• Content count

  • Joined

  • Last visited

  • Battles


1 Follower

About RamirezKurita

Recent Profile Visitors

268 profile views
  1. That's the problem with MM based entirely on WR, it tends to be extremely swingy as the more it responds the more it destroys it's own balance data. It's also the reason why most online games use an entirely separate metric, usually called by some variation of "match making rating", to judge their players. Win Rate should be the goal, not the method, of a good matchmaker.
  2. That's pretty much what a good matchmaker would do, it would ensure that every single game would be a close match. A theoretically perfect matchmaker would ensure that in every game, everyone would have an equal chance of victory. Each player should fight tooth and nail the entire time to the best of their abilities if they want to achieve victory, rather than skilled players just cruising through and farming newbies. Obviously, in practice this is impossible because of factors outside the matchmakers control or detectability (first game of the night, buggy internet, playing at 2 AM during a drunken bender, family/friends playing on your account etc), but it's still reasonable to aim for the matchmaker to account for all accountable circumstances - it's what a good chunk of most other competitive online games do. Obviously, going simply by WR would be a pointless metric, as the goal of a matchmaker is to keep everyone at 50% - if everyone's at 50% then it's impossible to determine relative player skill. That's why most matchmakers use some kind of variation on the elo system, usually termed as some kind of Match Making Rating (MMR). This has seen usage In pretty much every major competitive online game in the last decade, particularly in MOBAs which often have extremely elaborate matchmakers. There's a very interesting article on the matchmaker in Overwatch by a couple of the Blizzard Devs here, which goes into some of the stuff they have to factor in when trying to balance games (plus some of the player psychology behind it all), and from what I remember reading from developer blogs League of Legends used quite a similar system to factor in a variety of different things to try to ensure as close games as possible. I would have probably stopped playing League of Legends years before I did if it had WoWS's matchmaker, one of the game's greatest saving graces is the very good matchmaker which does a very good job of ensure nearly all games are good and hard-fought. Sure, WoWS has a few additional variables to play with compared to most online games, due to larger teams than the norm as well as the variety of ships, classes and mechanics, but that can all be dealt with via the correct algorithms and sufficient sample sizes - there's players with hundreds of games on a single ship and thousands of games overall and that's enough to give a reasonable estimate for a player's relative performance in a variety of ships. The problem here is that WG aren't even attempting to create balanced teams, they are just throwing people together at random while endlessly trying to make teams have identical ships rather than actually balancing said teams. I don't really understand the resistance players in WoWS have to a proper matchmaker that ensures close games. Poorer players get to enjoy close games that doesn't result in them getting stomped, skilled players get to keep the game's challenge rather than just steamrolling seals until they meet another skilled player. The only player group I could see that would be vehemently opposed to a proper matchmaker would be those that play only to curb stomp seals and want to avoid any sense of challenge.
  3. Would be nice to see, except that the players who rage because they are bottom tier will have a severe aneurism and proceed to rage on the forums for days on end if the matchmaker throws their Fuso into a game where an enemy has a T10 ship. No matter how actually balanced the teams are, these players will enter a mouth frothing fury the moment that they are not the centerpiece of their team.
  4. Something that's worth mentioning is that soontm the Pensacola will be downtiered to T6 once the USN cruiser split happens. She's a respectable performer at T7 (if not particularly strong), so I'd imagine that she'll wreck faces at T6. I'm still holding onto my personal theory that WG realised they screwed up with the powercreep on the newer cruisers (which still isn't as big as the powercreep on the battleships, mind you) and so are planning on downtiering most of the older lines in some way to restore balance.
  5. Yeah, the trend is general across tiers for a given ship class and nation, there will always be some ships that are overall better at fire starting than others. Notably, the Ibuki suffers from the IJN cruisers barely scaling from T7-9, while the Zao represents the first major increase in firepower for a few tiers. Even then, the Zao's effective 5.02 FPM does compare favourably to the Ibuki's effective 3.85 FPM but the 203mm Mogami has 4.31 FPM (155mm Mogami has 5.70, but that's a CL and so isn't a fair comparison) and the Myoko has 5.61 FPM - note that the effective fires per minute greatly favour the T7 Myoko and not the alleged super flamethrowing Zao. Going further back, the Aoba's effective 4.30 FPM is improved compared to the Furutaka's 3.40 FPM, while the lower tier CLs Kuma has 5.40 FPM and Tenryuu's 3.87 FPM. So overall, keeping a range of effective fires per minute between 3.40 and 5.70 across 7 tiers I'd say is a pretty reasonable scaling (particularly as the 5.70 and the 5.40 FPM ships are two of the CLs in the line while the 3.40 FPM is the lowest tier CA in the game). Even just within the CA's, this leaves us with a range of 3.40-5.61, a gap of 65% between the bottom and top performers, without the FRC scaling we would instead have a range of 10.03 FPM down to 4.08 FPM - a performance where the top performer has 146% greater firestarting capacity than the bottom performer. I'm sure players can work around a 65% performance difference between ships over 6 tiers, but trying to choose a middle value that doesn't break when exposed to a 146% difference is much harder. Sure, there's a couple of individual ships that could do with tweaking (such as the Ibuki, which suffers from having basically the same guns as the Myokou from 2 tiers earlier, but I don't know how her overall performance is beyond slinging HE), but by and large the scaling FRC keeps the ever increasing number of shells per minute across the tiers.
  6. The Aoba can set more fires on T6 ships than the Ibuki can to T9 ships because the FRC is doing it's job. It keeps fires relevant across tiers by countering the scaling of the numbers, RoF and calibre of guns. It's no different how the Kuma can penetrate other T4 cruisers about as well as the Zao can penetrate other T10 cruisers - AP vs armour is not really any different to fire chance vs FRC scaling through tiers. Higher tier ships are an upgrade over lower tier ones (generally speaking, there are a couple that are more of a sidegrade), it's just that your opponent's ships also scale in their defensive capabilities too which pretty much cancels out any extra increase in offensive power.
  7. Sad to see them go. They needed balancing, not removal. The trouble is that WG have been pushing CVs further and further out of the game, diminishing the effect of fighters and attempting to get air superiority, rather than trying to get them more involved in the game.
  8. That's why I specified "accounting for ship type and nation", German ships tend to have poorer HE, RN BBs have strong HE, cruisers tend to have better HE etc. Take a single line for a specific nation, and generally speaking there will be an increase in base fire starting potential (of course there will be drops at certain specific points, but we are talking about general trends here), once that is all accounted for you generally see base fire starting potential scale roughly alongside the FRC. Leveling the FRC would partially close the gap, however HE is already really effective against relatively higher tier ships because it ignores armour and deals %hp damage. If you ever see a New Mexico dragged into a T8 game, there's a very real chance that they'll spend some time slinging HE at T8 BBs because their 14 inchers will struggle to do much to angled ships at T8 - this would be an even further buff to HE shells in a role that they are already very effective at. Also, more importantly, that sentence was if they followed through and rebalanced the fire chances to maintain consistent effectiveness across all tiers; low tier ships have their low fire chances balanced by the low FRCs they typically face, while high tier ships have their high base fire chances cancelled out by the high resistances they face, the only way to balance fire all the way from T3-T10 with the removal of the FRC would be by making base fire starting potential equal across all tiers. You can't remove the scaling resistance while also keeping the scaling fire chances or else the balance would be broken (particularly at the extremes), if one goes the other will have to go too. That 14% damage on the Monty could indeed by healed, just like the 14% on the NoCal could be, but remember that the Monty is a T10 while the NoCal is just a T8. T10s are meant to be stronger, they occupy a greater proportion of a team's allocated resources and power budget than the T8s. A damage repair party on a Montana is a more valuable resource for the team than a T8's repair party, the spending of one represents a much greater expenditure of team resources. To take an extreme example, if there was a simple finger of God attack that did 100% damage to any ship at the beginning of a game, what ship would you target? By your logic, targeting any ship would make sense as you are still dealing 100% of a health bar. However, in practice most players would choose to strike down a T10 (barring certain special cases such as the Missouri, or if there's certain ships that can perform a required role on the team such as a team's only DD being a T8), because 100% of the health bar of a T10 is much stronger than 100% of the health bar of a T8. It doesn't matter if it is repairable or not as it is repairable in both cases with battleships, being repairable only matters across tier gaps where ships get an extra charge of the repair party or if you are comparing ships where one of them has a super heal - what matters is that every single % hp and every single consumable of a T10 matters more than a T8 because the T10 is itself much stronger.
  9. The thing is that because of the fire resistance stat, a lower tier ship doesn't have a particularly low fire chance in practice despite what the stats say. If they remove the resistance coefficient, then to maintain fire's balance throughout the tiers they would also have to make all ships (accounting for ship type and nation) equal in firestarting across all tiers, which both makes flamethrowing ships exceptionally powerful against higher tier ships (they already work better than AP marksmen when bottom tier) as well as making absolutely no sense as to why 12 inch guns have the same fire chance as 16 inch guns, or why a Des Moines has a worse fire chance than the Baltimore despite firing the same shells. I agree it shouldn't be a hidden stat though, there's already far too many hidden stats in the game that make for large amounts of difference in performance. For carriers, I don't quite know how it will turn out, but they are a slightly different scenario as fires affect them differently to the rest of the ships as well as they aren't often in the line of fire in the first place. The fire damage has relatively hurt the Montana more in that scenario, as an unmitigated fire deals 16K or so damage while the NC only takes 11K damage. That means the Montana has effectively taken an extra few overpens compared to the NC, taking out more of the team's health pool. Even just by going by percentages, 10% of a T10's hp is worth more to a team than 10% of a T8's hp because the T10 is worth more. It's a widely known thing in MOBA games and RPGs that % hp damage exists as the tank shredder, with single-player and coop games usually deliberately introducing damage caps or extra resistance onto bosses to prevent them being just as fragile as the regular enemies.
  10. Yes, the overall scaling of the fire performance would still generally work upwards through the tiers (albeit with a notable reduction in how performance scales through the tiers, giving bottom tiers in a particular game far more bite), but depending on the position it is balanced around it would create extremely unhealthy trends at both extremes. Balancing around T8 would be great for T8s, workable for T7s and T9s, but would result in T3-6 having almost completely ineffective HE (a St. Louis would cause 34% less fires than before!) while the T10s would proceed to mulch everything with fire because they would enjoy a 27% increase in fires against their fellow T10s. No matter what value you choose to make the base fire resistance, either or both of the two extremes in tier would end up broken. The scaling ensures that fires remain useful and balanced throughout all the tiers, all the way from the T2-3 protected cruisers all the way through to the T10 CAs. %health does make a very large difference, as things with higher health tend to be the higher tier ships. Taking 10% of a T10 is a much larger blow than taking 10% off a T8 (except the T8 cruisers, due to lack of heal) because a T10 represents a much larger proportion of the team's allocated power budget. % health damage completely ignores how durable something is, compare this to AP shells where higher health targets can absorb far more fire from AP. Tougher ships don't take less % hp damage to bring down, they take the same amount of % hp damage, but far more flat damage to stop - relatively speaking they are more vulnerable to the percentage damage. To take an extreme example, a ship with 1000 hp in T10 would effectively shrug off fires, it would only worry about direct damage as HE shells would still destroy it with flat damage; conversely, a ship with 1,000,000 hp could almost ignore AP shells but would remain very vulnerable to fires - weaknesses are relative.
  11. Okay, a bit of hyperbole on my part, but the point is that they are not meant to be equal. No ship is meant to be an equal to a ship 2 tiers higher. Sure, individual ships might have certain advantages over their higher tier cousins, but in general the higher tier ships are meant to be stronger. If the Des Moines and the NO had the same fire starting potential (which would involve ramping it's fire chance up to 28-30%!), then it would indeed be a close matchup in a HE slugfest, with the DM winning very slightly due to an increased health pool and more direct damage coupled with the repair party. It's much the same way a Fuso vs an Amagi would result in an incredibly lopsided matchup, as no matter whether the Fuso uses HE or AP it is outclassed on pure specs and is reliant on the Amagi doing something stupid. The other important thing for fires is that they do % health damage, so a higher tier ship generally suffers more from each individual fire than a lower tier one, which is balanced by it being harder to set them on fire in the first place. Looking at it another way, a Montana has a base fires per minute of 8.64 fires, which is reduced against a fellow TX to a reasonable 4.32 fires. A NoCal has a base fires per minute of 6.48, which is reduced against a fellow TVIII to 4.10 - a number actually very close to the Monty's fires in practice. This shows that the scaling in the upper tiers for these two ships is actually pretty accurate. However, the Montana can set 5.48 fires a minute on average against a NoCal and receive 3.24 in return, which sounds like an advantage for the Montana and it should be, but it's actually far closer than it seems because the Montana takes about 50% more damage from each fire; this means that going purely by fire damage, the a Montana in a HE slugging match against a North Carolina would only just come out on top in terms of damage dealt, dealing 12% more fire damage than the NoCal despite having twice as many guns because it takes 50% more damage per fire but is only 21% more resistant to the starting of fires. Even with this scaling, HE is still far more effective against bigger ships than small ones because it deals % hp damage, a North Carolina is better off shooting a TX with HE than a fellow T8 as Creamgravy above me has stated.
  12. It doesn't work out like either of the two scenarios I proposed because of the scaling fire resistance. The lower tier ships have the much lower firestarting potential but in practice it doesn't matter too much because they also have much lower fire resistance, the two effects cancel each other out nicely; the higher tier ships have great firestarting potential but they are reigned in because their usual targets have much higher fire resistance. The only effect they have is to ensure that HE doesn't become a universal cure-all for dealing with relatively higher tier ships and that shooting HE at relatively lower tier ships isn't a waste of time. Yes, it does leave the bottom tiers in a match relatively weak and vulnerable, but that's working as intended. Bottom tier ships aren't meant to be unstoppable juggernauts, they are there to support and take advantage of misplays and mistakes while leaving the heavy lifting to the top tier ships.
  13. Because if you don't offer some kind of fire resistance you get one of two scenarios: If you balance fires around the lower tiers, with lower tier ships with their small numbers of slow firing low calibre guns, then the upper tiers will just turn into fire spewing slugfests. If you instead balance fires around the upper tiers, where ships like the Conqueror, the Zao and the Des Moines are the benchmarks of firestarting performance, then lower tier ships might as well not even bothering to ever use HE shells to start fires. T8 ships should have a harder job setting fires on a T10 than the other way around, T10 ships are meant to be more powerful overall. It's just like how a T10 ship is expected to land more penetrating hits on a T8 than vice-versa. Offering the scaling resistance means that against equal tier ships, overall firestarting performance remains roughly the same throughout all tiers despite the base fire potential increasing, while also offering a scaling increase in combat performance throughout those same tiers rather than HE stagnating in performance.
  14. Furutaka all the way through to the Ibuki use the same 203mm guns though, it's just that the Ibuki has more of them (while the Zao has even more guns than the Ibuki). Similarly, all the USN 203mm guns have 14% fire chances but the Baltimore has 6 RPM for it's guns and the Des Moines has 10.9 RPM compared to the Pensacola's 4 - all other things being equal the Des Moines has about double the firestarting potential. There is a notable increase in firestarting potential throughout almost all ships as you tier up in terms of fires per minute, even if there isn't that much of an improvement per shell.
  15. The trouble is that either you end up with higher tier ships with their larger number of bigger guns having the same (or lower) fire chance than smaller ships with their pop guns, or you end up with massive permafires in the upper tiers. The fire resistance scaling prevents this, as it both gives higher tier ships reasonable stats for their guns as well as making their opponents suitable fire resistant. However, I dislike how fire chance is just a rigid factor based entirely on tier. It just feels gamey and out of place, I'd much rather if they actually tweaked individual ships fire chances (using the current values as a rough guide) to make a ship's individual fire chance a meaningful balance metric rather than just another scaling factor based on tier.