competition.

If the combination of an extremely potent Prowler Primary / Secondary setup is the new benchmark for asymmetrical balance, then both the NC and VS need tremendous buffs to other AI / AV units, either MAX AI and AV, ES RLs, ESF buffs, etc. If the potent combination of Prowler Primary / Secondary performance we currently have is not the new benchmark, then something is overperforming and must be nerfed, followed by buffs for it's other AI / AV systems to compensate for the lack of performance in other areas, most notably long range AV.

(The Prowler is all the TR had for long range AV, stacking more AV on the same unit has made their performance lopsided to the point where we see lopsided usage hours like these. The TR currently only has one set of well-performing options, and these options are on the Prowler / Harasser. With well-performing infantry based AV, this would not have to be the case. )


But most of the above is off topic, I digress.

Then yes, this stat is inherently flawed, and means that KPH of secondary weapons cannot be accurate when comparing them to even other secondary weapons. Until recently, Prowler secondary gunners were used at a rate of up to 30-35% at most. In the last few weeks, it has risen to 80%, to match where the VS has been for years. With that much of a difference in use time, I can't see the stat being useful in any scenario that involves seats that may be empty, although I do see it's validity for infantry weapons, one of which will always be out in combat, for the most part.

Yep, I wish there was a way to break this down by class, i'd be interested in seeing numbers for Stalker infils, specifically.

For KPH? I agree, that's a good reason to not use it to compare vehicle weapons. People frequently solo gun their own vehicles, and tank pilots are inflating the hours to kills ratio, dropping the stat significantly. The usage rate of secondary gunners between factions skews this even more, further reducing it's validity. Until recently, the VS used the most secondary gunners, the NC were 10-20% behind, and the TR very significantly fewer secondary gunners overall.

KPU works as I described in my original post (I went and confirmed it with the PS2 oracle site creator a while back) If 1 unique user gets even 1 kill, they count for the stat for that day - however they will continue to only count for one unique user, stacking up kills that count towards that day's stat. I can see this skewing the outcome in scenarios where an individual gets a single kill and simply logs off, as that will lower the average, but a gunner can sit around all day without artificially reducing performance stats. KPU gives us a better indication as to average performance for people that are actually using the gun, rather then people who pull it and sit around AFK, or gunners that hop in for a few minutes, fire randomly, and leave without getting any kills (My most frequent gunner type, when I drive by clumps of friendlies and V5 for randoms)

But I mean what's the point of making this comparison or observation?
So we know that the PPA dropped in effectiveness. That's what nerfs tend to do. We don't need a thread that simply says, "hey, the PPA got worse when it was nerfed".

Instead, the question is usually one of degree.

1) Should the PPA have been nerfed in the first place?
2) Should the PPA have been nerfed that hard?

And to that, your observations between the PPA-H and Prowler primary cannon can't help us. Because the Prowler primary cannons are not the PPA equivalents. Whether the PPA-H ended up better or worse than the Prowler primary cannons is pretty irrelevant to where the PPA should be, as the PPA can only be compared to equivalent weapons (the Marauder and Canister, and to a lesser extent the Kobalts).

I guess I would disagree as to this "intent". I don't feel that the Prowler having horrible secondaries was by design, but by accident. In my view, all MBTs ought to have equally performing primaries and equally performing secondaries. This simply means that in the case of the Prowler, the primary cannon need to be toned down (or more preferably competing weapons buffed) while the secondaries need to be buffed.

I'm not sure that your data on 2/2 usage rates is correct. I ran the numbers myself, both in 2014 and fairly recently too.

That later data was the monthly averages from mid-late September 2015, after the introduction of the latest wave of ESAV secondaries. The difference is not significant enough to suggest much seat-switching shennanigans, at least in a pattern that is much different between MBTs.

Again, the problem is usage time per unique, which is where KPU falls apart. Harassers spend little time using their weapons, so if they get fewer kills that's okay. MBTs spend more time using their weapons, so if they get more kills that's okay.

KPU does not give an accurate impression of how long a player must invest to get a single kill. As my figures show, the PPA has to spend only 50 seconds or so to get a kill, which is about a third of the time a Prowler primary gunner must spend to get a kill.