AndyHill

Weekend Tester
  • Content count

    355
  • Joined

  • Last visited

  • Battles

About AndyHill

  • Rank
    Petty Officer
  • Birthday
  • Portal profile AndyHill
  1. Don't be too obsessed with win rate, if you want to improve just focus on playing well and don't hesitate to ask for advice when you're having trouble. Playing divisions really only helps your winrate if you're with good players, but then again the best reason to play with other people is to have fun together. Of course if you can div up with really good players that's probably the best way to learn new tricks and improve your gameplay.
  2. Well carriers are of course a special case. Mostly I want to know how aggressively I can play against certain opponents. Am I likely to be able to outplay someone with the oldest tricks in the book, especially when playing DDs I want to know who has a suitable "fish soup index". If the odds are heavily against my team I either just focus on farming damage or if the situation presents itself I go for a super high risk / super high reward play to try to hard carry the day. Sometimes they work when the red team is caught off guard, sometimes I just get faster to the next game (fairly often as a top-5 performer in the team). And yes, I also check out who I want to fight with, support or lay smokes for. I also try to focus down strong opponents whenever an opportunity presents itself or look for weaknesses in the enemy lineup.
  3. [In reference to someone with lots of dmg but lowish win% allegedly mostly playing high tiers] Well we don't really know if he could, since he mostly plays high tiers. I don't really understand what you're getting at here, since I don't know why someone's performance in a one single game should carry any real weight (and I think I know who you are talking about). Especially when he's facing a team with four unicum to near-unicum players and a very solid carrier. Unfortunately I don't remember how he did in the game, he wasn't one of the ships I engaged on the flank I went to. As for different ships demanding different playstyles that's definitely true, but on the other hand usually players who are very good in one ship type tend to do at least pretty good in most others.
  4. I took a look at your stats with those ships and I'd say that by the time you reach 400-500 combined battles with the kind of damage you have now you will have unicum-level win% or close to that or you must have discovered the most epic way to farm stats without influencing the game.
  5. First of all, nothing puts you in a place to belittle others. Bad behavior and especially certain kinds of not very constructive rant threads (we all know and love) can expose your stats to be belittled a bit perhaps. A mid-range 50% win rate (depending on other stats) makes you a good player anyway and because of how inaccurate statistical predictions are in the case of one battle anyway, the actual difference between a 55% and 60% player isn't that great. Both are expected to carry their weight in your team or punish you hard for your mistakes if they're in the opposing team.
  6. Well I use stats software and you will probably have some difficulty trying to find my stats anywhere. For a number of reasons.
  7. 20 Battles does not make a meaningful sample. And statistics don't guarantee anything about a single battle, that's simply not how they work. They are just a very rough estimate as far as any single battle is concerned. Actually it does. I have some kind of a gut feel for what I'd expect to see from a normal player and I am pretty decent at spotting rerolls. If someone has way too good stats for his number of games I expect it to not be a first account and although still a good player, he's probably not quite as godlike his stats show since he has taken the potato period we all have out of the equation.
  8. Well not really, except in some cases but I can kind of almost pick that up with the data given. If someone has very low damage compared to his win% I tend to check to see his stats just to see why I'm not getting the ratio I expect and usually it's the case of someone being carried in divisions, playing lots of low tiers or a very focused DD player. These are all something I can't see from my stats alone, though, and the first two options will lead to a false conclusion if he is in fact actually the third case (which is somewhat rare). If you can get 70% wins in a division you have to be a very good player playing with very good players. CVs are something of a special case, but usually players who are superb at one type of a ship tend not to be bad at other ship types. Cruisers and especially destroyers tend to be more demanding than battleships, though, so I pay more attention to those generally.
  9. The problem with your analysis is that 23 battles is basically a meaningless sample as far as win% is concerned. However, 1.78 kills and enormous damage is a bit different and although you might be correct and he farms those after game is lost, I just don't really see that. When someone has relatively low win% for his average damage it's almost invariably someone who plays a lot of high tiers. Now if he had 230 games in that ship with low win% and otherwise good stats I'd say your analysis is probably correct. So he's a pretty good player I'd say. Why do you think his win% at tier 10 is low? Sorry I'm a bit busy right now I won't try to make the guess right now but if it's really important for you let me know and I'll make the time later. I can't possibly remember his performance in the battle, though, since that was basically a random game I picked and I don't remember details of it.
  10. This is precisely the point. I don't think it is, but I can't prove it because it's just my personal experience. The software I use is actually open source so if I had any kind of programming skills I could make it do some data gathering for me, but I don't. I'll continue my point a bit later... As was mentioned before by someone (sorry I forget and am too lazy to check by who, but it was a very good point) there's no such thing as a T4 player. Even very experienced players can play low tiers and they tend to farm wins and dump the win% for less experienced players. Just about everyone will play T4 from time to time, as a statistics genius I now make a bold prediction that for example let's say in two weeks we will see a disproportionate amount of experienced players at T4. As one of my not so greatest recent gaming achievements I was told some time ago that I was among something like top 20 Katori players on the server. But it was the only ship I could use for retraining Japanese carrier captains because of torpedo acceleration (not going to play FujinKaze with that) and and... And experience is one of the key elements here. There are bound to be a lot of players who play a few games, never get past tier 1, 2, etc. and one of the things I'm absolutely sure of is that I see an enormous difference in average games played at low tiers vs high tiers. As you can see in one of my screenshots, more than half the players have less than 1k games, which I consider complete beginners. At high tiers that does happen too (for reasons we all know and love), but it's an anomaly whereas at low tiers it's the norm. And those inexperienced people simply can not be very good - unless they are rerolls, which you can quite easily spot from stats. At high tiers the players who are resistant to learning tend to fail pretty constantly. At low and especially very low tiers, the majority of players have never had the chance to learn. That alone contributes hugely to the difference in skill at different tiers. As you mentioned yourself, I'd see a disproportionately high (but not hugely so) Win% compared to damage with not too many games. In all likelihood a very dangerous opponent or a reliable ally. No, not really. Your number of games played makes you not the most experienced player I usually see at T9, but that's not a huge factor. Everything else you mentioned is kind of inconsequential for me when I'm making my assumptions. I've shifted through vast masses of data to kind of get a feel for what to expect and all the other data points go through a similar or similarly different development history and nothing you mentioned makes me think you wouldn't be a dangerous opponent / good ally. The important thing to understand is that I'm not making a precise analysis on exactly how good someone is under all circumstances, just a rough estimate on general skill level. And when we're talking about one game a super unicum can fail hard and the AFK farmer can get solo warrior. I won't go into wall of skill argument here, I don't think it's the time or place. Let's just say I've played lots of FujinKaze and some Shima and if I need to farm BB kills for missions, I'm not taking the Shima. However, your point on the maps is a very good one and definitely a valid reason to not like high tiers gameplay. Yes it is true that the engagements are often at longer ranges at high tiers, but not always. Depending on the ship and your playstyle you can get into brawls where you need to try to hit specific weak points in armor etc. Also when I mentioned lower accuracy I was mostly referring to the very low tier ships that mostly tend to have very slow shells and inaccurate guns.
  11. From ship stats: Pretty much nothing. Overall: Potentially a very good player if the sample size is high enough.
  12. I agree completely - that's actually pretty much exactly what I at least tried to say earlier in this thread. At low tiers silly moves don't stand out in the general chaos, but at high tiers they tend to stand out much more when players on average have a much better idea of what they are doing.
  13. I don't see why win rate is not relevant in each case. The only time win rate isn't relevant is when you have very few battles and that's really the only reason why I want to see how many battles someone has. The combination of win% and average damage is the most powerful tool for quickly figuring out what kind of a player I'm looking at. A high win% with a high average damage is almost invariably a good player. A high win% player with little damage and a statistically relevant number of games played is either a seal clubber or has been carried in divisions - or in more rare cases a completely destroyer focused player, those tend to get lower damage numbers. Someone with low win rate and low average damage is highly likely not very good. Note that I'm not saying that with only those stats you can make an accurate assessment of just how good someone will play in a specific situation. That's not really possible no matter what, so I'm only looking for a tool that gives me quick and dirty estimates that may or may not provide information useful in the game at hand. This is actually the key thing to understand about statistics. There will always be exceptions, but that doesn't mean you can't make accurate predictions based on suitable data. I'm not claiming I can make accurate analysis for each and every individual, all I'm saying that I tend to see on average far better players in high tier games. It is true that high tier ships are more demanding. They have more gimmicks, but they are also more accurate, faster etc. - which is the reason I personally like high tier gameplay more (in addition to generally higher player skill level). It's generally more skill and less luck based than at low tiers. And if I'm not missing something obvious I think we kind of agree on the latter bolded part. As I mentioned, the reason bad plays are more obvious at high tiers is because they get punished much more heavily. At low tiers you can get away with murder and your opponents are likely to give you more opportunities for massive carrying. Just in case you're curious, here are a few examples of what it looks like for me when I get the stats for a game (and there is no intent of naming and shaming, these are simply the actual stats for a group of players and none of them are even especially bad considering how few games many of them have). This is the latest game I played so basically a random example of a low tier game: Both teams have approximately equal numbers of beginners, but our superior carrier and three very high skill players (Iron Duke, Clemson and Arkansas) made our team clear favorites for the game (and to very carefully avoid naming and shaming I won't even reveal what actually happened in this particular game. Here's another example from a fairly high level ranked game (I can't remember any details of this one nor why I even have that screenshot): I haven't checked but if you check out any of the players with high win rates and high average damage, you are likely to find good to great performance in all relevant categories.
  14. I see overall win rate, average damage and games played. And you are correct, I have not done statistical research and it is more of a personal opinion. However, it is the personal opinion (based on experience) by someone who has actually seen the stats of almost everyone in the battles he has played in the last 6+months or so.
  15. From what I've seen the players in bigger tier games are on average by far better than in lower tier matches. Note that I have not done any statistical analysis on the subject, it is purely my own observation, but the difference is significant. Of course things like differences between specific tiers are not really accurate and I can't say I'm correct about that with any real kind of certainty. However, the difference between average players at low tiers and high tiers is enormous and even between mid tiers and high tiers it's significant enough to notice.