Well then...
10:11 AM
Posted by Reygahnci on Monday, June 9, 2008
It has been a while since I posted. I had a great trip in Maui and did entirely too much. Most of my days were filled with riding in the car and seeing sites that I have seen a thousand times before, but we did it for the sake of my girlfriend who had never been to the island. This was a mistake because we barely had any time given to us just laying on the beach enjoying the sun and surf. At least, the last few days of the trip were devoted to such idle pleasures.
In my last post I said that if I could come up with a better arena rating system, I would discuss it in full here. However, I came to the lazy conclusion that a better system would require more work than it is likely worth. Sure, the current system is broken, or perhaps it is simply "not perfect", yet we have to admit that there are more pressing matters concerning the arena system than the Elo ratings that make it flawed. Without getting too entrenched in a discussion on fairness in competitive environments, let me just say that team composition, battlegroup population, battlegroup play times, and class balance are at the core of this discussion and would likely be another semi-essay a la my previous post concerning the Elo system. Suffice-to-say, that is not what I want to talk about.
SOOOO, I had to take a three hour break there because of work, so I am going to switch directions here and talk about the alpha stuff that is being released at the moment. Wikidot has been really good about keeping up to date with the alpha information that has been trickling out since it began. I am actually rather surprised that it has not complied with a take down notice as of yet... seeing how Blizzard likes to hand those out like candy concerning unreleased information (in fact, I got a perma-ban from the WoW forums for simply posting in a thread with unreleased information... they want everyone to stay away from it, I guess). SOOOO, because my account MIGHT (I don't really remember) be connected to this blog via my email address, I am going to hold off from posting discussions about alpha information and, instead, only post that link to Wikidot where all said information can be found. I think that this is sufficient to not be violating any terms of service that I have engaged with Blizzard about (and I might even be well within my rights to talk about information on the wikidot site... but I want to reread my contracts with Blizzard on the matter before I go off the deep end).
So, until I get more understanding on my agreements, I guess that I will only talk about current stuff...
Such as my 3s team. I am rolling warrior+rogue+druid and doing rather well. Obviously, we have trouble with the harder makeups and are usually beating the easier makeups. For instance, we can beat most teams like ours, dps+utility/dps+healer because of my ability to heal on the move, but we tend to lose (every time) to PMR. Also, we fought a very odd warlock+pally+pally team that was 2k rated and beat us... although we lasted a lot longer than they were expecting. The difference between our teams (I was restokin at the time) is that while I would not OOM if they only targeted one of my team mates throughout the entire match, they would never OOM because FoL is nigh-free and they had two people casting it on a warlock who would cut-self to sustain infinite mana.
We beat a bunch of teams that would kill one of my team mates right off the bat. For instance, shamang+beastwithin+priest would start off by bloodlusting and simply destroying my warrior, but we would end up winning once beast within and bloodlust wore off because my rogue is almost full s3 with comparable pve weapons while I was restokin, and would never OOM while the priest would eventually die. We also faced hunter+hunter+restoshamang which would try the basic same idea of nuke the warrior and hope to have enough left over to take down a restokin+rogue team. They did it once, but then I figured out to cyclone one of the warriors before he could beast within, and he would also miss the bloodlust (because the shamang was terrible in s3 shoulders... that pissed me off). So, the splitting of that damage and my warrior's ability to stay out of LoS with me via intervene was enough to beat that team the second time. We beat a mirror-match team because their druid was full resto, not restokin so he eventually OOM'd and I kept healing. Pretty snazzy if you ask me. I am probably going to try some 2s with my rogue friend to see if we can't hit 1850 to get him the s3 weapons... which will act as a bootstrap to get our 3s team to 1850 so my warrior can get the s3 weapon. The only problem now is that I am no longer spec'd restokin... which is the claim-to-fame with a rogue in 2s.
Also, our warlock+spriest team got some more video in this week. It turns out that we just destroy most warrior+healer teams, but we don't face enough of them anymore. We face rogue+healer a lot, which is becoming a problem as neither of us can take a rogue on (as if someone actually could). We should have won 5-6 of our 10 matches this week, but I was rusty having not played my warlock in over two weeks, so I ended up costing us two matches. One of those matches was a definite win if I had played better, while the other could have been a win, but there was no definitive "we should have won except I didn't cast shadowburn" moment to decide whether it would obviously be a win. So, our rating dropped a little, but it really could matter less as brother was so close to the cap that he had to buy the Vengeful End Game, which cost him 1000 arena points and only gained him two stam and a couple spell damage. The fact that we are missing gear and cannot buy anything else is starting to irk me.
However, we got a good video of us beating druid+rogue that was half due to a less-than-amazing rogue and a brilliant play on my behalf that gamed the druid. We were on Blade's Edge, fighting on top of the bridge with my felguard on the druid (who is at about 50%) and the rogue on the spriest after getting me to 50%. The rogue is 100%, my spriest is 90%, I am 50%, the druid is 50%. I notice that the druid is at about 40% mana, so I figure he would love to go and heal+drink to get back to a reasonable state of the fight. So, I sick my pet on the rogue, who is just tearing the priest apart, and the druid is good enough to notice this and run off, heal, and drink. However, that is exactly what I wanted, because the usual thing happens, the druid is a nelf, so he runs behind one of the pillars on the ground starts to drink, and shadowmelds, putting the pillar directly between himself and his rogue and us. At this point, I just start unloading shadow bolts (along with all my dots) onto the rogue, who had just recently used CloS, so I knew he would get the big hits. Then it happened, the druid noticed his rogue taking damage, started running up the opposite side to try and cyclone me then start healing, but I saw him coming. So, I fire off my last shadow bolt, it crits for about 2k, but also puts shadow vulnerability up, so I unload deathcoil and shadowburn right before getting cycloned... not that it matters at this point, I have done my damage. I see SW:D land and take the rogue to 1%, THEN cheat death procs, and the dots kill him before the druid can swiftmend. To say the least, that match was very close at the end, and we got a killshot off luckily against a rogue, which doesn't happen much anymore because of cheat death (I cannot wait for abilities that might go through immunities/resists/etc... I hope they add something like that as the warlock's 51 point talent in destro lawl).
So, now we have probably 150MB worth of video captured, not all of it is great, but it is enough to start composing a preliminary video without sound, trying to work them into a flow. I will probably begin doing that this week, maybe next depending on how busy everything is. Check back for more.
June 9, 2008 at 11:41 PM
A perfect system is impossible, but it should be possible to make a system that at least works well enough to satisfy either the most players or the most important players (whatever criteria for that may be). So who does the current system most help?
Not the casuals, they get mowed over by much higher rated players all the time with the current system. How about the top tier players? The system doesn't help them; there's no incentive to stay at the top since even a 2k 2v2 team will be getting enough points each week to be completely fitted with the top PvP gear in short order. And if they ever change teams or play in another bracket, they're right back at 1500. The mid-range players? They get hit the worst by higher rated players playing at a lower rating.
There is a group that benefits from the current system though: the point sellers. The system screws a lot of people, and there's top dollar to be had for the people who can game it for someone else's benefit.
So while I agree that perfection is impossible, I think it shouldn't be too much to ask for something better than the current system.
June 10, 2008 at 9:57 AM
I would have to agree that a better system is in order, however I would like to put a bit of a divide between the rating system and the match-making system. There is definitely something wrong with the fact that a team who goes 5-5 each week and loses rating because of a rerolled 2k team at the 1500 rating, but is that the fault of the rating system, or the match-making system?
I would tend to lean toward the latter rather than the former. I would think that the rating system is working perfectly well at the point of "you lost to a 1500 team at a 1700 rating, therefore the other team should gain a lot and you should lose a lot." Where it all falls apart, however, is when the gear comes into play. If a 1700 team in 1700 gear lost to a 1500 team in 1500 gear, then the system works perfectly. However, if the 2k reroll team in full 2k gear beats a 1700 team when they have a 1500 team, then the system has been corrupted. This understanding of the separation of systems should give us a foothold for progress. In this case, the rating system is fine, but the match-making system is flawed.
A few things that might change how the match-making system works...
you know what, this deserves a real post.