U34 Pickrate Analytics

Are you a fan of numbers? Well look no further, there are a LOT of numbers in this blog post. Continuing the tradition started at the end of Update 31, this blog post is going to be an analysis of all the loadouts either created or updated during Update 34: Modest Expectations (aka "U34"), which ran from April 22nd, 2021 through November 4th, 2021. Technically there's another 2 weeks' worth of loadouts included in this, because the Karl.gg devs weren't ready to add the U35 weapons until Nov 18th.

In total, there were at least 2,354 loadouts submitted to this website during the 30 weeks that U34 was live (28 + 2 extra). U32 had an average of 8.83 loadouts per day, U33 increased that to 12.29, and U34's average was 11.21 loadouts per day. Considering that 210 days is almost three times as long as U33's 76 day period, it's pretty impressive that the average number stayed so high.

In keeping with the format that I've used for the past pickrates blog posts, I'm going to place my observations and conclusions up at the top of the blog post so that people who aren't interested in wading through all of the numbers can still read the interesting stuff without having to scroll ALL the way to the bottom. For anyone who hasn't read one of these before, brace yourself for a wild yet somewhat monotonous adventure through numbers and percentages.

Part 1: Observations and Conclusions

Miscellaneous observations

One of the first things that stood out to me was the apparent "jump" in popularity of Gunner loadouts. In the past updates, the four classes all had a pretty similar number of loadouts submitted (within about 10% variance from least number to most). Yet during U34, Gunner had almost 25% more loadouts submitted than the other three classes (which were all around the same number, as expected). I don't have any definitive reason as to why there's such a discrepancy; only a speculation. A few months after U34 was released, the Modding Support Update came out too. It wasn't a full update in its own right, but rather a completion of the what U34 had promised. Since that update, the number of mod-users has increased at least tenfold. My best estimate, based off of Mod.io's listing for Blueprint Mod Manager, is close to 16x more mod users post-Modding Support. Some of the most popular mods to use are Custom Enemy Cap, Swarm Size Control, and Starship Troopers -- all of which drastically increase how many enemies that your team has to fight. It's plausible that with so many more people fighting so many more enemies in missions, that Gunner's popularity started increasing due to the deep ammo wells of his Primary weapons as well as the strength of his Shield Generator. I'm not certain that modded difficulty alone is sufficient cause for Gunner's loadouts to be nearly 25% more frequent than the other three classes, but it's the best reason I can think of. Typically Gunner has lagged behind the other classes, with Driller being the most popular by a small margin.

Continuing about Gunner in particular, I noticed a massive change in trends. Before U34, typically I would see Autocannon + Carpet Bomber combined with BRT + Composite Casings; Carpet Bomber for medium DPS waveclear and Composite Casings for medium DPS single-target damage. During U34, however, the go-to combo was Autocannon + Neurotoxin Payload with BRT + Lead Spray; a potent combo of low DPS waveclear and slows from NTP and close-range high DPS from Lead Spray. It might be worth doing more research into other loadouts of this style, trying to ascertain what lets it successfully contradict the prevailing wisdom about weapon builds. At first glance, it sounds like stock Engineer Shotgun & PGL but on Gunner.

Next on the list: Scout's M1000 OC "Supercooling Chamber" (aka "SCC"). Historically, this Overclock has always struggled to find its place in the game. The complaints commonly lodged against it are along the lines of "clunky to use, punishing to stand still, not enough damage in exchange for how long it took to charge, took too much ammo away" etc etc. Yet, despite all that, it only received two very minor changes in U34: Max Ammo penalty reduced from x0.635 to x0.665, and Focus Speed penalty reduced from x0.5 to x0.6. Even more perplexingly, the pickrate almost seems to have doubled from 8.33% in U33 to 14.64% in U34! For six Overclocks you would expect a pickrate of ~16.67%, so that would seem to be a quite healthy pickrate. My confusion arises from this observation: the percentage of loadouts submitted that use SCC seems entirely unrelated to how little I've heard people talking about it in-game, or on Discord servers, or on the subreddit, let alone how infrequently I've seen it used in game. For the whole duration of U34, I never once played in a lobby with someone using SCC; not even as a test. I'm fully aware that "anecdotal evidence" isn't actual proof, but you would think that if it truly was being used in 1/6 of all loadouts that I would have seen it used at least once. This one Overclock's pickrate has cast a lot of doubt into my mind about how accurate the analyses I've performed in the past have been. After a couple of days of thinking it over, I still don't have any good answers for this dilemma; only questions still linger in my mind.

What balance changes worked?

• Flamethrower baseline changes (more DPS, stronger Burn DoT): along with the other Flamethrower balance changes, this raised Flamethrower's usage rate from 43.51% to 48.53%.
• Subata OC "Tranquilizer Rounds": for such a small change, that extra 2 Rate of Fire seems to be all that this OC needed. 3.45% pickrate rose to 13.57%. The Elite enemies may have affected the pickrate too; long Stuns and Weakpoint damage are both useful vs Elites.
• EPC baseline Charged Shot changes: reducing the Heat per Charged Shot from 100% to 40% of the Heat Meter successfully removed the need for TCF in order to fire multiple Charged Shots back-to-back. It had no discernable affect on pickrates, but it did succeed from a design perspective.
• EPC T2.A "Heat Shield": moving this from Tier 4 to Tier 2 definitely helped this mod out. Along with the changes made to Heat Pipe, this mod's pickrate jumped from a paltry 9.23% to 38.85%. It's still not quite within the bounds of what I would consider "healthy" pickrate for one of two mods in a tier, but that's a massive improvement for just moving to Tier 2 without changing its effect.
• EPC OC "Heat Pipe": for sure one of the most painful nerfs that Driller mains had to endure, Heat Pipe was one of the most-complained about changes at the time. Now, 6 months later, it could be argued that it was a justified and healthy change. Its pickrate dropped from an extremely un-healthy 56.35% down to a very healthy 13.71%, as well as having one of its new penalties incentivize picking the aforementioned T2.A "Heat Shield". I think that this OC works pretty well for comboing Flying Nightmare with Frozen enemies, but no longer has quite as strong of a synergy with Thin Containment Field (what it was known and loved for).
• EPC OC "Persistent Plasma": U34 did a massive overhaul to this Overclock, and it had quite a positive effect! Jumping from virtually unused at 2.38% pickrate all they way up to 14.11% during U34, the new Persistent Plasma has a bunch of synergy with both the Flamethrower and Sludge Pump, to stack the slows and increase the damage dealt by all of the DoTs dealt by the weapons.
• Warthog T5.A "Turret Whip": also massively buffed by U34, it went from "barely healthy" at 28.24% to "vibrant" at 41.39%. I don't think that it's ever supposed to be better than Miner Adjustments, but I think it's in a really good state right now both in terms of design, balance, and popularity. It may not be for everybody, but it's a great addition to Engineer's kit. Give it a try if you haven't already!
• SMG OC "Turret Arc": literally every aspect of the lightning beam got buffed, and I've heard scattered reports that it's situationally very good nowadays. It went from just a touch underpicked at 8.6% to 12.86%, which is a healthy pickrate for something designed to be situational.
• SMG OC "Turret EM Discharge": this one has a little bit of conflicting data. In U33 it had a pickrate of 17.2%, so it was already "healthy" by my standards. Others have said that it was already situationally strong in U33, before its buff. On the other hand, many of the veterans I consult with thought it was pretty much never worth equipping and that it penalized the base weapon too much. When U34 buffed it (quite excessively, too) it jumped up to 22.38% pickrate, but it also started generating clips of two turrets soloing a swarm on Escort Duty and questions of "why was it buffed?" At the end of the day, I think it's not yet Overpicked and to a degree only situationally strong. However you want to slice it, U34 succeeded at getting more of the community to experiment with this Overclock and start running it.
• PGL T3.C "High Velocity Grenades": simply by moving it away from a tier that contained both Ammo and Area Damage, the pickrate of this mod rose from 5.74% to 17.36%. Still on the low side, but a step in the right direction. I've heard people talking about how they can now use it with Fat Boy. I imagine that if and when Hyper Propellant gets nerfed into a balanced state, the usage of this mod will start increasing as people move back to using PGL as a, you know, grenade launcher.
• PGL T5.A "Proximity Trigger": a simple change was all it needed. x1.1 AoE Radius raised its pickrate from 14.05% to 24.76%. It's still a little on the underpicked side, but I don't think it needs any more buffs. It's in a great state from design and balance perspective, so I think that once Hyper Propellant is no longer the dominant force behind PGL usage this will start getting a lot more love.
• PGL OC "RJ250 Compound": at the time, everyone thought that it was being over-buffed. However, the pickrate only rose from 5.74% to 13.95%, a very healthy place to be for a six-Overclock weapon.
• Minigun T3.B "Improved Stun": simultaneously buffing this mod while also lowering the baseline Stun of the Minigun seems to have done the trick. 11.61% to 23.13%; not quite the 33% that you could hope for, but much better balanced against the Armor Breaking at least. As long as Blowthrough Rounds stays in T3.C, it's going to dominate the pickrates of Tier 3.
• Minigun T5.A "Aggressive Venting": such a simple change -- make the mod reward Overheating, instead of penalizing the player. Without a cooling mod, it's actually faster to just Overheat the weapon and wait 5 seconds than it is to wait 6+ seconds to cool back down to zero. Jumped from a 6.56% pickrate to 22.8%, and it's in a really good state from design and balance perspective. I don't think it will ever be more popular than Cold as the Grave, but it's in a much better state now.
• Autocannon baseline RoF changes and T2.C "Lighter Barrel Assembly": one of the more resounding successes was how GSG changed the Autocannon Rate of Fire rampup mechanic. If you kept T2.C equipped, then it's exactly the same as it was pre-U34, but it's about 50% better baseline. Apparently this threshold was good enough for the community, because T2.C's pickrate dropped from an extremely unhealthy 61.05% to a much more reasonable 40.96%. It's still a little on the high side for a 3-mod Tier, but much healthier overall.
• Autocannon OC "Carpet Bomber": all that GSG did was tickle this Overclock with a feather duster nerf (-0.1m AoE Radius, -1 Direct Damage) and buff some of its neighbors, and its pickrate fell from 39.78% to 23.24%. It's still a bit on the high side, but nowhere near as dominant as it used to be.
• Autocannon OC "Combat Mobility": I wrote a bit about this Overclock's stuggle to have an identity, but it seems that U34 finally found a good one for it: the T2.C alternative. From 2.15% pickrate to 12.23%, it appears to have found a home amongst the other Overclocks.
• Autocannon OC "Neurotoxin Payload": similar to SMG's "Turret EM Discharge", this Overclock already had a pretty healthy pickrate in U33 at 17.2%. Nonetheless, it was generally regarded as a little underpowered so GSG decided to give it some buffs. And by "some" I mean "a lot". It rapidly stole the spotlight of U34, and has become the new dominant Overclock on the Autocannon. It may need to be slightly nerfed in the future, but for now it enjoys a 25.08% pickrate, just entering the Overpicked territory.
• GK2 baseline Ammo buff: back when U34 arrived, people were incredulous that GSG would only give it +25 Ammo. People were clamoring for 50 - 100 more ammo, trying to push it to compete with M1000's numbers. Yet despite "only" +25 ammo, it changed the balance of power in Tier 2 from a 38.78%/61.22% split to 44.89%/55.11%. Unforunately the GK2/M1000 split was worsened by U34; more on that later.
• M1000 baseline changes to Damage, WP Bonus, and Charge Speed: my nomination for the "least well-received" changes in U34 -- even moreso than Thin Containment Field and Heat Pipe -- was the changes made to M1000. For over a year, it had been referred to as the "sniper rifle" of the game, but it seems like GSG wanted to remind everybody that it was a "marksman rifle". It wasn't even a nerf, just a change that very few people wanted or liked. Despite its bad press, I am forced to recognize the positive effects it's had on the mod tiers. Tiers 2 through 5 all have had a noticeable improvement in pickrate balance, and the Focused Shot build is no longer dominating at a 3:1 ratio. Tier 1 actually is a little worse balanced than it used to be, but I suspect OC "Hipster" is to blame for that, not the baseline changes.
• M1000 T2.C "Hardened Rounds": what a strange and fickle effect Armor Breaking is. On Tier 4, it languished in obscurity at a 10.1% pickrate. Yet, simply by moving it to Tier 2 it became the "talk of the town", threw everyone into a hipfired-build frenzy, and now is sitting pretty at a 52.76% pickrate. Overpicked. Something GreyHound suggested was moving it again, this time from Tier 2 to Tier 3. Regardless, U34's change was too successful.
• M1000 OC "Hipster": don't be surprised; everyone knew this was coming. Its meager 11.46% pickrate in U33 skyrocketed to 36.43% in U34, another case of balance changes being too successful. Between making its Direct Damage penalty less severe, boosting its ammo gain, and letting it hit multiple Breakpoints with only one or two mods, no one is surprised that Hipster has dominated the M1000 meta.
• Zhukovs OC "Custom Casings": changing its penalty from Direct Damage to Rate of Fire was all this needed; its pickrate rose from an Underpicked 8.43% to a more healthy 14.35%. Not quite as good as I would have hoped, but a step in the right direction.

What balance changes failed?

• Cryo Cannon OC "Perfectly Tuned Cooler": it doesn't take any special insight to recognize why adding two penalites to the previously-Clean Overclock "Perfectly Tuned Cooler" would barely affect its pickrate. It still offers both its +1 Cold per Particle and +0.8 Rate of Fire, both of which let players Freeze enemies significantly faster. Its pickrate remains far too high, only going from 50.4% to 45.6%. At its core, Tuned Cooler suffers from a design problem, not a numbers problem. Until the identity of Cryo Cannon gets moved away from "braindead freezy combo gun" and turned into a proper weapon in its own right, Tuned Cooler's two bonuses will mean that it's always going to be the go-to Overclock.
• EPC T5.B "Thin Containment Field": for how intense and controversial the discussion around TCF was during updates 31-33, surprisingly little was changed about this mod during U34. Its Heat per Charged Shot multiplier was changed from x0.25 to x0.8 (functionally from 25% Heat per Charged Shot to 32%), and the minerals mined by its implosion now get scattered around the cave instead of dropping in a neat little pile. From how everyone started complaining about it and saying outrageous things like "TCF is dead, REEE", you'd think that it was removed from the game entirely! It's clear that the devs at GSG were trying to keep the high reward for skillful players who use TCF in combat, continue letting people use it to mine minerals (just not as conveniently), and also remove the "need" for it in order to chain multiple Charged Shots together back-to-back. In that respect, they achieved their goals. From an actual balance perspective, though, it failed to do anything note-worthy. Everyone knows that neither TCF's damage nor its radius are affected by the mods, and almost nobody uses Charged Shots for the Charged Shots. It's all about the sparkles that get added, like Flying Nightmare, TCF, or Persistent Plasma. As a result, people just go for ammo and charge speed; nothing else affects TCF, so in their eyes nothing else matters. That's evident by the abysmal pickrates of T1.C, T4.A, and T4.C, combined with the absurd pickrates of T1.B and T4.B. If U34 was GSG posing the question "is the playerbase willing to use EPC in ways other than TCF?" the answer has been a resounding "no". There's nothing left to do now but wait for the inevitable weapon rework/redesign and the gutting of TCF. Any chance that TCF had to skate by the next balance patch unscathed (like it did in U34) has gone out the window.
• EPC OC "Heat Pipe" (again): one of several changes I'm listing in both of the "Success" and "Failure" lists, Heat Pipe has been a very mixed bag during U34. Its charge speed and ammo saving bonuses have a great synergy with Flying Nightmare, and really anything other than TCF. Unfortunately, before U34 Heat Pipe was exclusively because of its synergy with TCF, so this balance change gave so much whiplash that it generated a bunch of backlash. I've already made the case above that the change was "healthy", but I'm not sure anyone in the community has been convinced that the change was "good". Honestly, one of the biggest standards to hold this balance change to is the effect it had on Energy Rerouting: it went from 7.94% pickrate to 31.05%, simply because it was the next-best OC to synergize with TCF. That's a failing of TCF moreso than Heat Pipe, but it just goes to show that Heat Pipe's change didn't really balance the EPC's Overclocks, it just moved the community's focus to a different one that worked with TCF.
• Autocannon OC "Neurotoxin Payload" (again): of the four buffs that NTP received, it probably only needed 2 of them at most. As it stands, it's perceived as "equal or superior to Carpet Bomber" due to an equal AoE Radius, stronger Crowd Control, similar total damage, and only at the cost of a little less DPS. I'm glad that NTP is enjoying its time in the spotlight, but I think that it was overbuffed and probably needs to be dialed back a bit in order for the Overclocks as a whole to be in a healthier state.
• Revolver T3.C "Hollow-Point Bullets": one of the goals GSG had for the Revolver changes was to make the Ammo mods more attractive so that players would stop taking both Damage mods, and the other goal was to give players less incentive to pick T3.C. The change made was to increase the baseline WP Bonus from 15% to 25% and decrease T3.C from +50% to +35% (net -5% change), but that change seemingly had a positive effect on pickrate: 61.02% to 68.33%. The truth of the matter is it was probably affected more by Volatile Bullets, but that's not the point. The goal was to balance Tier 3, and the change implemented failed to do that.
• Revolver OC "Chain Hit": a massive buff, from 33% Weakpoint Ricochet to 75%, and yet its pickrate remains only changed from 3.53% to 4.53%. At its core, I think that the bonus provided by Chain Hit is antithetical to how the playerbase wants to use the Revolver. When all of the other Overclocks provide a way to boost its damage per shot to ridiculous numbers, something that gives a chance to hit one more enemy on Weakpoint hit just seems lackluster in comparison. I wonder what would happen if Chain Hit was changed to work like Drak-25's new "Impact Deflection" Overclock...
• Revolver OC "Homebrew Powder": admittedly, a clever attempt to make "random damage" be worth picking. Observably, it had an effect: 0.88% pickrate to 2.42%. Fundamentally, the playerbase is unwilling to leave killing blows to chance. No amount of number tweaking can fix a design issue.
• Revolver OC "Volatile Bullets": once more, it's no secret that VB has always been a strong Overclock. It had never been popular because it required enemies to be Burning (something that could only be achieved via teammates, Minigun T5.C Hot Bullets, Minigun OC Inferno, or Incendiary Grenade), and that effect is mutually exclusive with the very-popular Frozen status effect. Then, when U34 came along and pretty much made VB identical to U33's Damage per Shot without Burning, its popularity exploded. We hadn't even gotten a taste of Hurricane T5.A yet, and already its pickrate rose from 12.39% to 22.36% across U34's duration. There's a difference between buffing something to make it more appealing, and making something potentially "best in slot". I'm extremely worried about what pickrate Volatile Bullets will have when U36 comes around and I write the next post in this series...
• BRT baseline changes to Damage and RoF: it's quite obvious what GSG tried to do with these changes: provide situations where players wouldn't pick both Damage mods and T2.C's RoF. Unfortunately, these changes had no effect on the mods' pickrates, and just inadvertently made Lead Spray even stronger.
• M1000 T2.C "Hardened Rounds" (again): as I alluded to above, the Armor Breaking mod on M1000 seems to have gone from Underpicked to Overpicked simply by changing places. It's still the same +220% AB that it had always been, it's just no longer competing against Weakpoint Bonus or Blowthroughs. As a result, its popularity has skyrocketed and people have enthusiastically embraced the "bodyshot hipfire" playstyle enabled by AB. I'm really at a loss as to why this change failed; from a design perspective the concept of "faster charge time (dps), higher accuracy (less damage lost), armor breaking (less damage lost)" seems like it should work better than it does in practice. I don't think nerfing the numerical value (+220% AB) will solve the problem; I'd be curious to see how GreyHound's suggestion to move it to Tier 3 would work out.
• M1000 OC "Minimal Clips": from Overpicked at 37.5% because it gave +16 ammo, to Underpicked at 7.14% because it only gives +2 or +4. From a balance perspective, I think that its numbers are tuned fine; it's just no longer "the best OC to pick for every build" anymore. So, much like Woody's nightmare about Andy in Toy Story, the playerbase collectively seems to have dropped Minimal Clips and said "I don't want to play with you any more" to go play with the new Buzz Lightyear: Hipster. I'm not sure that any amount of reasonable number buffing would raise the pickrate of this Overclock, but neither do I think of it as having a design flaw to overcome. It's in a very weird state, where its popularity is solely dependent upon its competitors being properly balanced.
• M1000 OC "Hipster" (again): once more unto the gates of Overpicked, into the very mouth of Meta itself. Out of the frying pan and into the fire. All these idioms and more; Hipster got way over-buffed. We just traded Minimal Clips for Hipster as the only Overclock that the playerbase wants to equip on the weapon. It's easy to see why; it has high DPS, high total damage, and due to how it reshuffles the damage per shot and the total ammo it's not as punishing to miss your shots. I think the principle of changing the Direct Damage penalty from multiplicative to additive is good, but it should have had some kind of accompanying RoF penalty (from +3 to +2) to keep DPS in check, and at the very least keep the Ammo at its old multiplier (or potentially less) instead of increasing it like U34 did.
• Boomstick T3.A "Improved Stun": despite it now being an amazing Stun delivery system now (50% chance to Stun for 6 seconds per pellet), its prickrate barely moved from 6.86% to 8.79% during U34. Fundamentally, the playerbase is using the Boomstick as either a mobility tool via Special Powder (so they take ammo) or as a high burst DPS weapon (so they take pellets per shot). No one seems to want to use it for Stunning enemies because "death is the ultimate crowd control". Neither design nor numerical balance is the problem here, it's just not what players want to use.
• Zhukovs baseline Ammo change: although it was intended to reduce the pickrate of T4.C by moving 50 ammo into baseline, instead it seems to have just pushed both Embedded Detonators and Cryo Minelets. Not only are those two the strongest OCs on the weapon right now, they also rely heavily on ammo. In spite of moving ammo from T4.C to baseline, the pickrate of both ammo mods increased during U34 because the community values Zhukovs for Embed Dets' DPS or for Cryo Minelet's crowd control and damage boost.

Closing thoughts

This blog post has already ballooned far beyond my intended length, and you haven't even gotten to the data yet, so I'll try to be brief as I wrap up. As much as it pains me to say it, from a high-level perspective I don't think that U34 had very much success in balancing the weapons. In comparison to U33's pickrates, there are only 5 fewer Overpicked options (59 to 54) and only 6 fewer Underpicked options (68 to 62). In the scope of 303 Mods and Overclocks between the 16 weapons, that's a 1.8% change. Similarly, there are 80 mod tiers and 16 Overclock slots -- only 8 of the "most popular" mods changed, and 6 of the "most popular" overclocks, 16.67% sum difference. It's not reasonable to expect all of the most popular mods to change, but I was hoping for a bigger difference than just 10%. Comparing the 318 U33 pickrates to the 319 from U34, very little was meaningfully changed and it makes me kind of sad. GSG put in a ton of work on the balance changes, received a lot of flak for their efforts, and at the end of the update there was very little reward for all their trouble. The things that were popular, simple, and effective in U31 still are in U35, and the playerbase is indifferent to trying new builds, experimenting to find other things that they might like, and potentially just unwilling to change.

I'm extremely worried that U34 might have provided evidence that stagnation has set in; the first step in a playerbase becoming disinterested in a game. So little changed in U33 that I didn't have anything meaningful to say in its pickrate blog post. In contrast, despite 103 balance changes being done when U34 arrived, still very little has actually changed in how we play the game. The trends that were started in U31 and U32 have just continued right along through U34, as if most of that work was for naught. I guess there's not much that I can do except just repeat the message I've been championing for two years now: weapon build variety is crucial to DRG's longevity.

To anyone who has read this far, I challenge you to try a new weapon build every single mission for at least two weeks. It's really simple; just pick a Mod or Overclock that you've either never used or haven't used in a long time, figure out the rest of the build that complements your selection, and launch a mission. You might be surprised how much fun you might have with things you would have never thought to try if your brain was still in "min/max mode"! Especially with 4 new weapons and 26 new Overclocks to unlock, there's no better time to try new things!

Rock & Stone, -MS

Part 2: Weapon popularity per class

• Driller (509 primary weapon builds, 482 secondary)
  • Flamethrower 48.53%
  • Cryo Cannon 51.47%
  • Subata 45.44%
  • EPC 54.56%
• Engineer (523 primary weapon builds, 520 secondary)
  • Shotgun 58.13%
  • SMG 41.87%
  • Grenade Launcher 59.81%
  • Breach Cutter 40.19%
• Gunner (644 primary weapon builds, 595 secondary)
  • Minigun 48.45%
  • Autocannon 51.55%
  • Revolver 57.31%
  • BRT 42.69%
• Scout (515 primary weapon builds, 504 secondary)
  • GK2 43.69%
  • M1000 56.31%
  • Boomstick 54.17%
  • Zhukovs 45.83%

Part 3: Most popular build per weapon

This section is just aggregating most popular mods and overclock for each weapon into one "build"; don't misinterpret this as "the same distinct build that was repeat-submitted the most number of times from many users".

• Driller
  • Flamethrower AABCA + Sticky Fuel
  • Cryo Cannon CABCB + Tuned Cooler
  • Subata BABAA + Automatic Fire
  • EPC BBBBB + Energy Rerouting
• Engineer
  • Shotgun ABCBB + Magnetic Pellet Alignment
  • SMG BBBBB + EM Refire Booster
  • Grenade Launcher BABBB + Hyper Propellant
  • Breach Cutter AAABB + High-Voltage Crossover
• Gunner
  • Minigun CBCAB + Lead Storm
  • Autocannon CBBBA + Neurotoxin Payload
  • Revolver BCCAB + Magic Bullets
  • BRT ACBCA + Lead Spray
• Scout
  • GK2 BBCAC + Bullets of Mercy
  • M1000 BCABA + Hipster
  • Boomstick AACAC + Special Powder
  • Zhukovs ACBCB + Cryo Minelets

Part 4: Overpicked mods and OCs

In order for a Mod or Overclock to qualify as "Overpicked", it had to be selected at least 1.5 times more often than it would be if every mod/OC at that tier was picked equally. To phrase it another way: if there are two mods, then for a balanced pickrate each one should be picked around 50% of the time. In order for one of those two to qualify as Overpicked, it would have to have a pickrate of 1.5 * 50% = 75% or higher. Tiers with 3 mods need a 50% pickrate, 5 Overclocks would need 30%, 6 Overclocks 25%, and 7 Overclocks would need a 21.43% or higher pickrate.

54 Overpicked, 303 total Mods and Overclocks

• Flamethrower
  • T2.A "Unfiltered Fuel"
  • T4.C "More Fuel"
  • OC "Face Melter"
  • OC "Sticky Fuel"
• Cryo Cannon
  • T2.A "Larger Reserve Tank"
  • T4.C "Larger Reserve Tank"
  • OC "Tuned Cooler"
• Subata
  • T1.B "High Capacity Magazine"
  • OC "Automatic Fire"
• EPC
  • T1.B "Larger Battery"
  • T4.B "High Density Battery"
  • T5.B "Thin Containment Field"
  • OC "Energy Rerouting"
  • OC "Heavy Hitter"
• Shotgun
  • T4.B "Bigger Pellets"
• SMG
  • T1.B "Upgraded Capacitors"
  • T2.B "Recoil Compensator"
  • OC "EM Refire Booster"
• Grenade Launcher
  • T1.B "Extra Ammo"
  • T2.A "Expanded Ammo Bags"
  • T3.B "Pressure Wave"
  • T5.B "Spiky Grenade"
  • OC "Hyper Propellant"
• Breach Cutter
  • T2.A "Expanded Ammo Bags"
  • T3.A "Quick Deploy"
  • T5.B "Plasma Trail"
  • OC "High-Voltage Crossover"
  • OC "Spinning Death"
• Minigun
  • T5.B "Cold as the Grave"
  • OC "Burning Hell"
  • OC "Lead Storm"
• Autocannon
  • T1.C "Expanded Ammo Bags"
  • T5.A "Feedback Loop"
  • OC "Neurotoxin Payload"
• Revolver
  • T2.C "Expanded Ammo Bags"
  • T3.C "Hollow-Point Bullets"
  • OC "Elephant Rounds"
  • OC "Magic Bullets"
• BRT
  • T1.A "High Velocity Rounds"
  • T2.C "Disabled Safety"
  • T3.B "Increased Caliber Rounds"
  • OC "Lead Spray"
• GK2
  • T1.B "Supercharged Feed Mechanism"
  • T4.A "Hollow-Point Bullets"
  • T5.C "Stun"
  • OC "Bullets of Mercy"
  • OC "AI Stability Engine"
• M1000
  • T2.C "Hardened Rounds"
  • T5.A "Hitting Where it Hurts"
  • OC "Hipster"
• Boomstick
  • T4.A "Super Blowthrough Rounds"
  • OC "Special Powder"
• Zhukovs
  • T1.A "Expanded Ammo Bags"
  • OC "Cryo Minelets"

Part 5: Underpicked mods and OCs

Conversely, for a Mod or Overclock to be considered "Underpicked", it had to be selected at least 0.5 times less often than it would be if every mod/OC at that tier was picked equally. 2 mods would need a pickrate of 25% or less, 3 mods 16.67%, 5 Overclocks 10%, 6 Overclocks 8.33%, and 7 Overclocks would need a pickrate of 7.14% or less. In my opinion, a well-balanced system would have pickrates between 80-120% of a perfect distribution for each option in a tier.

62 Underpicked, 303 total

• Flamethrower
  • T4.B "Sticky Flame Duration"
  • OC "Sticky Additive"
  • OC "Fuel Stream Diffuser"
• Cryo Cannon
  • T1.A "Larger Pressure Chamber"
  • T2.C "Bypassed Integrity Check"
  • T4.A "Hard Mixture"
  • OC "Flow Rate Expansion"
• Subata
  • T1.C "Quickfire Ejector"
  • OC "Chain Hit"
• EPC
  • T1.C "Higher Charged Plasma Energy"
  • T4.A "Expanded Plasma Splash"
  • T4.C "Reactive Shockwave"
  • T5.A "Flying Nightmare"
  • OC "Magnetic Cooling Unit"
• Shotgun
  • T2.C "Choke"
  • T4.A "Tungsten Coated Buckshot"
• SMG
  • T1.C "Expanded Ammo Bags"
  • OC "Super-Slim Rounds"
  • OC "Well Oiled Machine"
  • OC "Light-Weight Rounds"
• Grenade Launcher
  • T1.A "Fragmentary Shell"
  • T2.B "Larger Payload"
  • T5.C "Disabled Inertia Inhibitor" (but I'm not surprised -- this one is really niche)
  • OC "Compact Rounds"
• Breach Cutter
  • T2.C "Loosened Node Cohesion"
  • T3.B "Improved Case Ejector"
  • T5.A "Explosive Goodbye"
  • OC "Roll Control"
  • OC "Inferno"
• Minigun
  • T1.A "Magnetic Refrigeration"
  • T4.C "Magnetic Bearings"
  • OC "Thinned Drum Walls"
  • OC "Compact Feed Mechanism"
  • OC "Exhaust Vectoring"
• Autocannon
  • T1.A "Increased Caliber Rounds"
  • T2.A "Tighter Barrel Alignment"
  • T5.C "Damage Resistance at Full RoF"
  • OC "Composite Drums"
• Revolver
  • T2.B "Floating Barrel"
  • T3.A "Super Blowthrough Rounds"
  • OC "Homebrew Powder"
  • OC "Chain Hit"
• BRT
  • T1.C "Blowthrough Rounds"
  • T2.B "Quickfire Ejector"
  • T3.A "High Capacity Magazine"
  • T4.A "Hardened Rounds"
  • OC "Full Chamber Seal"
• GK2
  • T1.A "Gyro Stabilisation"
  • T3.A "Floating Barrel"
  • T4.B "Hardened Rounds"
  • T5.B "Battle Cool"
  • OC "Gas Rerouting"
  • OC "Homebrew Powder"
• M1000
  • T2.B "Better Weight Balance"
  • T5.B "Precision Terror"
  • OC "Minimal Clips"
• Boomstick
  • T3.A "Improved Stun"
  • T4.B "Tungsten Coated Buckshot"
  • OC "Double Barrel"
  • OC "Shaped Shells"
• Zhukovs
  • T1.B "High Velocity Rounds"
  • T4.B "Hollow-Point Bullets"

Part 6: Raw data